技术领域technical field
本发明涉及大数据以及电商领域，尤其涉及一种基于大数据的电商商品推荐方法。The invention relates to the fields of big data and e-commerce, in particular to a method for recommending e-commerce commodities based on big data.
背景技术Background technique
国内最大的电商平台淘宝网每日访问用户达6000万，每日在线商品数目已经超过了8亿件。面对急速增长的数据规模，用户正面临着“信息超载问题”，如果不借助于搜索引擎、推荐系统或者信息分类等辅助技术，用户从海量的互联网资源中找到自己真正感兴趣的信息是一件非常困难的事情,使得信息的有效利用率反而降低了。搜索引擎和个性化推荐系统是解决“信息超载”问题的两种手段。搜索引擎根据用户输入的关键字反馈给用户查询的结果，由于搜索引擎根据的是所有人的行为规律返回搜索结果，无法根据每个用户提供个性化服务，使得可能用户真正感兴趣的内容被海量的搜索结果所掩盖。个性化推荐在此问题上弥补了搜索引擎的不足，即代替用户评估其所有未看过的产品，并通过分析用户的兴趣爱好和历史行为，主动推荐符合用户喜好的项目。Taobao, the largest e-commerce platform in China, has 60 million daily visitors and more than 800 million daily online products. Facing the rapidly growing data scale, users are facing the "information overload problem". Without the help of auxiliary technologies such as search engines, recommendation systems, or information classification, it is a difficult task for users to find the information they are really interested in from massive Internet resources. This is a very difficult thing, which reduces the effective utilization of information. Search engines and personalized recommendation systems are two means to solve the "information overload" problem. The search engine feeds back the results of the user's query based on the keywords entered by the user. Since the search engine returns search results based on the behavior rules of all people, it cannot provide personalized services for each user, so that the content that the user is really interested in may be massively searched. covered by search results. Personalized recommendation makes up for the shortcomings of search engines on this issue, that is, instead of users evaluating all the products they have not seen, and by analyzing users' hobbies and historical behaviors, it actively recommends items that meet users' preferences.
在大数据时代下的推荐系统会面临海量的训练规模，传统单机环境下的推荐系统不能满足大数据时代推荐的需求。因此以分布式计算平台作为模型计算平台的推荐系统渐次诞生。进入Web2.0时代后，实时推荐的需求越来越多，而传统推荐系统，都是定期对数据进行分析，然后对模型进行更新，进而使用新的模型进行个性化推荐，训练效率低下，同时因为没有完善的机制配合对实时用户做出反馈，因此存在着推荐满意度以及交易转化率低下的问题。因此构建基于新型分布式流并行处理技术,能够分析实时用户行为并且做出实时推荐反馈的系统是非常有研究意义的。The recommendation system in the era of big data will face a massive training scale, and the recommendation system in the traditional stand-alone environment cannot meet the needs of recommendation in the era of big data. Therefore, the recommendation system that uses the distributed computing platform as the model computing platform is gradually born. After entering the era of Web 2.0, there are more and more demands for real-time recommendation, while the traditional recommendation system analyzes the data regularly, then updates the model, and then uses the new model for personalized recommendation, the training efficiency is low, and at the same time Because there is no perfect mechanism to cooperate with real-time user feedback, there are problems of recommendation satisfaction and low transaction conversion rate. Therefore, it is of great research significance to build a system based on new distributed stream parallel processing technology that can analyze real-time user behavior and make real-time recommendation feedback.
发明内容Contents of the invention
发明目的：Purpose of the invention:
针对上述问题，本发明提供一种基于大数据的电商商品推荐方法。In view of the above problems, the present invention provides a method for recommending e-commerce commodities based on big data.
技术方案：Technical solutions:
一种基于大数据的电商商品推荐方法，包括：数据采集模块、数据处理模块、数据推荐模块，所述方法包括以下步骤：A method for recommending e-commerce commodities based on big data, comprising: a data acquisition module, a data processing module, and a data recommendation module, the method comprising the following steps:
S010：数据采集模块根据用户输入的关键词搜索采集商品数据；S010: The data collection module searches and collects commodity data according to the keyword input by the user;
S020：根据用户所使用的终端所在位置的地域划分，数据采集模块采集该地域的对商品特征喜好的数据；S020: According to the geographical division of the location of the terminal used by the user, the data collection module collects the data on the characteristics and preferences of the products in this region;
S030：数据处理模块根据该地域基于以往的购买记录进行商品特征喜好数据的一次排序；S030: The data processing module performs a sorting of the commodity feature preference data based on the previous purchase records in the region;
S040：数据处理模块根据地域喜好特征数据的排序进行商品数据的筛选以及一次排序；S040: The data processing module performs screening and one-time sorting of commodity data according to the sorting of regional preference feature data;
S050：数据采集模块根据用户年龄所在的年龄段搜索采集该年龄段对商品特征喜好的数据；S050: The data collection module searches and collects the data on product characteristics preferences of the age group according to the age group of the user;
S060：数据处理模块根据数据采集模块针对用户年龄搜索采集的商品特征喜好的数据进行针对地域的商品喜好特征的数据筛选；S060: The data processing module performs data screening of product preference characteristics for regions according to the data of commodity characteristics and preferences collected by the data collection module for the user's age search;
S070：数据处理模块根据筛选结果对商品特征喜好数据进行二次排序；S070: The data processing module performs secondary sorting on the product feature preference data according to the screening result;
S080：数据处理模块根据商品特征喜好数据的二次排序进行商品数据二次排序；S080: The data processing module performs secondary sorting of commodity data according to the secondary sorting of commodity feature preference data;
S090：数据处理模块根据一次排序以及二次排序进行商品数据最终排序；S090: The data processing module performs the final sorting of the commodity data according to the primary sorting and the secondary sorting;
S100：数据推荐模块根据商品数据最终排序的结果进行商品推荐；S100: The data recommendation module recommends products according to the final sorting results of the product data;
作为本发明的一种优选方式，所述步骤S090包括以下步骤：As a preferred mode of the present invention, the step S090 includes the following steps:
S091：数据处理模块提取同时出现在一次排序以及二次排序的商品数据；S091: The data processing module extracts commodity data that appears in both the primary sorting and the secondary sorting;
S092：数据处理模块根据同一商品数据在一次排序以及二次排序中的排列位置计算最终排序位置。S092: The data processing module calculates the final sorting position according to the arrangement position of the same commodity data in the primary sorting and the secondary sorting.
作为本发明的一种优选方式，所述数据处理模块中设置有算法，所述算法用于计算商品特征喜好数据的一次排序、二次排序以及最终排序。As a preferred mode of the present invention, the data processing module is provided with an algorithm, and the algorithm is used to calculate the primary ranking, secondary ranking and final ranking of commodity feature preference data.
作为本发明的一种优选方式，对于步骤S100，当数据推荐模块进行商品数据推荐时，数据推荐模块将排名最靠前的五个商品数据以及最靠后的五个商品特征喜好数据进行推荐。As a preferred mode of the present invention, for step S100, when the data recommendation module recommends commodity data, the data recommendation module recommends the top five commodity data and the bottom five commodity feature preference data.
作为本发明的一种优选方式，还包括以下步骤：As a preferred mode of the present invention, the following steps are also included:
S110：数据处理模块根据用户对推荐商品的删除进行批量删除；S110: The data processing module deletes the recommended products in batches according to the user's deletion of the recommended products;
S120：数据处理模块进行推荐的补充；S120: The data processing module makes recommended supplements;
S130：数据推荐模块根据数据处理模块的推荐的补充进行二次商品数据推荐。S130: The data recommendation module recommends secondary commodity data according to the supplement recommended by the data processing module.
作为本发明的一种优选方式，数据处理装置根据用户删除的推荐商品数据的商品特征喜好数据进行批量删除。As a preferred mode of the present invention, the data processing device deletes in batches according to the commodity feature preference data of the recommended commodity data deleted by the user.
作为本发明的一种优选方式，当所述数据处理模块执行步骤S120时，数据处理模块提取未推荐的商品数据中在最终排序中最靠前或者最靠后的商品数据进行补充推送。As a preferred mode of the present invention, when the data processing module executes step S120, the data processing module extracts the product data that is the highest or the lowest in the final sorting among the unrecommended product data for supplementary push.
作为本发明的一种优选方式，所述方法还包括以下步骤：As a preferred mode of the present invention, the method also includes the following steps:
S140：数据采集模块采集用户该次购买的商品数据以及商品特征喜好数据；S140: the data collection module collects the commodity data purchased by the user this time and commodity feature preference data;
S150：数据处理模块将数据采集模块采集的该次购买数据写入大数据。S150: The data processing module writes the purchase data collected by the data acquisition module into big data.
本发明实现以下有益效果：The present invention realizes following beneficial effect:
1.通过地域、年龄段对用户喜好进行分析，并进行商品推荐。1. Analyze user preferences by region and age group, and make product recommendations.
2.通过定性、定量的分析影响不同用户购买行为的各种因素，判断每种因素在用户购买时对其影响的权重大小，从而选择合理的商品信息推送给用户，从而有效解决了仅根据用户浏览历史选择推送商品的单一化的、盲目的推送的方式，大大提高了电商对于不同层次消费人员的购买意向的把握，有效提高了电商在运营中的核心竞争力。2. Through qualitative and quantitative analysis of various factors that affect the purchase behavior of different users, determine the weight of each factor’s influence on the user’s purchase, so as to select reasonable product information to push to the user, thus effectively solving the problem of only relying on the user Browsing the history to choose a single and blind method of pushing products has greatly improved the e-commerce company's grasp of the purchase intentions of consumers at different levels, and effectively improved the core competitiveness of e-commerce companies in operation.
附图说明Description of drawings
此处的附图被并入说明书中并构成本说明书的一部分，示出了符合本公开的实施例，并于说明书一起用于解释本公开的原理。The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description serve to explain the principles of the disclosure.
图1为本发明工作步骤图；Fig. 1 is a working step diagram of the present invention;
图2为实施例一步骤图；Fig. 2 is a step diagram of embodiment one;
图3为系统框架图。Figure 3 is a system frame diagram.
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，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。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 only some, not all, embodiments of the present invention.
实施例一：Embodiment one:
参考图为图1-3。一种基于大数据的电商商品推荐方法，包括：数据采集模块、数据处理模块、数据推荐模块，所述方法包括以下步骤：The reference picture is Figure 1-3. A method for recommending e-commerce commodities based on big data, comprising: a data acquisition module, a data processing module, and a data recommendation module, the method comprising the following steps:
S010：数据采集模块根据用户输入的关键词搜索采集商品数据；S010: The data collection module searches and collects commodity data according to the keyword input by the user;
S020：根据用户所使用的终端所在位置的地域划分，数据采集模块采集该地域的对商品特征喜好的数据；S020: According to the geographical division of the location of the terminal used by the user, the data collection module collects the data on the characteristics and preferences of the products in this region;
S030：数据处理模块根据该地域基于以往的购买记录进行商品特征喜好数据的一次排序；S030: The data processing module performs a sorting of the commodity feature preference data based on the previous purchase records in the region;
S040：数据处理模块根据地域喜好特征数据的排序进行商品数据的筛选以及一次排序；S040: The data processing module performs screening and one-time sorting of commodity data according to the sorting of regional preference feature data;
S050：数据采集模块根据用户年龄所在的年龄段搜索采集该年龄段对商品特征喜好的数据；S050: The data collection module searches and collects the data on product characteristics preferences of the age group according to the age group of the user;
S060：数据处理模块根据数据采集模块针对用户年龄搜索采集的商品特征喜好的数据进行针对地域的商品喜好特征的数据筛选；S060: The data processing module performs data screening of product preference characteristics for regions according to the data of commodity characteristics and preferences collected by the data collection module for the user's age search;
S070：数据处理模块根据筛选结果对商品特征喜好数据进行二次排序；S070: The data processing module performs secondary sorting on the product feature preference data according to the screening result;
S080：数据处理模块根据商品特征喜好数据的二次排序进行商品数据二次排序；S080: The data processing module performs secondary sorting of commodity data according to the secondary sorting of commodity feature preference data;
S090：数据处理模块根据一次排序以及二次排序进行商品数据最终排序；S090: The data processing module performs the final sorting of the commodity data according to the primary sorting and the secondary sorting;
S100：数据推荐模块根据商品数据最终排序的结果进行商品推荐；S100: The data recommendation module recommends products according to the final sorting results of the product data;
作为本发明的一种优选方式，所述步骤S090包括以下步骤：As a preferred mode of the present invention, the step S090 includes the following steps:
S091：数据处理模块提取同时出现在一次排序以及二次排序的商品数据；S091: The data processing module extracts commodity data that appears in both the primary sorting and the secondary sorting;
S092：数据处理模块根据同一商品数据在一次排序以及二次排序中的排列位置计算最终排序位置。S092: The data processing module calculates the final sorting position according to the arrangement position of the same commodity data in the primary sorting and the secondary sorting.
作为本发明的一种优选方式，所述数据处理模块中设置有算法，所述算法用于计算商品特征喜好数据的一次排序、二次排序以及最终排序。As a preferred mode of the present invention, the data processing module is provided with an algorithm, and the algorithm is used to calculate the primary ranking, secondary ranking and final ranking of commodity feature preference data.
在具体实施过程中，针对用户所使用的终端所在位置的地域划分，在大数据中储存有在该地域划分中的商品特征数据，数据采集模块根据各个商品特征数据是否体现在用户输入的关键词对应的商品中判断是否采集该商品特征数据，若有体现，则采集该商品特征数据。数据处理模块根据被采集的商品特征数据的购买次数进行商品特征数据的一次排序，数据处理模块根据商品特征喜好数据进行商品数据的筛选，在筛选过后，数据处理模块根据商品特征喜好数据在以往的区域购买记录中的购买次数进行商品数据的一次排序。In the specific implementation process, for the geographical division of the location of the terminal used by the user, the product feature data in the geographical division is stored in the big data, and the data acquisition module is based on whether each product feature data is reflected in the keywords entered by the user. In the corresponding commodity, it is judged whether to collect the characteristic data of the commodity, and if so, collect the characteristic data of the commodity. The data processing module sorts the product feature data based on the purchase times of the collected product feature data. The data processing module screens the product data according to the product feature preference data. After the screening, the data processing module uses the product feature preference data in the previous The number of purchases in the regional purchase records is used to sort the product data once.
针对用户登录电商系统的账户，根据账户的出生日期判断用户的年龄以及该年龄所在的年龄段，数据采集模块从大数据中采集该年龄段中出现的商品特征，数据处理模块根据基于年龄段的商品特征数据以及基于地域的商品特征数据进行商品特征数据的筛选，数据处理模块根据该年龄段中某些商品特征被购买的次数进行商品特征喜好数据的二次排序，数据处理模块根据商品特征喜好数据的二次排序进行商品数据的筛选以及商品数据的二次排序。For the account of the user logging into the e-commerce system, the age of the user and the age group of the age are judged according to the date of birth of the account. The data collection module collects the characteristics of commodities appearing in this age group from the big data, and the data processing module based on the age group Commodity feature data and region-based product feature data are used to screen product feature data. The data processing module performs secondary sorting of product feature preference data according to the number of times certain product features are purchased in the age group. The data processing module is based on product features. Secondary sorting of preference data Screening of commodity data and secondary sorting of commodity data.
数据处理模块根据商品数据的一次排序以及二次排序进行商品数据的最终排序。The data processing module performs final sorting of commodity data according to the primary sorting and secondary sorting of commodity data.
对于商品数据的一次排序以及二次排序，数据处理模块计算不同商品特征数据量在总数据量中的占比，基于地域的商品特征喜好占比的排序即商品特征喜好数据的一次排序，基于年龄段的商品特征喜好占比的排序即商品特征喜好数据的二次排序。对于同一件商品在一次排序或者二次排序中会带有多种商品特征，因此数据处理模块中写入算法，算法如下：For the primary sorting and secondary sorting of commodity data, the data processing module calculates the proportion of different commodity feature data in the total data volume, and the sorting of commodity feature preference ratio based on region is the primary sorting of commodity feature preference data, based on age The sorting of the product feature preference ratio of the segment is the secondary sorting of the product feature preference data. For the same product, there will be multiple product features in the first sorting or secondary sorting, so the algorithm is written in the data processing module, and the algorithm is as follows:
Per=1-(1-a)*(1-b)*……*(1-n)，其中，Per为商品总占比，a、b...n为不同商品特征数据的占比。利用该算法计算得到的商品占比的排序即一次排序或者二次排序。Per=1-(1-a)*(1-b)*……*(1-n), where Per is the total proportion of products, and a, b...n are the proportions of different product characteristic data. The sorting of the commodity proportion calculated by this algorithm is primary sorting or secondary sorting.
对于最终排序，算法与上述算法一致，不同点在于Per为商品的最终占比，a为某商品在一次排序中的占比，b为同一件商品在二次排序中的占比。根据最终占比的排序即最终排序。For the final sorting, the algorithm is the same as the above algorithm, the difference is that Per is the final proportion of the product, a is the proportion of a certain product in the first sorting, and b is the proportion of the same product in the second sorting. The sorting according to the final proportion is the final sorting.
在数据推荐模块进行商品推荐时，数据推荐模块进行分榜单推荐，即分别根据商品数据一次排序、二次排序以及最终排序进行不同的推荐。When the data recommendation module recommends products, the data recommendation module recommends by list, that is, different recommendations are made according to the primary ranking, secondary ranking and final ranking of product data.
实施例二：Embodiment two:
参考图为图1、图3。针对实施例一，本实施例的不同点在于：Refer to Figure 1 and Figure 3. With regard to Embodiment 1, the difference of this embodiment lies in:
作为本发明的一种优选方式，对于步骤S100，当数据推荐模块进行商品数据推荐时，数据推荐模块将排名最靠前的五个商品数据以及最靠后的五个商品特征喜好数据进行推荐。As a preferred mode of the present invention, for step S100, when the data recommendation module recommends commodity data, the data recommendation module recommends the top five commodity data and the bottom five commodity feature preference data.
作为本发明的一种优选方式，还包括以下步骤：As a preferred mode of the present invention, the following steps are also included:
S110：数据处理模块根据用户对推荐商品的删除进行批量删除；S110: The data processing module deletes the recommended products in batches according to the user's deletion of the recommended products;
S120：数据处理模块进行推荐的补充；S120: The data processing module makes recommended supplements;
S130：数据推荐模块根据数据处理模块的推荐的补充进行二次商品数据推荐。S130: The data recommendation module recommends secondary commodity data according to the supplement recommended by the data processing module.
作为本发明的一种优选方式，数据处理装置根据用户删除的推荐商品数据的商品特征喜好数据进行批量删除。As a preferred mode of the present invention, the data processing device deletes in batches according to the commodity feature preference data of the recommended commodity data deleted by the user.
作为本发明的一种优选方式，当所述数据处理模块执行步骤S120时，数据处理模块提取未推荐的商品数据中在最终排序中最靠前或者最靠后的商品数据进行补充推送。As a preferred mode of the present invention, when the data processing module executes step S120, the data processing module extracts the product data that is the highest or the lowest in the final sorting among the unrecommended product data for supplementary push.
作为本发明的一种优选方式，所述方法还包括以下步骤：As a preferred mode of the present invention, the method also includes the following steps:
S140：数据采集模块采集用户该次购买的商品数据以及商品特征喜好数据；S140: the data collection module collects the commodity data purchased by the user this time and commodity feature preference data;
S150：数据处理模块将数据采集模块采集的该次购买数据写入大数据。S150: The data processing module writes the purchase data collected by the data acquisition module into big data.
在具体实施过程中，商品推荐分为基于商品数据的一次排序、二次排序以及最终排序进行推荐、基于商品特征喜好数据的一次排序、二次排序进行推荐。对于基于商品数据的排序，数据推荐模块直接将所有的商品按照正序以及倒序分别进行推荐，即用户可以直观的看到针对商品本身的商品排序，对于有目的性的用户来说比较方便。In the specific implementation process, commodity recommendation is divided into recommendations based on primary sorting, secondary sorting, and final sorting of commodity data, and recommendation based on primary sorting and secondary sorting of commodity feature preference data. For sorting based on product data, the data recommendation module directly recommends all products in forward and reverse order, that is, users can intuitively see the product sorting for the products themselves, which is more convenient for purposeful users.
针对基于商品特征喜好数据的排序，数据推荐模块选择将排名最靠前的5个商品特征以及最靠后的5个商品特征进行商品的推荐，在推荐过程中，数据推荐模块交叉的推荐不同的商品特征对应的商品。当用户对于某一种商品特征不感兴趣时，用户可以直接删除该商品特征对应的某几件商品，数据处理模块会根据该几件商品同时拥有的商品特征进行该商品特征数据的删除，在删除过后，数据推荐模块会选择在为推荐的商品中排名最靠前或者最靠后的商品进行推荐。For the sorting of preference data based on product features, the data recommendation module selects the top 5 product features and the bottom 5 product features for product recommendation. During the recommendation process, the data recommendation module cross-recommends different products The product corresponding to the product feature. When the user is not interested in a certain product feature, the user can directly delete some products corresponding to the product feature, and the data processing module will delete the product feature data according to the product features that these several products have at the same time. Afterwards, the data recommendation module will select the product that ranks the highest or the lowest among the recommended products for recommendation.
在用户选择并购买完商品时，数据采集模块采集用户该次购买的商品数据以及商品特征喜好数据，数据处理模块将上述的商品数据以及商品特征数据写入大数据。When the user selects and purchases the product, the data acquisition module collects the product data and product feature preference data purchased by the user this time, and the data processing module writes the above product data and product feature data into the big data.
上述实施例只为说明本发明的技术构思及特点，其目的是让熟悉该技术领域的技术人员能够了解本发明的内容并据以实施，并不能以此来限制本发明的保护范围。凡根据本发明精神实质所作出的等同变换或修饰，都应涵盖在本发明的保护范围之内。The above-mentioned embodiments are only to illustrate the technical concept and characteristics of the present invention, and its purpose is to enable those skilled in the technical field to understand the content of the present invention and implement it accordingly, and not to limit the protection scope of the present invention. All equivalent changes or modifications made according to the spirit of the present invention shall fall within the protection scope of the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810088583.6ACN110163693A (en) | 2018-01-30 | 2018-01-30 | A kind of electric business Method of Commodity Recommendation based on big data |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810088583.6ACN110163693A (en) | 2018-01-30 | 2018-01-30 | A kind of electric business Method of Commodity Recommendation based on big data |
| Publication Number | Publication Date |
|---|---|
| CN110163693Atrue CN110163693A (en) | 2019-08-23 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810088583.6APendingCN110163693A (en) | 2018-01-30 | 2018-01-30 | A kind of electric business Method of Commodity Recommendation based on big data |
| Country | Link |
|---|---|
| CN (1) | CN110163693A (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105809451A (en)* | 2016-02-29 | 2016-07-27 | 江苏大学 | Big data based e-commerce company evaluating, analyzing and predicting method and system for online shopping |
| CN106327227A (en)* | 2015-06-19 | 2017-01-11 | 北京航天在线网络科技有限公司 | Information recommendation system and information recommendation method |
| US20170188102A1 (en)* | 2015-12-23 | 2017-06-29 | Le Holdings (Beijing) Co., Ltd. | Method and electronic device for video content recommendation |
| CN107203518A (en)* | 2016-03-16 | 2017-09-26 | 阿里巴巴集团控股有限公司 | Method, system and device, the electronic equipment of on-line system personalized recommendation |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106327227A (en)* | 2015-06-19 | 2017-01-11 | 北京航天在线网络科技有限公司 | Information recommendation system and information recommendation method |
| US20170188102A1 (en)* | 2015-12-23 | 2017-06-29 | Le Holdings (Beijing) Co., Ltd. | Method and electronic device for video content recommendation |
| CN105809451A (en)* | 2016-02-29 | 2016-07-27 | 江苏大学 | Big data based e-commerce company evaluating, analyzing and predicting method and system for online shopping |
| CN107203518A (en)* | 2016-03-16 | 2017-09-26 | 阿里巴巴集团控股有限公司 | Method, system and device, the electronic equipment of on-line system personalized recommendation |
| Publication | Publication Date | Title |
|---|---|---|
| CN107562818B (en) | Information recommendation system and method | |
| CN102629257B (en) | Commodity recommending method of e-commerce website based on keywords | |
| CN112632405B (en) | Recommendation method, recommendation device, recommendation equipment and storage medium | |
| CN109918563B (en) | Book recommendation method based on public data | |
| CN107220365B (en) | Precise recommendation system and method based on collaborative filtering and parallel processing of association rules | |
| Sivapalan et al. | Recommender systems in e-commerce | |
| CN109684538A (en) | A kind of recommended method and recommender system based on individual subscriber feature | |
| KR102227552B1 (en) | System for providing context awareness algorithm based restaurant sorting personalized service using review category | |
| CN109190043A (en) | Recommended method and device, storage medium, electronic equipment and recommender system | |
| CN102411754A (en) | Personalized recommendation method based on commodity property entropy | |
| CN107391687A (en) | A kind of mixing commending system towards local chronicle website | |
| US20040049514A1 (en) | System and method of searching data utilizing automatic categorization | |
| CN102663627A (en) | Personalized recommendation method | |
| CN111460251A (en) | Data content personalized push cold start method, device, equipment and storage medium | |
| US20130173583A1 (en) | Keyword index pruning | |
| US9552415B2 (en) | Category classification processing device and method | |
| CN118172138A (en) | Intelligent recommendation system for electronic commerce | |
| Chen et al. | Product recommendation system for small online retailers using association rules mining | |
| CN116431895A (en) | Personalized recommendation method and system for safety production knowledge | |
| CN111199422A (en) | Improved LFM Collaborative Filtering Method with Temporal Decay Factors | |
| Das et al. | Leveraging collaborative tagging for web item design | |
| Perugini et al. | A connection-centric survey of recommender systems research | |
| Singh et al. | Implementation of knowledge based collaborative filtering and machine learning for e-commerce recommendation system | |
| CN111581505B (en) | Flight recommendation method and system based on combined recommendation | |
| CN110163693A (en) | A kind of electric business Method of Commodity Recommendation based on big data |
| 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 | ||
| RJ01 | Rejection of invention patent application after publication | Application publication date:20190823 | |
| RJ01 | Rejection of invention patent application after publication |