CN116108269A - Search result display method and device, electronic equipment and storage medium - Google Patents
Search result display method and device, electronic equipment and storage mediumDownload PDFInfo
- Publication number
- CN116108269A CN116108269ACN202211733384.9ACN202211733384ACN116108269ACN 116108269 ACN116108269 ACN 116108269ACN 202211733384 ACN202211733384 ACN 202211733384ACN 116108269 ACN116108269 ACN 116108269A
- Authority
- CN
- China
- Prior art keywords
- search
- sample
- user
- user interest
- search term
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9538—Presentation of query results
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- User Interface Of Digital Computer (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
Translated fromChinese技术领域technical field
本发明涉及端智能技术领域，尤其涉及一种搜索结果展示方法、装置、电子设备及存储介质。The present invention relates to the field of terminal intelligence technology, and in particular to a search result display method, device, electronic equipment and storage medium.
背景技术Background technique
随着互联网技术的不断发展，端智能成为逐步兴起并广泛应用的技术。目前，可以在端侧通过模型对用户搜索出来的搜索结果进行用户偏好打分，并根据用户偏好对搜索结果进行重新排列，以使用户更容易看到自己的偏好搜索结果。With the continuous development of Internet technology, terminal intelligence has gradually emerged and is widely used. At present, the user preference scoring can be performed on the search results searched by the user through the model on the terminal side, and the search results can be rearranged according to the user preference to make it easier for users to see their preferred search results.
目前针对搜索结果进行用户偏好打分的模型均为通过与搜索结果有关的训练样本进行训练，但此方法训练出来的用户打分模型的预测效果较差，对搜索结果的估值准确度不高，无法精准预测用户的喜好。At present, the models for scoring user preferences for search results are all trained by training samples related to search results, but the prediction effect of the user scoring model trained by this method is poor, and the accuracy of the estimation of search results is not high. Accurately predict user preferences.
发明内容Contents of the invention
本发明实施例提供一种搜索结果展示方法、装置、电子设备及存储介质，以新的维度分析用户偏好，即以用户针对搜索词的行为数据来训练用户感兴趣度预测模型，提升了模型对搜索结果的估值准确度。Embodiments of the present invention provide a search result display method, device, electronic device, and storage medium, which analyze user preferences in a new dimension, that is, use user behavior data for search terms to train a user interest prediction model, and improve the model's ability to The estimated accuracy of the search results.
本发明实施例第一方面提供了一种搜索结果展示方法，应用于客户端，所述方法包括：The first aspect of the embodiment of the present invention provides a method for displaying search results, which is applied to a client, and the method includes:
获取服务端返回的对应于目标搜索词的多个搜索结果；Obtain multiple search results corresponding to the target search term returned by the server;
通过本地预先训练的用户感兴趣度预测模型，得到所述多个搜索结果对应的用户感兴趣度；所述用户感兴趣度预测模型的训练样本至少包括：样本搜索词的用户行为特征；所述样本搜索词包括：用户历史输入的搜索词，以及，用户历史输入的搜索词的联想搜索词；Through the local pre-trained user interest degree prediction model, the user interest degree corresponding to the plurality of search results is obtained; the training samples of the user interest degree prediction model include at least: user behavior characteristics of sample search terms; The sample search terms include: the search term entered by the user in history, and the associated search term of the search term entered by the user in history;
按照用户感兴趣度从高到低的顺序，对所述多个搜索结果进行重排；Rearranging the plurality of search results in descending order of user interest;
展示重排后的多个搜索结果。Display multiple search results after rearrangement.
可选的，所述用户感兴趣度预测模型的训练样本还包括：样本搜索结果的用户行为特征；所述样本搜索结果是所述服务端返回的对应于目标样本搜索词的搜索结果，所述目标样本搜索词为：所述用户历史输入的搜索词，或，所述用户历史输入的搜索词的联想搜索词。Optionally, the training samples of the user interest degree prediction model further include: user behavior characteristics of sample search results; the sample search results are search results corresponding to target sample search terms returned by the server, and the The target sample search term is: a search term entered by the user in history, or an associative search term of the search term entered by the user in history.
可选的，所述方法还包括：Optionally, the method also includes:
获取对应于用户输入的搜索词的多个联想搜索词；Obtaining a plurality of predicted search terms corresponding to the search term entered by the user;
通过所述用户感兴趣度预测模型，得到所述多个联想搜索词对应的用户感兴趣度；Through the user interest degree prediction model, the user interest degree corresponding to the plurality of associative search words is obtained;
按照用户感兴趣度从高到低的顺序，对所述多个联想搜索词进行重排；Rearranging the plurality of associated search terms in descending order of user interest;
展示重排后的多个联想搜索词。Show multiple predicted search terms after reranking.
可选的，所述方法还包括：Optionally, the method also includes:
在检测到用户对所述样本搜索词的操作的情况下，存储一条正训练样本，所述正训练样本包括以下至少一种数据：所述操作的发生时间、所述样本搜索词的特征、所述样本搜索词对应的业务类型；In the case that the user's operation on the sample search term is detected, a positive training sample is stored, and the positive training sample includes at least one of the following data: the occurrence time of the operation, the characteristics of the sample search term, the The business type corresponding to the above sample search terms;
在未检测到用户对所述样本搜索词的操作的情况下，存储一条负训练样本，所述负训练样本包括以下至少一种数据：展示所述样本搜索词的发生时间、所述样本搜索词的特征、所述样本搜索词对应的业务类型。If the user's operation on the sample search term is not detected, a negative training sample is stored, and the negative training sample includes at least one of the following data: showing the occurrence time of the sample search term, the sample search term features, and the business type corresponding to the sample search term.
可选的，所述方法还包括：Optionally, the method also includes:
在检测到用户对所述样本搜索结果的操作的情况下，存储一条正训练样本，所述正训练样本包括以下至少一种数据：所述点击操作的发生时间、所述样本搜索结果对应的标签特征、所述样本搜索结果的浏览时长、所述样本搜索结果对应的业务类型；In the case where the user's operation on the sample search result is detected, a positive training sample is stored, and the positive training sample includes at least one of the following data: the occurrence time of the click operation, the label corresponding to the sample search result characteristics, browsing time of the sample search results, and business type corresponding to the sample search results;
在未检测到用户对所述样本搜索结果的操作的情况下，存储一条负训练样本，所述负训练样本包括以下至少一种数据：展示所述样本搜索结果的发生时间、所述样本搜索结果对应的标签特征、所述样本搜索结果为零的浏览时长、所述样本搜索结果对应的业务类型。If no user operation on the sample search result is detected, a negative training sample is stored, and the negative training sample includes at least one of the following data: the occurrence time of displaying the sample search result, the sample search result The corresponding tag feature, the browsing time when the sample search result is zero, and the business type corresponding to the sample search result.
可选的，所述方法还包括：Optionally, the method also includes:
响应于用户的启动操作，拉取已存储的训练样本；In response to the user's startup operation, pull the stored training samples;
利用所述已存储的训练样本，对所述用户感兴趣度预测模型的模型参数进行更新；updating the model parameters of the user interest degree prediction model by using the stored training samples;
所述通过本地预先训练的用户感兴趣度预测模型，得到所述多个搜索结果对应的用户感兴趣度，包括：The user interest degree corresponding to the plurality of search results obtained through the local pre-trained user interest degree prediction model includes:
通过更新后的用户感兴趣度预测模型，得到所述多个搜索结果对应的用户感兴趣度。The user interest degrees corresponding to the plurality of search results are obtained through the updated user interest degree prediction model.
本发明实施例第二方面提供了一种搜索结果展示装置，应用于客户端，所述装置包括：The second aspect of the embodiment of the present invention provides a device for displaying search results, which is applied to a client, and the device includes:
第一获取模块，用于获取服务端返回的对应于目标搜索词的多个搜索结果；The first obtaining module is used to obtain a plurality of search results corresponding to the target search term returned by the server;
第一预测模块，用于通过本地预先训练的用户感兴趣度预测模型，得到所述多个搜索结果对应的用户感兴趣度；所述用户感兴趣度预测模型的训练样本至少包括：样本搜索词的用户行为特征；所述样本搜索词包括：用户历史输入的搜索词，以及，用户历史输入的搜索词的联想搜索词；The first prediction module is used to obtain the user interest degree corresponding to the plurality of search results through the local pre-trained user interest degree prediction model; the training samples of the user interest degree prediction model include at least: sample search terms The user behavior characteristics; the sample search term includes: the search term input by the user history, and the associated search term of the search term input by the user history;
第一重排模块，用于按照用户感兴趣度从高到低的顺序，对所述多个搜索结果进行重排；The first rearrangement module is used to rearrange the plurality of search results in order of user interest from high to low;
第一展示模块，用于展示重排后的多个搜索结果。The first display module is used to display multiple search results after rearrangement.
可选的，所述用户感兴趣度预测模型的训练样本还包括：样本搜索结果的用户行为特征；所述样本搜索结果是所述服务端返回的对应于目标样本搜索词的搜索结果，所述目标样本搜索词为：所述用户历史输入的搜索词，或，所述用户历史输入的搜索词的联想搜索词。Optionally, the training samples of the user interest degree prediction model further include: user behavior characteristics of sample search results; the sample search results are search results corresponding to target sample search terms returned by the server, and the The target sample search term is: a search term entered by the user in history, or an associative search term of the search term entered by the user in history.
可选的，所述装置还包括：Optionally, the device also includes:
第二获取模块，用于获取对应于用户输入的搜索词的多个联想搜索词；The second obtaining module is used to obtain a plurality of associated search words corresponding to the search words input by the user;
第二预测模块，用于通过所述用户感兴趣度预测模型，得到所述多个联想搜索词对应的用户感兴趣度；The second prediction module is used to obtain the user interest degree corresponding to the plurality of associative search words through the user interest degree prediction model;
第二重排模块，用于按照用户感兴趣度从高到低的顺序，对所述多个联想搜索词进行重排；The second rearrangement module is used to rearrange the plurality of associative search terms in order of user interest from high to low;
第二展示模块，用于展示重排后的多个联想搜索词。The second display module is used to display a plurality of rearranged associative search words.
可选的，所述装置还包括：Optionally, the device also includes:
第一存储模块，用于在检测到用户对所述样本搜索词的操作的情况下，存储一条正训练样本，所述正训练样本包括以下至少一种数据：所述操作的发生时间、所述样本搜索词的特征、所述样本搜索词对应的业务类型；The first storage module is configured to store a positive training sample when detecting the user's operation on the sample search word, and the positive training sample includes at least one of the following data: the occurrence time of the operation, the The characteristics of the sample search term and the business type corresponding to the sample search term;
第二存储模块，用于在未检测到用户对所述样本搜索词的操作的情况下，存储一条负训练样本，所述负训练样本包括以下至少一种数据：展示所述样本搜索词的发生时间、所述样本搜索词的特征、所述样本搜索词对应的业务类型。The second storage module is used to store a negative training sample when no user operation on the sample search term is detected, and the negative training sample includes at least one of the following data: showing the occurrence of the sample search term Time, features of the sample search term, and service type corresponding to the sample search term.
可选的，所述装置还包括：Optionally, the device also includes:
第三存储模块，用于在检测到用户对所述样本搜索结果的操作的情况下，存储一条正训练样本，所述正训练样本包括以下至少一种数据：所述点击操作的发生时间、所述样本搜索结果对应的标签特征、所述样本搜索结果的浏览时长、所述样本搜索结果对应的业务类型；The third storage module is configured to store a positive training sample when the user's operation on the sample search result is detected, and the positive training sample includes at least one of the following data: the occurrence time of the click operation, the The tag feature corresponding to the sample search result, the browsing time of the sample search result, and the business type corresponding to the sample search result;
第四存储模块，用于在未检测到用户对所述样本搜索结果的操作的情况下，存储一条负训练样本，所述负训练样本包括以下至少一种数据：展示所述样本搜索结果的发生时间、所述样本搜索结果对应的标签特征、所述样本搜索结果为零的浏览时长、所述样本搜索结果对应的业务类型。The fourth storage module is used to store a negative training sample when no user operation on the sample search result is detected, and the negative training sample includes at least one of the following data: showing the occurrence of the sample search result time, the label feature corresponding to the sample search result, the browsing time when the sample search result is zero, and the business type corresponding to the sample search result.
可选的，所述装置还包括：Optionally, the device also includes:
第三获取模块，用于响应于用户的启动操作，拉取已存储的训练样本；The third acquisition module is used to pull the stored training samples in response to the user's startup operation;
模型更新模块，用于利用所述已存储的训练样本，对所述用户感兴趣度预测模型的模型参数进行更新；A model update module, configured to update the model parameters of the user interest degree prediction model by using the stored training samples;
所述第一预测模块，包括：The first prediction module includes:
第一预测子模块，用于通过更新后的用户感兴趣度预测模型，得到所述多个搜索结果对应的用户感兴趣度。The first prediction submodule is configured to obtain the user interest degree corresponding to the plurality of search results through the updated user interest degree prediction model.
本发明实施例第三方面提供一种电子设备，包括处理器、存储器及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述计算机程序被所述处理器执行时实现如本发明第一方面所述的搜索结果展示方法的步骤。The third aspect of the embodiments of the present invention provides an electronic device, including a processor, a memory, and a computer program stored on the memory and operable on the processor, and the computer program is implemented when executed by the processor. The steps of the search result display method described in the first aspect of the present invention.
本发明实施例第四方面提供一种计算机可读存储介质，所述计算机可读存储介质上存储有计算机程序，所述计算机程序被处理器执行时实现如本发明第一方面所述的搜索结果展示方法的步骤。The fourth aspect of the embodiment of the present invention provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the search result according to the first aspect of the present invention is realized Show the steps of the method.
本发明实施例通过客户端获取服务端返回的针对目标搜索词的多个搜索结果，通过客户端本地预先训练的用户感兴趣度预测模型，预测得到多个搜索结果对应的用户感兴趣度，从而按照用户感兴趣度从高到低的顺序，对多个搜索结果进行重排及展示；其中，本实施例中的用户感兴趣度预测模型的训练样本至少包括：样本搜索词的用户行为特征。通过本实施例的搜索结果展示方法，客户端以一个新的维度分析用户偏好，即用户针对搜索词的行为数据更能体现用户的实际偏好，因此，客户端事先在本地根据采集到的用户针对样本搜索词(用户历史输入的搜索词，以及，用户历史输入的搜索词的联想搜索词)的用户行为特征，进行模型训练，得到训练好的用户感兴趣度预测模型，该用户感兴趣度预测模型对搜索结果估值预测更准确，更偏向于用户的真实喜好，较好地提升了模型的预测效果，从而在客户端获取到服务端返回的搜索结果时，可以通过该训练好的用户感兴趣度预测模型对多个搜索结果进行预测，并按照预测得到的用户感兴趣度对多个搜索结果进行重排及展示，使得重排结果更倾向于用户喜好，让用户更容易找到自己的偏好目标。In the embodiment of the present invention, the client obtains multiple search results for the target search term returned by the server, and uses the local pre-trained user interest prediction model of the client to predict the user interest corresponding to the multiple search results, thereby Rearrange and display multiple search results in order of user interest from high to low; wherein, the training samples of the user interest prediction model in this embodiment at least include: user behavior characteristics of sample search terms. Through the search result display method of this embodiment, the client analyzes user preferences in a new dimension, that is, the user's behavior data for search terms can better reflect the user's actual preferences. The user behavior characteristics of the sample search terms (the search terms entered by the user in the history, and the associated search terms of the search terms entered by the user in the history) are used for model training to obtain the trained user interest degree prediction model, and the user interest degree prediction The model is more accurate in predicting the valuation of search results, and is more inclined to the user's real preferences, which improves the prediction effect of the model, so that when the client obtains the search results returned by the server, it can use the trained user experience. The interest degree prediction model predicts multiple search results, and rearranges and displays multiple search results according to the predicted user interest, making the rearranged results more inclined to user preferences and making it easier for users to find their preferences Target.
附图说明Description of drawings
为了更清楚地说明本发明实施例的技术方案，下面将对本发明实施例的描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the accompanying drawings that need to be used in the description of the embodiments of the present invention. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention , for those skilled in the art, other drawings can also be obtained according to these drawings without paying creative labor.
图1是本发明一实施例示出的一种搜索结果展示方法的流程图；FIG. 1 is a flow chart of a method for displaying search results shown in an embodiment of the present invention;
图2是本发明一实施例提供的搜索结果展示装置的结构框图；Fig. 2 is a structural block diagram of a search result display device provided by an embodiment of the present invention;
图3是本发明一实施例示出的一种电子设备的示意图。Fig. 3 is a schematic diagram of an electronic device according to an embodiment of the present invention.
具体实施方式Detailed ways
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.
请参考图1，图1是本发明一实施例示出的一种搜索结果展示方法的流程图。本实施例提供的模型运行方法可应用于客户端。在本实施例中，客户端可以是指为用户提供本地服务的程序，客户端可以以不同的程序形态来实现同一软件功能；例如，客户端可以是应用软件APP、小程序、网页web浏览器等。如图1所示，本实施例的搜索结果展示方法可以包括以下步骤：Please refer to FIG. 1 , which is a flowchart of a method for displaying search results according to an embodiment of the present invention. The model running method provided in this embodiment can be applied to a client. In this embodiment, the client may refer to a program that provides local services for users, and the client may implement the same software function in different program forms; for example, the client may be an application software APP, a small program, or a web browser wait. As shown in Figure 1, the search result display method in this embodiment may include the following steps:
步骤S11：获取服务端返回的对应于目标搜索词的多个搜索结果。Step S11: Obtain multiple search results corresponding to the target search term returned by the server.
本实施例中，用户在使用客户端的搜索功能时，客户端可以针对目标搜索词向服务端发送搜索请求，以向服务端请求与该目标搜索词对应的多个搜索结果。其中，目标搜索词为用户确定的需要进行搜索功能的词语。服务端在接收到该搜索请求后，根据搜索请求携带的目标搜索词确定出对应于目标搜索词的多个搜索结果并返回给客户端，客户端获取到服务端返回的对应于目标搜索词的多个搜索结果。In this embodiment, when the user uses the search function of the client, the client may send a search request to the server for the target search term, so as to request the server for multiple search results corresponding to the target search term. Wherein, the target search word is a word determined by the user that needs to perform a search function. After receiving the search request, the server determines multiple search results corresponding to the target search term according to the target search term carried in the search request and returns them to the client, and the client obtains the search results corresponding to the target search term returned by the server. Multiple search results.
步骤S12：通过本地预先训练的用户感兴趣度预测模型，得到所述多个搜索结果对应的用户感兴趣度。Step S12: Obtain user interest degrees corresponding to the plurality of search results through a local pre-trained user interest degree prediction model.
本实施例中，客户端预先训练有用户感兴趣度预测模型，该用户感兴趣度预测模型用于预测用户针对搜索结果的用户感兴趣度。在本实施例中，客户端训练用户感兴趣度预测模型的训练样本中至少包括样本搜索词的用户行为特征，样本搜索词包括：用户历史输入的搜索词，以及，用户历史输入的搜索词的联想搜索词。In this embodiment, the client is pre-trained with a user interest degree prediction model, and the user interest degree prediction model is used to predict the user interest degree of the user with respect to the search results. In this embodiment, the training samples of the client training user interest degree prediction model at least include the user behavior characteristics of the sample search words, and the sample search words include: the search words entered by the user history, and the search words entered by the user history Predictive search terms.
其中，联想搜索词是与搜索词具有关联关系的相关、相似的词语。例如，搜索词为“土豆”，则联想搜索词可以为：洋芋、马铃薯、土豆红烧肉、土豆网等与搜索词“土豆”关联关系的相关、相似的词语。其中，联想搜索词可以是客户端直接根据搜索词确定的，也可以是服务器根据搜索词确定并返回给客户端的，而且客户端和服务端可以通过任何可用方式确定每个搜索词的联想搜索词，对此本申请实施例不加以限定。例如，可通过已训练的联想词预测模型，或者是通过其他预测方式，确定每个搜索词的多个联想词以及相关度，进而基于相关度从中选择TopN或者相关度超过一定阈值的的联想词作为相应搜索词的联想搜索词，N可以为正整数。Wherein, the associative search term is a related and similar term that has an association relationship with the search term. For example, if the search term is "potato", then the associative search term may be: potato, potato, potato braised pork, Tudou.com and other related and similar words related to the search term "potato". Wherein, the predictive search term can be directly determined by the client according to the search term, or can be determined by the server based on the search term and returned to the client, and the client and the server can determine the predictive search term for each search term by any available means , which is not limited in this embodiment of the present application. For example, through the trained associative word prediction model, or through other prediction methods, multiple associative words and correlations for each search term can be determined, and then TopN or associative words whose correlation exceeds a certain threshold can be selected based on the correlation As an associated search term for the corresponding search term, N may be a positive integer.
用户历史输入的搜索词是客户端采集到的用户在客户端中输入的历史搜索词，用户历史输入的搜索词的联想搜索词是与用户历史输入的搜索词具有关联关系的相关、相似的词语，该用户历史输入的搜索词的联想搜索词可以是服务端根据用户在客户端中输入的历史搜索词确定并返回的，也可以是客户端自身根据用户在客户端中输入的历史搜索词确定的，本实施例对此不作限制。The search words entered by the user in the history are the historical search words entered by the user in the client collected by the client, and the associated search words of the search words entered by the user in the past are related and similar words that are associated with the search words entered by the user in the past , the associated search term of the search term entered by the user in history may be determined and returned by the server based on the historical search term input by the user in the client, or determined by the client itself based on the historical search term input by the user in the client , which is not limited in this embodiment.
客户端将服务器返回的对应于目标搜索词的多个搜索结果输入至该用户感兴趣度预测模型，进行模型预测，得到用户感兴趣度预测模型输出的该多个搜索结果各自对应的用户感兴趣度。The client inputs the multiple search results corresponding to the target search words returned by the server into the user interest degree prediction model, performs model prediction, and obtains the corresponding user interest values of the multiple search results output by the user interest degree prediction model. Spend.
本实施例中，用户使用客户端的搜索功能时，用户在客户端输入搜索词后，客户端还会确定出该搜索词的多个联想词并展示，以供用户从输入的搜索词和该搜索词的多个联想词中确定出目标搜索词进行搜索。用户在确定目标搜索词的过程中，客户端会记录下针对这些搜索词的用户行为特征，每次记录下的搜索词的用户行为特征均可以作为上述样本搜索词的用户行为特征进行用户感兴趣度预测模型的训练。本实施例中通过客户端进行用户感兴趣度预测模型的训练，由于数据采集(即训练样本的采集)和使用都在客户端，更具备实时性且可避免隐私泄漏，同时数据全部来自用户个体，预测结果更倾向于单个用户的喜好，从而进一步提升了模型的预测准确性。In this embodiment, when the user uses the search function of the client, after the user enters the search term on the client, the client will also determine and display multiple associated words of the search term for the user to learn from the input search term and the search term. A target search word is determined from a plurality of associated words of the word for searching. When the user determines the target search words, the client will record the user behavior characteristics for these search words, and the user behavior characteristics of the search words recorded each time can be used as the user behavior characteristics of the above sample search words to determine whether the user is interested. training of predictive models. In this embodiment, the training of the user interest degree prediction model is carried out through the client. Since the data collection (that is, the collection of training samples) and the use are all in the client, it is more real-time and can avoid privacy leakage. At the same time, all data comes from individual users. , the prediction result is more inclined to the preference of a single user, which further improves the prediction accuracy of the model.
步骤S13：按照用户感兴趣度从高到低的顺序，对所述多个搜索结果进行重排。Step S13: Rearrange the multiple search results in descending order of user interest.
本实施例中，可以理解，服务端返回的多个搜索结果是按顺序排列的，客户端通过本地存储的用户感兴趣度预测模型，得到多个搜索结果对应的用户感兴趣度后，可以按照用户感兴趣度从高到低的顺序，对该多个搜索结果进行重新排列。In this embodiment, it can be understood that the multiple search results returned by the server are arranged in order. After the client obtains the user interest corresponding to the multiple search results through the locally stored user interest degree prediction model, it can follow the The multiple search results are rearranged in descending order of user interest.
步骤S14：展示重排后的多个搜索结果。Step S14: displaying the rearranged search results.
本实施例中，客户端对多个搜索结果进行重新排列后，可以在客户端中展示该重排后的多个搜索结果，从而让用户更容易看到自己的目标搜索结果，提高用户体验。In this embodiment, after the client rearranges the multiple search results, the rearranged multiple search results can be displayed on the client, so that users can more easily see their target search results and improve user experience.
在本实施例中，客户端事先在本地根据采集到的用户针对样本搜索词(用户历史输入的搜索词，以及，用户历史输入的搜索词的联想搜索词)的用户行为特征，进行模型训练，得到训练好的用户感兴趣度预测模型。客户端以一个新的维度分析用户偏好，该新维度(用户针对搜索词的行为数据)更能体现用户的实际偏好，因此，客户端训练出来的用户感兴趣度预测模型对搜索结果估值预测更准确，更加偏向于用户的真实喜好，较好地提升了模型的预测效果，从而在客户端获取到服务端返回的搜索结果时，可以通过该训练好的用户感兴趣度预测模型对多个搜索结果进行预测，并按照预测得到的用户感兴趣度对多个搜索结果进行重排及展示，使得重排结果更倾向于用户喜好，让用户更容易找到自己的偏好目标，提升了用户体验。In this embodiment, the client performs model training locally in advance according to the collected user behavior characteristics of the user for the sample search words (search words entered by the user in history, and associated search words of the search words entered by the user in history), Obtain a trained user interest degree prediction model. The client analyzes user preferences in a new dimension. This new dimension (user behavior data for search terms) can better reflect the user's actual preferences. Therefore, the user interest prediction model trained by the client predicts the search results. It is more accurate, more biased towards the user's real preferences, and better improves the prediction effect of the model, so that when the client obtains the search results returned by the server, the trained user interest prediction model can be used to predict multiple Search results are predicted, and multiple search results are rearranged and displayed according to the predicted user interest, making the rearranged results more inclined to user preferences, making it easier for users to find their preferred targets, and improving user experience.
结合以上实施例，在一实施方式中，本发明实施例还提供了一种搜索结果展示方法。在该方法中，所述用户感兴趣度预测模型的训练样本还包括：样本搜索结果的用户行为特征；所述样本搜索结果是所述服务端返回的对应于目标样本搜索词的搜索结果，其中，目标样本搜索词为用户在样本搜索词(用户历史输入的搜索词，以及，用户历史输入的搜索词的联想搜索词)中确定出来的一个搜索词，以针对确定出来的搜索词进行搜索。所述目标样本搜索词为：所述用户历史输入的搜索词，或，所述用户历史输入的搜索词的联想搜索词。With reference to the above embodiments, in an implementation manner, the embodiment of the present invention also provides a method for displaying search results. In this method, the training samples of the user interest degree prediction model further include: user behavior characteristics of sample search results; the sample search results are search results corresponding to target sample search terms returned by the server, wherein , the target sample search term is a search term determined by the user among the sample search terms (the search term entered by the user in history, and the associated search term of the search term entered by the user in history), so as to search for the determined search term. The target sample search term is: a search term entered by the user in history, or an associative search term of the search term entered by the user in history.
本实施例中，为了进一步提高用户感兴趣度预测模型的预测准确度，避免分析用户偏好的数据太片面，客户端用于训练用户感兴趣度预测模型的训练样本除了样本搜索词的用户行为特征外，还包括有样本搜索结果的用户行为特征。In this embodiment, in order to further improve the prediction accuracy of the user interest degree prediction model and avoid too one-sided analysis of user preference data, the training samples used by the client to train the user interest degree prediction model except the user behavior characteristics of the sample search terms In addition, it also includes user behavior characteristics with sample search results.
其中，样本搜索结果为服务端针对目标样本搜索词返回的对应于目标样本搜索词的搜索结果，目标样本搜索词为客户端历史采集到的目标搜索词，目标样本搜索词为：用户历史输入的搜索词，或，用户历史输入的搜索词的联想搜索词。也就是说，样本搜索结果为客户端历史获取到的、服务端返回的对应于目标搜索词的搜索结果。Among them, the sample search result is the search result corresponding to the target sample search term returned by the server for the target sample search term, the target sample search term is the target search term collected in the history of the client, and the target sample search term is: The search term, or, the predicted search term for the search term entered by the user in the past. That is to say, the sample search result is the search result corresponding to the target search term that is obtained historically by the client and returned by the server.
本实施例中，用户使用客户端的搜索功能时，客户端展示搜索结果后，用户会对搜索结果进行相关操作，如对搜索结果进行浏览或点击。此时，客户端会记录下针对这些搜索结果的用户行为特征，每次记录下的搜索结果的用户行为特征均可以作为上述样本搜索结果的用户行为特征进行用户感兴趣度预测模型的训练。In this embodiment, when the user uses the search function of the client, after the client displays the search results, the user performs related operations on the search results, such as browsing or clicking on the search results. At this time, the client will record the user behavior characteristics of these search results, and the user behavior characteristics of each recorded search result can be used as the user behavior characteristics of the above sample search results to train the user interest degree prediction model.
在本实施例中，客户端通过样本搜索词的用户行为特征和样本搜索结果的用户行为特征一起训练用户感兴趣度预测模型，以对用户偏好进行多方位各角度的相关分析，避免分析用户偏好的数据太片面，从而进一步提高用户感兴趣度预测模型的预测准确度，对搜索结果重排起到更好的效果。In this embodiment, the client uses the user behavior characteristics of the sample search words and the user behavior characteristics of the sample search results to train the user interest degree prediction model to conduct multi-angle and various angle correlation analysis on user preferences, avoiding analyzing user preferences The data is too one-sided, so as to further improve the prediction accuracy of the user interest prediction model, and have a better effect on the rearrangement of search results.
结合以上实施例，在一实施方式中，本发明实施例还提供了一种搜索结果展示方法。在该方法中，除上述步骤外，该方法还可以包括步骤S21：With reference to the above embodiments, in an implementation manner, the embodiment of the present invention also provides a method for displaying search results. In this method, in addition to the above steps, the method may also include step S21:
步骤S21：获取对应于用户输入的搜索词的多个联想搜索词。Step S21: Obtain a plurality of associated search terms corresponding to the search term input by the user.
本实施例中，用户在使用客户端的搜索功能时，用户可以在客户端中输入搜索词，客户端确定用户输入的搜索词后可以针对用户输入的搜索词向服务端发送搜索词请求，以向服务端请求与用户输入的搜索词对应的多个联想搜索词。服务端在接收到该搜索词请求后，根据搜索词请求携带的用户输入的搜索词确定出对应于用户输入的搜索词的多个联想搜索词并返回给客户端，客户端获取到服务端返回的对应于用户输入的搜索词的多个联想搜索词。In this embodiment, when the user uses the search function of the client, the user can input a search word in the client, and after the client determines the search word input by the user, it can send a search word request to the server for the search word input by the user, so as to The server requests a plurality of predictive search terms corresponding to the search term input by the user. After receiving the search term request, the server determines a plurality of associated search terms corresponding to the search term input by the user according to the search term carried by the user in the search term request and returns them to the client, and the client obtains the search term returned by the server. A number of predicted search terms corresponding to the search term entered by the user.
步骤S22：通过所述用户感兴趣度预测模型，得到所述多个联想搜索词对应的用户感兴趣度。Step S22: Obtain user interest degrees corresponding to the plurality of associative search words through the user interest degree prediction model.
本实施例中，客户端本地预先训练好的用户感兴趣度预测模型还可以对联想搜索词进行用户感兴趣度的预估。客户端将服务器返回的与用户输入的搜索词对应的多个联想搜索词输入至该用户感兴趣度预测模型，进行模型预测，得到用户感兴趣度预测模型输出的该多个联想搜索词对应的用户感兴趣度。In this embodiment, the user interest degree prediction model pre-trained locally on the client side may also estimate the user interest degree for the associated search term. The client inputs a plurality of associative search words corresponding to the search words input by the user returned by the server into the user interest degree prediction model, performs model prediction, and obtains the output corresponding to the plurality of associative search words output by the user interest degree prediction model. user interest.
步骤S23：按照用户感兴趣度从高到低的顺序，对所述多个联想搜索词进行重排。Step S23: Rearrange the plurality of associative search terms in descending order of user interest.
本实施例中，可以理解，服务端返回的多个联想搜索词是按顺序排列的，客户端通过本地存储的用户感兴趣度预测模型，得到多个联想搜索词对应的用户感兴趣度后，可以按照用户感兴趣度从高到低的顺序，对该多个联想搜索词进行重新排列。In this embodiment, it can be understood that the multiple associated search words returned by the server are arranged in order, and the client obtains the user interest degrees corresponding to the multiple associated search words through the locally stored user interest degree prediction model. The plurality of associated search terms may be rearranged in descending order of user interest.
步骤S24：展示重排后的多个联想搜索词。Step S24: displaying the rearranged multiple associated search terms.
本实施例中，客户端对多个联想搜索词进行重新排列后，可以在客户端中展示该重排后的多个联想搜索词。In this embodiment, after the client rearranges the multiple associated search words, the rearranged multiple associated search words may be displayed on the client.
在本实施例中，客户端中训练的用户感兴趣度预测模型还可以服务端返回的多个联想搜索词进行用户感兴趣度的预估和重排，从而可以让用户在客户端中同时看见用户输入的搜索词和已经根据用户偏好重新排序的多个联想搜索词，从而使得用户更容易找到自己想要进行搜索的目标搜索词，让用户更容易找到自己的偏好目标，提高了用户体验，促进了搜索业务的发展。In this embodiment, the user interest degree prediction model trained in the client can also predict and rearrange the user interest degree with multiple associated search words returned by the server, so that the user can simultaneously see in the client The search terms entered by the user and multiple associated search terms that have been reordered according to user preferences make it easier for users to find the target search terms they want to search for, making it easier for users to find their preferred targets, and improving user experience. Promoted the development of the search business.
结合以上实施例，在一实施方式中，本发明实施例还提供了一种搜索结果展示方法。该方法中，除上述步骤外，该方法还可以包括步骤S31和S32：With reference to the above embodiments, in an implementation manner, the embodiment of the present invention also provides a method for displaying search results. In this method, in addition to the above steps, the method may also include steps S31 and S32:
步骤S31：在检测到用户对所述样本搜索词的操作的情况下，存储一条正训练样本，所述正训练样本包括以下至少一种数据：所述操作的发生时间、所述样本搜索词的特征、所述样本搜索词对应的业务类型。Step S31: In the case of detecting the user's operation on the sample search word, store a positive training sample, the positive training sample includes at least one of the following data: the occurrence time of the operation, the time of the sample search word feature, and the business type corresponding to the sample search term.
本实施例中，客户端在用户输入一次搜索词后针对样本搜索词的展示期间，检测到用户针对某个样本搜索词的操作的情况下，确定用户针对该样本搜索词对应的用户行为特征，并将该样本搜索词对应的用户行为特征作为一条正训练样本进行记录存储，其中，该正训练样本包括以下至少一种数据：用户针对样本搜索词的操作时的发生时间、样本搜索词的特征，样本搜索词对应的业务类型。In this embodiment, when the client detects the user's operation on a certain sample search word during the display of the sample search word after the user enters the search word once, it determines the user behavior characteristics corresponding to the sample search word. And record and store the user behavior feature corresponding to the sample search term as a positive training sample, wherein the positive training sample includes at least one of the following data: the occurrence time of the user's operation on the sample search term, the characteristics of the sample search term , the business type corresponding to the sample search term.
其中，上述操作可以是用户针对样本搜索词(用户历史输入的搜索词，以及，用户历史输入的搜索词的联想搜索词)的点击或确认操作，其中确认操作可以是语音控制确认操作等。本实施例的样本搜索词对应的业务类型可以指的是样本搜索词所代表的分类。Wherein, the above-mentioned operation may be the user's click or confirmation operation on the sample search words (the search words entered by the user in history, and the associated search words of the search words entered by the user in history), wherein the confirmation operation may be a voice control confirmation operation or the like. The service type corresponding to the sample search term in this embodiment may refer to the category represented by the sample search term.
步骤S32：在未检测到用户对所述样本搜索词的操作的情况下，存储一条负训练样本，所述负训练样本包括以下至少一种数据：展示所述样本搜索词的发生时间、所述样本搜索词的特征、所述样本搜索词对应的业务类型。Step S32: When no user operation on the sample search term is detected, store a negative training sample, the negative training sample includes at least one of the following data: showing the occurrence time of the sample search term, the The characteristics of the sample search term and the service type corresponding to the sample search term.
本实施例中，客户端在用户输入一次搜索词后针对样本搜索词的展示期间，未检测到用户针对一个或多个样本搜索词的操作的情况下，客户端针对每个未被用户进行上述操作的样本搜索词，确定用户针对每个样本搜索词对应的用户行为特征，并将每个样本搜索词对应的用户行为特征均作为一条负训练样本进行记录存储，其中，该负训练样本包括以下至少一种数据：客户端展示样本搜索词的发生时间、样本搜索词的特征、样本搜索词对应的业务类型。In this embodiment, when the client does not detect the user's operation on one or more sample search words during the presentation of the sample search words after the user inputs a search word, the client performs the above-mentioned The sample search terms of the operation determine the user behavior characteristics corresponding to each sample search term, and record and store the user behavior characteristics corresponding to each sample search term as a negative training sample, wherein the negative training sample includes the following At least one type of data: the client displays the occurrence time of the sample search term, the characteristics of the sample search term, and the business type corresponding to the sample search term.
在本实施例中，客户端在每次检测到用户针对样本搜索词的操作的用户行为特征作为正训练样本进行存储，客户端在每次未检测到用户针对样本搜索词的操作的用户行为特征作为负训练样本进行存储，从而持续采集用户在客户端中发生的针对样本搜索词的用户行为数据，以作为客户端内部训练用户感兴趣度预测模型的训练样本，从而训练出更加偏向用户喜好的模型。In this embodiment, the client detects the user behavior characteristics of the user's operation on the sample search word each time as a positive training sample, and the client does not detect the user behavior characteristics of the user's operation on the sample search word each time. It is stored as a negative training sample, so as to continuously collect user behavior data for sample search terms that occur in the client, and use it as a training sample for training the user interest prediction model inside the client, so as to train a more biased user preference. Model.
结合以上实施例，在一实施方式中，本发明实施例还提供了一种搜索结果展示方法。该方法中，除上述步骤外，该方法还可以包括步骤S41和S42：With reference to the above embodiments, in an implementation manner, the embodiment of the present invention also provides a method for displaying search results. In this method, in addition to the above steps, the method may also include steps S41 and S42:
步骤S41：在检测到用户对所述样本搜索结果的操作的情况下，存储一条正训练样本，所述正训练样本包括以下至少一种数据：所述点击操作的发生时间、所述样本搜索结果对应的标签特征、所述样本搜索结果的浏览时长、所述样本搜索结果对应的业务类型。Step S41: In the case of detecting the user's operation on the sample search result, store a positive training sample, the positive training sample includes at least one of the following data: the occurrence time of the click operation, the sample search result The corresponding tag feature, the browsing time of the sample search result, and the business type corresponding to the sample search result.
本实施例中，客户端在一次搜索的多个样本搜索结果的展示期间，检测到用户针对一个或多个样本搜索结果的操作的情况下，确定用户针对该样本搜索结果对应的用户行为特征，并将该样本搜索结果对应的用户行为特征作为一条正训练样本进行记录存储，其中，该正训练样本包括以下至少一种数据：用户针对样本搜索结果的点击操作时的发生时间、样本搜索结果对应的标签特征，样本搜索结果的浏览时长、样本搜索结果对应的业务类型。In this embodiment, when the client detects the user's operation on one or more sample search results during the display of multiple sample search results in one search, it determines the user behavior characteristics corresponding to the sample search results. And record and store the user behavior feature corresponding to the sample search result as a positive training sample, wherein the positive training sample includes at least one of the following data: the time when the user clicks on the sample search result, the corresponding time of the sample search result The label features of the sample search results, the browsing time of the sample search results, and the business type corresponding to the sample search results.
其中，本实施例的样本搜索结果的浏览时长为客户端进入样本搜索结果的详情页至返回的时长；样本搜索结果对应的业务类型可以指的是样本搜索结果的标签所代表的分类。Wherein, the browsing time of the sample search result in this embodiment is the time period from the client entering the details page of the sample search result to returning; the business type corresponding to the sample search result may refer to the category represented by the label of the sample search result.
步骤S42：在未检测到用户对所述样本搜索结果的操作的情况下，存储一条负训练样本，所述负训练样本包括以下至少一种数据：展示所述样本搜索结果的发生时间、所述样本搜索结果对应的标签特征、所述样本搜索结果为零的浏览时长、所述样本搜索结果对应的业务类型。Step S42: When no user operation on the sample search result is detected, store a negative training sample, the negative training sample includes at least one of the following data: the occurrence time of the sample search result, the The label feature corresponding to the sample search result, the browsing time when the sample search result is zero, and the business type corresponding to the sample search result.
本实施例中，客户端在一次搜索的多个样本搜索结果的展示期间，未检测到用户针对一个或多个样本搜索结果的操作的情况下，客户端针对每个未被用户进行点击操作的样本搜索结果，确定用户针对每个样本搜索结果对应的用户行为特征，并将每个样本搜索结果对应的用户行为特征均作为一条负训练样本进行记录存储，其中，该负训练样本包括以下至少一种数据：客户端展示样本搜索结果的发生时间、样本搜索结果对应的标签特征，样本搜索结果为零的浏览时长、样本搜索结果对应的业务类型。In this embodiment, when the client does not detect the user's operation on one or more sample search results during the display of multiple sample search results in one search, the client Sample search results, determine the user behavior characteristics corresponding to each sample search result, and record and store the user behavior characteristics corresponding to each sample search result as a negative training sample, wherein the negative training sample includes at least one of the following Types of data: The client displays the occurrence time of the sample search results, the label features corresponding to the sample search results, the browsing time when the sample search results are zero, and the business type corresponding to the sample search results.
在本实施例中，客户端在每次检测到用户针对样本搜索结果的点击操作的用户行为特征作为正训练样本进行存储，客户端在每次未检测到用户针对样本搜索结果的点击操作的用户行为特征作为负训练样本进行存储，从而持续采集用户在客户端中发生的针对样本搜索结果的用户行为数据，以作为客户端内部训练用户感兴趣度预测模型的训练样本，从而训练出更加偏向用户喜好的模型。In this embodiment, each time the client detects the user behavior characteristics of the user's click operation on the sample search results, it is stored as a positive training sample, and each time the client does not detect the user's click operation on the sample search results. Behavioral features are stored as negative training samples, so as to continuously collect user behavior data for sample search results that occur in the client, as training samples for training the user interest prediction model inside the client, so as to train more user-biased favorite model.
在一实施方式中，在用户感兴趣度预测模型的训练样本包括：样本搜索词的用户行为特征和样本搜索结果的用户行为特征时时，训练样本中的一条负训练样本可以包括以下至少一种数据：客户端展示样本搜索词的发生时间、样本搜索词的特征，样本搜索词对应的业务类型；或，一条负训练样本可以包括以下至少一种数据：客户端展示样本搜索结果的发生时间、样本搜索结果对应的标签特征，样本搜索结果为零的浏览时长、样本搜索结果对应的业务类型。In one embodiment, when the training samples of the user interest degree prediction model include: user behavior features of sample search terms and user behavior features of sample search results, a negative training sample in the training samples may include at least one of the following data : The client displays the occurrence time of the sample search term, the characteristics of the sample search term, and the business type corresponding to the sample search term; or, a negative training sample can include at least one of the following data: the occurrence time of the client display sample search result, the sample The label feature corresponding to the search result, the browsing time when the sample search result is zero, and the business type corresponding to the sample search result.
训练样本中的一条正训练样本可以至少包括以下至少一种数据：用户针对样本搜索词的操作时的发生时间、样本搜索词的特征，样本搜索词对应的业务类型；或，一条正训练样本可以包括以下至少一种数据：用户针对样本搜索结果的点击操作时的发生时间、样本搜索结果对应的标签特征，样本搜索结果的浏览时长、样本搜索结果对应的业务类型。A positive training sample in the training samples may include at least one of the following data: the occurrence time of the user's operation on the sample search term, the characteristics of the sample search term, and the business type corresponding to the sample search term; or, a positive training sample may At least one of the following data is included: the time when the user clicks on the sample search result, the tag feature corresponding to the sample search result, the browsing time of the sample search result, and the business type corresponding to the sample search result.
在一实施例中，用户感兴趣度预测模型的初始模型可以是一个线性回归器，用户感兴趣度预测模型的输入数据为：当前时间、搜索词的特征/搜索结果对应的标签特征，以及，搜索结果对应的业务类型/搜索词对应的业务类型；用户感兴趣度预测模型的输出数据为用户可能产生的浏览时间。而由于用于训练的样本搜索词没有浏览时间，因此，可以在训练时，将样本搜索词的正负训练样本中的样本搜索词的浏览时间的权重设置为最低，将样本搜索词的浏览时间对模型的影响减到最低，如可以将搜索词的浏览时间的权重设置为0以作为校准。In an embodiment, the initial model of the user interest degree prediction model may be a linear regressor, and the input data of the user interest degree prediction model is: the current time, the characteristics of the search word/the label characteristics corresponding to the search results, and, The business type corresponding to the search result/the business type corresponding to the search term; the output data of the user interest degree prediction model is the possible browsing time generated by the user. And because the sample search term used for training has no browsing time, therefore, during training, the weight of the browsing time of the sample search term in the positive and negative training samples of the sample search term can be set to the lowest, and the browsing time of the sample search term The impact on the model is minimized, for example, the weight of the browsing time of the search term can be set to 0 as a calibration.
结合以上任一实施例，在一实施方式中，本发明实施例还提供了一种搜索结果展示方法。在该方法中，除上述步骤外，该方法还可以包括步骤S51和S52，上述步骤S12可以具体包括步骤S53：In combination with any of the above embodiments, in an implementation manner, the embodiment of the present invention further provides a method for displaying search results. In this method, in addition to the above steps, the method may also include steps S51 and S52, and the above step S12 may specifically include step S53:
步骤S51：响应于用户的启动操作，拉取已存储的训练样本。Step S51: In response to the user's start-up operation, the stored training samples are pulled.
本实施例中，客户端响应于用户的启动操作，会从本地存储中拉取历史存储的训练样本，该训练样本包括正训练样本和负训练样本。In this embodiment, in response to the user's start operation, the client will pull historically stored training samples from the local storage, and the training samples include positive training samples and negative training samples.
步骤S52：利用所述已存储的训练样本，对所述用户感兴趣度预测模型的模型参数进行更新。Step S52: Using the stored training samples, update the model parameters of the user interest degree prediction model.
本实施例中，客户端拉取到本地历史存储的训练样本后，可以将历史存储的训练样本中的新的训练样本作为训练数据，对本地存储的最新的用户感兴趣度预测模型的模型参数进行更新，得到更新后的用户感兴趣度预测模型，该更新后的用户感兴趣度预测模型可以进行搜索结果或搜索词的用户感兴趣度的预估。In this embodiment, after the client pulls the training samples stored in the local history, the new training samples in the training samples stored in the history can be used as the training data, and the model parameters of the latest user interest degree prediction model stored locally The update is performed to obtain an updated user interest degree prediction model, and the updated user interest degree prediction model can estimate the user interest degree of search results or search words.
步骤S53：通过更新后的用户感兴趣度预测模型，得到所述多个搜索结果对应的用户感兴趣度。Step S53: Obtain the user interest degrees corresponding to the plurality of search results through the updated user interest degree prediction model.
本实施例中，客户端将服务器返回的对应于目标搜索词的多个搜索结果输入至该更新后的用户感兴趣度预测模型，进行模型预测，得到更新后的用户感兴趣度预测模型输出的该多个搜索结果各自对应的用户感兴趣度。In this embodiment, the client inputs a plurality of search results corresponding to the target search words returned by the server into the updated user interest degree prediction model, performs model prediction, and obtains the output of the updated user interest degree prediction model. The degree of user interest corresponding to each of the plurality of search results.
本实施例中，客户端可根据每一次用户在客户端中针对搜索词或搜索结果的用户行为数据对用户感兴趣度预测模型进行更新，是其预测结果更加倾向于用户个人的习惯和爱好，从而提升用户体验，促进业务发展。In this embodiment, the client can update the user interest degree prediction model according to the user behavior data of the search word or search result each time the user uses the client, so that the prediction results are more inclined to the user's personal habits and hobbies, Thereby improving user experience and promoting business development.
在一种实施例中，在客户端启动后，当客户端中检测到本地没有保存有用户感兴趣度预测模型时，客户端可以从本地存储中拉取历史存储的所有训练样本，该训练样本包括正训练样本和负训练样本。客户端拉取到本地历史存储的所有训练样本后，可以将历史存储的所有训练样本作为训练数据，进行模型训练，以训练得到用户感兴趣度预测模型，并将该用户感兴趣度预测模型保存在客户端的存储空间中。In one embodiment, after the client is started, when the client detects that there is no user interest degree prediction model stored locally, the client can pull all the training samples stored in the history from the local storage, the training samples Including positive training samples and negative training samples. After the client pulls all the training samples in the local historical storage, it can use all the training samples in the historical storage as training data for model training to obtain the user interest degree prediction model and save the user interest degree prediction model in the client's storage space.
在一实施方式中，搜索结果的展示方法可以包括以下步骤：In one embodiment, the method for displaying search results may include the following steps:
1、客户端在接收到用户输入的搜索词时，可以确定出该搜索词的多个联想词语，该联想词语为与用户输入的搜索词相关、相似的词语，联想词语会以列表的形式展示在客户端界面中；1. When the client receives the search term entered by the user, it can determine multiple associated words of the search term. The associated words are related and similar to the search term entered by the user, and the associated words will be displayed in the form of a list in the client interface;
2、用户通过直接在搜索栏旁边点击确认按键，以确定搜索本身输入的搜索词，或者，通过点击联想词语来搜索点击的联想词语，来确定自己想搜索的内容(即目标搜索词)后，客户端与服务器交互，接收服务器返回的目标搜索词所对应的搜索结果，并以列表的形式展示搜索结果；2. The user directly clicks the confirm button next to the search bar to confirm the search term entered by the search itself, or clicks the associated word to search for the clicked associated word to determine the content he wants to search (ie the target search term), The client interacts with the server, receives the search results corresponding to the target search terms returned by the server, and displays the search results in the form of a list;
3、用户可以对搜索结果的内容进行浏览，遇见想看的内容可以进行点击，点击后进入对应的详情页面，遇见不想看的内容可以不点击。3. Users can browse the content of the search results, click on the content they want to see, and enter the corresponding details page after clicking, and don’t need to click on the content they don’t want to see.
4、客户端将用户每一次点击确认按键时所输入的搜索词所对应的用户行为数据，和用户每一次点击的联想词语所对应的用户行为数据作为一条正样本，该正样本数据包括但不限于：点击发生的时间、词语的内容，词语对应的业务类型；此外，客户端还会将用户选择搜索词时，已经展示但未被点击的联想词或输入的搜索词的用户行为数据作为一条负样本，该负样本数据包括但不限于：展示发生的时间、词语的内容，词语对应的业务类型。4. The client takes the user behavior data corresponding to the search words entered each time the user clicks the confirmation button, and the user behavior data corresponding to the associated words each time the user clicks as a positive sample. The positive sample data includes but not Limited to: the time when the click occurred, the content of the word, and the business type corresponding to the word; in addition, the client will also use the associated words that have been displayed but not clicked or the user behavior data of the entered search word when the user selects the search word as a piece of information. Negative sample, the negative sample data includes but is not limited to: the time when the display occurs, the content of the word, and the business type corresponding to the word.
5、客户端将用户针对客户端所展示的搜索结果的点击行为数据会被作为一条正样本，该正样本数据包括但不限于：点击搜索结果发生的时间、搜索结果的标签词语、搜索结果的浏览时长、访问搜索结果的业务类型等；此外，在整个搜索功能使用结束后，其他未被点击的搜索结果所对应的数据会被作为一条负样本数据，该负样本数据包括但不限于：搜索结果展示发生的时间、搜索结果的标签词语、搜索结果的浏览时长为0、未访问的搜索结果的业务类型等。5. The client will regard the user's click behavior data on the search results displayed on the client as a positive sample, which includes but is not limited to: the time when the search result was clicked, the label words of the search result, the Browsing time, business type of access to search results, etc.; in addition, after the entire search function is used, the data corresponding to other unclicked search results will be used as a piece of negative sample data, which includes but is not limited to: search The time when the results are displayed, the tag words of the search results, the browsing time of the search results is 0, the business type of the search results that have not been visited, etc.
6、客户端在每次启动后，会拿到本地保存的所有正负样本作为训练数据进行模型训练，并将训练好的模型保存在本地存储空间中。而如果本地中已经保存了模型，则读取该模型，并根据新的训练样本对模型进行更新。其中，该模型是一个线性回归器，输入为当前时间、联想词特征或搜索结果的标签特征、业务类型(词语或搜索结果对应的业务类型)，输出为用户可能产生的浏览时间。而联想词样本没有浏览时间，因此，可以在训练时将搜索词样本数据的浏览时间的权重设置为最低，将其对模型的影响降到最低，将如将搜索词样本数据的浏览时间的权重为0以作为校准。6. After each startup, the client will get all the positive and negative samples stored locally as training data for model training, and save the trained model in the local storage space. And if the model has been saved locally, read the model and update the model according to the new training samples. Among them, the model is a linear regressor, the input is the current time, associative word features or label features of search results, business type (the business type corresponding to the words or search results), and the output is the possible browsing time generated by the user. The associative word sample has no browsing time, therefore, the weight of the browsing time of the search word sample data can be set to the lowest during training to minimize its impact on the model, such as setting the weight of the browsing time of the search word sample data 0 for calibration.
7、至此，用户在客户端使用搜索功能，客户端展示联想词时，客户端可以通过模型对联想词的列表数据中的每一条联想词进行预测，依照得到的预测结果按从大到小的排列方式将对应的联想词数据重新排列，重排后的联想词列表，即为从数据上按用户感兴趣值的排列。7. So far, when the user uses the search function on the client, when the client displays the associated words, the client can use the model to predict each associated word in the list data of the associated words, according to the obtained prediction results in order from large to small The arrangement method rearranges the corresponding associative word data, and the rearranged associative word list is an arrangement according to the user's interest value from the data.
8、同样，客户端在展示搜索结果时，可以通过模型对搜索结果的列表数据中的每一条搜索结果进行预测，依照得到的预测结果按从大到小的排列方式将对应的搜索结果数据重新排列，重排后的搜索结果列表，即为从数据上按用户感兴趣值的排列。8. Similarly, when the client displays the search results, it can use the model to predict each search result in the list data of the search results, and rearrange the corresponding search result data in descending order according to the obtained prediction results. Arrangement, the search result list after rearrangement is the arrangement according to the user's interest value from the data.
9、而每一次用户的搜索和后续的浏览行为都会更新客户端的该模型，使其预测结果更加倾向于用户个人的习惯和爱好，从而提升用户体验，促进业务发展。9. Each user's search and subsequent browsing behavior will update the model on the client side, making its prediction results more inclined to the user's personal habits and hobbies, thereby improving user experience and promoting business development.
本实施例中的服务端可以理解为服务器，该服务器可以是独立的服务器也能够是多个物理服务器工程的服务器集群或者分布式系统，还能够是提供云服务、云数据库、云计算、云函数、云存储、网络服务、云通信、中间件服务、域名服务、安全服务、CDN(ContentDelivery Network内容分发网络)、以及大数据和人工智能平台等基础云计算服务的云服务器。The server in this embodiment can be understood as a server, which can be an independent server or a server cluster or distributed system of multiple physical server projects, and can also provide cloud services, cloud databases, cloud computing, and cloud functions. Cloud servers for basic cloud computing services such as cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (ContentDelivery Network), and big data and artificial intelligence platforms.
需要说明的是，对于方法实施例，为了简单描述，故将其都表述为一系列的动作组合，但是本领域技术人员应该知悉，本发明实施例并不受所描述的动作顺序的限制，因为依据本发明实施例，某些步骤可以采用其他顺序或者同时进行。其次，本领域技术人员也应该知悉，说明书中所描述的实施例均属于优选实施例，所涉及的动作并不一定是本发明实施例所必须的。It should be noted that, for the method embodiment, for the sake of simple description, it is expressed as a series of action combinations, but those skilled in the art should know that the embodiment of the present invention is not limited by the described action sequence, because According to the embodiment of the present invention, certain steps may be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification belong to preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.
基于同一发明构思，本发明一实施例提供了一种搜索结果展示装置200，该搜索结果展示装置200可应用于客户端。参考图2，图2是本发明一实施例提供的搜索结果展示装置的结构框图。如图2所示，搜索结果展示装置200包括：Based on the same inventive concept, an embodiment of the present invention provides a search
第一获取模块201，用于获取服务端返回的对应于目标搜索词的多个搜索结果；The first obtaining
第一预测模块202，用于通过本地预先训练的用户感兴趣度预测模型，得到所述多个搜索结果对应的用户感兴趣度；所述用户感兴趣度预测模型的训练样本至少包括：样本搜索词的用户行为特征；所述样本搜索词包括：用户历史输入的搜索词，以及，用户历史输入的搜索词的联想搜索词；The
第一重排模块203，用于按照用户感兴趣度从高到低的顺序，对所述多个搜索结果进行重排；The
第一展示模块204，用于展示重排后的多个搜索结果。The
可选的，所述用户感兴趣度预测模型的训练样本还包括：样本搜索结果的用户行为特征；所述样本搜索结果是所述服务端返回的对应于目标样本搜索词的搜索结果，所述目标样本搜索词为：所述用户历史输入的搜索词，或，所述用户历史输入的搜索词的联想搜索词。Optionally, the training samples of the user interest degree prediction model further include: user behavior characteristics of sample search results; the sample search results are search results corresponding to target sample search terms returned by the server, and the The target sample search term is: a search term entered by the user in history, or an associative search term of the search term entered by the user in history.
可选的，所述装置200还包括：Optionally, the
第二获取模块，用于获取对应于用户输入的搜索词的多个联想搜索词；The second obtaining module is used to obtain a plurality of associated search words corresponding to the search words input by the user;
第二预测模块，用于通过所述用户感兴趣度预测模型，得到所述多个联想搜索词对应的用户感兴趣度；The second prediction module is used to obtain the user interest degree corresponding to the plurality of associative search words through the user interest degree prediction model;
第二重排模块，用于按照用户感兴趣度从高到低的顺序，对所述多个联想搜索词进行重排；The second rearrangement module is used to rearrange the plurality of associative search terms in order of user interest from high to low;
第二展示模块，用于展示重排后的多个联想搜索词。The second display module is used to display a plurality of rearranged associative search terms.
可选的，所述装置200还包括：Optionally, the
第一存储模块，用于在检测到用户对所述样本搜索词的操作的情况下，存储一条正训练样本，所述正训练样本包括以下至少一种数据：所述操作的发生时间、所述样本搜索词的特征、所述样本搜索词对应的业务类型；The first storage module is configured to store a positive training sample when detecting the user's operation on the sample search word, and the positive training sample includes at least one of the following data: the occurrence time of the operation, the The characteristics of the sample search term and the business type corresponding to the sample search term;
第二存储模块，用于在未检测到用户对所述样本搜索词的操作的情况下，存储一条负训练样本，所述负训练样本包括以下至少一种数据：展示所述样本搜索词的发生时间、所述样本搜索词的特征、所述样本搜索词对应的业务类型。The second storage module is used to store a negative training sample when no user operation on the sample search term is detected, and the negative training sample includes at least one of the following data: showing the occurrence of the sample search term Time, features of the sample search term, and service type corresponding to the sample search term.
可选的，所述装置200还包括：Optionally, the
第三存储模块，用于在检测到用户对所述样本搜索结果的操作的情况下，存储一条正训练样本，所述正训练样本包括以下至少一种数据：所述点击操作的发生时间、所述样本搜索结果对应的标签特征、所述样本搜索结果的浏览时长、所述样本搜索结果对应的业务类型；The third storage module is configured to store a positive training sample when the user's operation on the sample search result is detected, and the positive training sample includes at least one of the following data: the occurrence time of the click operation, the The tag feature corresponding to the sample search result, the browsing time of the sample search result, and the business type corresponding to the sample search result;
第四存储模块，用于在未检测到用户对所述样本搜索结果的操作的情况下，存储一条负训练样本，所述负训练样本包括以下至少一种数据：展示所述样本搜索结果的发生时间、所述样本搜索结果对应的标签特征、所述样本搜索结果为零的浏览时长、所述样本搜索结果对应的业务类型。The fourth storage module is used to store a negative training sample when no user operation on the sample search result is detected, and the negative training sample includes at least one of the following data: showing the occurrence of the sample search result time, the label feature corresponding to the sample search result, the browsing time when the sample search result is zero, and the business type corresponding to the sample search result.
可选的，所述装置200还包括：Optionally, the
第三获取模块，用于响应于用户的启动操作，拉取已存储的训练样本；The third acquisition module is used to pull the stored training samples in response to the user's startup operation;
模型更新模块，用于利用所述已存储的训练样本，对所述用户感兴趣度预测模型的模型参数进行更新；A model update module, configured to update the model parameters of the user interest degree prediction model by using the stored training samples;
所述第一预测模块202，包括：The
第一预测子模块，用于通过更新后的用户感兴趣度预测模型，得到所述多个搜索结果对应的用户感兴趣度。The first prediction submodule is configured to obtain the user interest degree corresponding to the plurality of search results through the updated user interest degree prediction model.
基于同一发明构思，本发明另一实施例提供一种电子设备300，如图3所示。图3是本发明一实施例示出的一种电子设备的示意图。该电子设备包括处理器301、存储器302及存储在存储器302上并可在处理器301上运行的计算机程序，所述计算机程序被所述处理器执行时实现本发明上述任一实施例所述的搜索结果展示方法中的步骤。示例的，上述任一实施例中的客户端可以是电子设备上运行的计算机程序。Based on the same inventive concept, another embodiment of the present invention provides an
基于同一发明构思，本发明另一实施例提供一种计算机可读存储介质，该计算机可读存储介质上存储有计算机程序，该计算机程序被处理器执行时实现如本发明上述任一实施例所述的搜索结果展示方法中的步骤。Based on the same inventive concept, another embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the computer program described in any one of the above-mentioned embodiments of the present invention can be realized. Steps in the above search result display method.
本说明书中的各个实施例均采用递进的方式描述，每个实施例重点说明的都是与其他实施例的不同之处，各个实施例之间相同相似的部分互相参见即可。Each embodiment in this specification is described in a progressive manner, each embodiment focuses on the difference from other embodiments, and the same and similar parts of each embodiment can be referred to each other.
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。It should be noted that, in this document, the term "comprising", "comprising" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article or apparatus comprising a set of elements includes not only those elements, It also includes other elements not expressly listed, or elements inherent in the process, method, article, or device. Without further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of additional identical elements in the process, method, article, or apparatus comprising that element.
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端(可以是手机，计算机，服务器，空调器，或者网络设备等)执行本发明各个实施例所述的方法。Through the description of the above embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus a necessary general-purpose hardware platform, and of course also by hardware, but in many cases the former is better implementation. Based on such an understanding, the essence of the technical solution of the present invention or the part that contributes to the prior art can be embodied in the form of software products, and the computer software products are stored in a storage medium (such as ROM/RAM, disk, CD) contains several instructions to make a terminal (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) execute the methods described in various embodiments of the present invention.
上面结合附图对本发明的实施例进行了描述，但是本发明并不局限于上述的具体实施方式，上述的具体实施方式仅仅是示意性的，而不是限制性的，本领域的普通技术人员在本发明的启示下，在不脱离本发明宗旨和权利要求所保护的范围情况下，还可做出很多形式，均属于本发明的保护之内。Embodiments of the present invention have been described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementations, and the above-mentioned specific implementations are only illustrative, rather than restrictive, and those of ordinary skill in the art will Under the enlightenment of the present invention, without departing from the gist of the present invention and the protection scope of the claims, many forms can also be made, all of which belong to the protection of the present invention.
本领域普通技术人员可以意识到，结合本发明实施例中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本发明的范围。Those of ordinary skill in the art can appreciate that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed in the embodiments of the present invention can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may use different methods to implement the described functions for each specific application, but such implementation should not be regarded as exceeding the scope of the present invention.
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。Those skilled in the art can clearly understand that for the convenience and brevity of the description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiment, which will not be repeated here.
在本发明所提供的实施例中，应该理解到，所揭露的装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。In the embodiments provided in the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or May be integrated into another system, or some features may be ignored, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
另外，在本发明各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit.
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本发明各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。If the functions described above are realized in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the essence of the technical solution of the present invention or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media capable of storing program codes such as U disk, mobile hard disk, ROM, RAM, magnetic disk or optical disk.
以上所述，仅为本发明的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本发明揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本发明的保护范围之内。因此，本发明的保护范围应以权利要求的保护范围为准。The above is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Anyone skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present invention. Should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims (14)
Translated fromChinesePriority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211733384.9ACN116108269A (en) | 2022-12-30 | 2022-12-30 | Search result display method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211733384.9ACN116108269A (en) | 2022-12-30 | 2022-12-30 | Search result display method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116108269Atrue CN116108269A (en) | 2023-05-12 |
Family
ID=86257481
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211733384.9APendingCN116108269A (en) | 2022-12-30 | 2022-12-30 | Search result display method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116108269A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116595256A (en)* | 2023-05-22 | 2023-08-15 | 毕加展览有限公司 | Method and system for data screening and immersive interaction of digital exhibition |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102523511A (en)* | 2011-11-09 | 2012-06-27 | 中国传媒大学 | Network program aggregation and recommendation system and network program aggregation and recommendation method |
| CN104462293A (en)* | 2014-11-27 | 2015-03-25 | 百度在线网络技术（北京）有限公司 | Search processing method and method and device for generating search result ranking model |
| CN109491516A (en)* | 2018-11-02 | 2019-03-19 | 北京字节跳动网络技术有限公司 | Input display methods, device, storage medium and the electronic equipment of associational word |
| CN112100529A (en)* | 2020-11-17 | 2020-12-18 | 北京三快在线科技有限公司 | Search content ordering method and device, storage medium and electronic equipment |
| CN113590917A (en)* | 2021-06-30 | 2021-11-02 | 五八有限公司 | Data searching method and device, electronic equipment and storage medium |
- 2022
- 2022-12-30CNCN202211733384.9Apatent/CN116108269A/enactivePending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102523511A (en)* | 2011-11-09 | 2012-06-27 | 中国传媒大学 | Network program aggregation and recommendation system and network program aggregation and recommendation method |
| CN104462293A (en)* | 2014-11-27 | 2015-03-25 | 百度在线网络技术（北京）有限公司 | Search processing method and method and device for generating search result ranking model |
| CN109491516A (en)* | 2018-11-02 | 2019-03-19 | 北京字节跳动网络技术有限公司 | Input display methods, device, storage medium and the electronic equipment of associational word |
| CN112100529A (en)* | 2020-11-17 | 2020-12-18 | 北京三快在线科技有限公司 | Search content ordering method and device, storage medium and electronic equipment |
| CN113590917A (en)* | 2021-06-30 | 2021-11-02 | 五八有限公司 | Data searching method and device, electronic equipment and storage medium |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116595256A (en)* | 2023-05-22 | 2023-08-15 | 毕加展览有限公司 | Method and system for data screening and immersive interaction of digital exhibition |
| CN116595256B (en)* | 2023-05-22 | 2023-11-03 | 毕加展览有限公司 | Method and system for data screening and immersive interaction of digital exhibition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| RU2725659C2 (en) | Method and system for evaluating data on user-element interactions | |
| US7941383B2 (en) | Maintaining state transition data for a plurality of users, modeling, detecting, and predicting user states and behavior | |
| US9171078B2 (en) | Automatic recommendation of vertical search engines | |
| RU2731335C2 (en) | Method and system for generating recommendations of digital content | |
| JP5638031B2 (en) | Rating method, search result classification method, rating system, and search result classification system | |
| US9934312B2 (en) | Recommendation search method for search engine, device and computer readable storage medium | |
| RU2580516C2 (en) | Method of generating customised ranking model, method of generating ranking model, electronic device and server | |
| JP5572596B2 (en) | Personalize the ordering of place content in search results | |
| RU2670494C2 (en) | Method for processing search requests, server and machine-readable media for its implementation | |
| CN103339623B (en) | Method and apparatus relating to internet searching | |
| KR101255406B1 (en) | Dynamic search with implicit user intention mining | |
| US8326826B1 (en) | Navigational resources for queries | |
| KR101330273B1 (en) | Context based resource relevance | |
| US20140372250A1 (en) | System and method for providing recommended content | |
| US9116992B2 (en) | Providing time series information with search results | |
| US20110208712A1 (en) | Automated tool for human assisted mining and capturing of precise results | |
| CN109168047B (en) | Video recommendation method and device, server and storage medium | |
| WO2016187705A1 (en) | System and method for ranking search results | |
| WO2010039537A2 (en) | Method and system for determining topical relatedness of domain names | |
| CN106447419B (en) | Visitor identification based on feature selection | |
| US20170161385A1 (en) | System And Method For Compiling Search Results Using Information Regarding Length Of Time Users Spend Interacting With Individual Search Results | |
| JP2015525929A (en) | Weight-based stemming to improve search quality | |
| US20120295633A1 (en) | Using user's social connection and information in web searching | |
| US20110029559A1 (en) | Method, apparatus, and program for extracting relativity of web pages | |
| JP4761460B2 (en) | Information search method, information search device, and information search processing program by search device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |