Detailed Description
In order to make the present application better understood by those skilled in the art, the following description will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings, and it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
It should be understood that in the present application, "at least one (item)" means one or more, and "a plurality" means two or more. "and/or" is used to describe an association relationship of an associated object, and indicates that three relationships may exist, for example, "a and/or B" may indicate that only a exists, only B exists, and three cases of a and B exist simultaneously, where a and B may be singular or plural. The character "/" generally indicates that the context-dependent object is an "or" relationship. "at least one of" or the like means any combination of these items, including any combination of single item(s) or plural items(s). For example, at least one of a, b or c may represent a, b, c, "a and b", "a and c", "b and c", or "a and b and c", wherein a, b, c may be single or plural.
In a conventional information recommendation method, information is generally recommended based on a label, a category or a keyword of information, and when the label, the category or the keyword of information to be recommended is the same as the label, the category or the keyword of information browsed by a target user, the information to be recommended is recommended to the target user. However, the inventor finds that the prior information recommendation method is not accurate or fine enough due to the coarse granularity of the labels, the categories or the keywords of the information, and has no diversity and expansibility of information recommendation and poor information recommendation effect.
Taking news as an example, when the target user browses the news on the market of the company a (i.e. the information browsed by the target user), in the conventional information recommendation method, because the labels of the news on the market of the company a are set as economy and company a, the labels are set to be not fine enough, all the news labeled as economy and company a are recommended to the target user when the information recommendation is performed, however, the target user actually pays attention to the news on the market of the company a, which results in inaccurate and fine information recommendation, and the news on the market of other companies or the news on the market of the company a or the news on the economy of other companies cannot be recommended to the target user, so that the diversity and the expansibility of the information recommendation cannot be satisfied, and the information recommendation effect is poor.
In order to solve the problem, the embodiment of the application provides an information recommendation method and an information recommendation device, which are used for determining information characteristics of information based on information browsing records of a plurality of users, wherein the information characteristics can reflect the correlation between corresponding information and other information in the browsing records of the users, and the user characteristics of the target users can be determined based on the information characteristics of each information in the information browsing records of the target users in a first preset time period, so that the information browsing tendency and trend of the target users can be reflected. Therefore, based on the similarity between the information characteristics of the information and the user characteristics of the target user, whether the information is related to the browsed information of the target user can be reflected, and whether the information accords with the information browsing tendency and trend of the user or not can be determined. And then, according to the similarity between the information characteristics of the information and the user characteristics of the target user, accurately recommending the information to the target user. In addition, the information recommending method and the information recommending device provided by the embodiment of the application are not limited by information types or labels, and the recommended information has diversity, so that the information recommending effect is improved.
For example, one of the scenarios of the embodiments of the present application may be applied to the scenario shown in fig. 1. The scenario may include a server 101 and a terminal 102. The user can browse information using the terminal 102, and the server 101 obtains information characteristics of the information and user characteristics of the target user based on information browsing records of a plurality of users and information browsing records of the target user, and recommends information to the target user according to the information characteristics of the information and the user characteristics of the target user.
In the above application scenario, although the actions of the embodiments of the present application are described as being performed by the server 101, these actions may also be performed by the client 102, or may also be performed in part by the client 102, in part by the server 101. The present application is not limited to the execution subject, and may be executed by performing the operations disclosed in the embodiments of the present application. Those skilled in the art will appreciate that the frame diagram shown in fig. 1 is but one example in which embodiments of the present application may be implemented. The scope of applicability of the embodiments of the application is not limited in any way by the framework.
It should be noted that the terminal 102 in embodiments of the present application may be any user device that is present, under development or future developed that is capable of interacting with the server 101 via any form of wired and/or wireless connection (e.g., wi-Fi, LAN, cellular, coaxial cable, etc.), including but not limited to, present, under development or future developed smartphones, non-smartphones, tablet computers, laptop personal computers, desktop personal computers, minicomputers, midrange computers, mainframe computers, and the like. It should also be noted that server 101 in embodiments of the present application may be one example of an existing, developing or future developed device capable of providing information recommendation services to users. Embodiments of the application are not limited in this respect.
It should be noted that, the information recommendation method and apparatus provided by the embodiment of the present application are applicable to various types of information recommendation, such as personalized recommendation. Recommended information includes, but is not limited to, news, merchandise, audio video, websites, entertainment, books, articles, and applications, among others.
The information recommendation method provided by the embodiment of the application will be described in detail below with reference to the accompanying drawings.
Referring to fig. 2, the flow chart of an information recommendation method provided by the embodiment of the application is shown.
The information recommendation method provided by the embodiment of the application comprises the following steps:
S201, acquiring user characteristics of a target user and information characteristics of information.
In the embodiment of the application, the information characteristics are obtained based on the correlation of the information in the information browsing records of a plurality of users, and the user characteristics are obtained based on the information characteristics of each information in the information browsing records of the corresponding users in a first preset time period. It should be noted that, the information browsing record includes, but is not limited to, any one or more of clicking, browsing, collecting and praying by the user, and the information may be included in the information browsing record as long as the user performs an operation on the information. The following operations refer to actions performed by a user on each piece of information in his information browsing record, including but not limited to clicking, browsing, collecting, praying, purchasing, etc., which are not listed here.
The information characteristic of the target information can reflect the correlation between the target information and other information that the user has operated when operated by the user. The user characteristics are obtained based on the information characteristics of each piece of information in the information browsing record corresponding to the user in the first preset time period, and the information operation trend and/or trend of the user can be reflected. Therefore, the correlation between the information and the information operation trend and/or trend of the target user can be determined by utilizing the user characteristics of the target user and the information characteristics of the information, whether the information is recommended to the target user or not is determined, and accurate recommendation of the information is realized.
In some possible implementations of the embodiments of the present application, the information browsing records of multiple users may be regarded as one text (i.e. reference text), and because the context of the target information in the text, that is, other information in the same information browsing record as the target information, the context feature can reverse the correlation between the target information and other information that the user has operated when operated by the user, the information feature of the information may include the text feature of the information in the text. I.e. the information features comprise the contextual features of the corresponding information in a reference text, which is obtained based on information browsing records of a plurality of users. In practical applications, any manner may be used to obtain a contextual feature of the information in the reference text, such as a word vector algorithm, etc., where the contextual feature may be represented by a feature vector, which is not limited herein. The information features of how to obtain the information will be described in detail with reference to a specific example, which is not described in detail.
In some possible implementations of the embodiments of the present application, when the information feature is a feature vector, the user feature is obtained by:
and taking the average value of the information characteristics of each piece of information in the information browsing record of the corresponding user in the first preset time period as the user characteristic of the corresponding user, or taking the average value of the information characteristics of each piece of information in the information browsing record of the corresponding user in the first preset time period multiplied by the corresponding coefficient as the user characteristic of the corresponding user.
In the embodiment of the present application, the information browsing record corresponding to the user may include information browsed in a last period of time (i.e. the first preset time period), or may include only a few pieces of information browsed in the last period of time, which is not limited herein. It should be noted that, the first preset time period may be set according to a specific application scenario and actual needs, because the information operation interest of the user has a certain timeliness, the first preset time period should not be set too long or too long as the current time interval. As an example, assuming that the current time is the kth day, the first preset time period may be from the kth-1 day to the kth-n day, where n is a positive integer, such as n=15.
In practical application, the coefficient corresponding to the information feature of each piece of information can be set according to specific needs to determine the tendency of information recommendation, which is not listed here. It will be appreciated that when an information feature is represented as a feature vector, the user feature is also a feature vector.
S202, recommending information to the target user according to the similarity of the information characteristics and the user characteristics of the target user.
It will be appreciated that, since the information characteristics of the information can reflect the correlation between the target information and other information that the user has browsed when operated by the user, and the user characteristics of the target user can reflect the information operation tendency and/or trend of the target user, the more similar the information characteristics of the information and the user characteristics of the target user, the more consistent the information operation tendency and/or trend of the target user. Therefore, according to the similarity between the information characteristics of the information and the user characteristics of the target user, whether the information accords with the information operation tendency and/or trend of the target user or not can be determined, and the accuracy and precision of information recommendation can be improved by recommending the information to the target user based on the similarity.
In addition, when the information is recommended, the recommended information is not limited by parameters such as labels, categories and the like, but is recommended based on the correlation between the information and other information when the information is operated and the information operation tendency and trend of the target user, so that on one hand, the recommended information has diversity, on the other hand, the information can be recommended by adapting to the development of the information operation interest of the target user, and the information recommendation effect is improved.
In practical application, any feature similarity calculation method may be used to obtain the similarity between the information feature and the user feature of the target user, for example, the similarity between the feature vectors and the distance representative feature is not limited herein.
In practical application, when the information feature and the user feature are feature vectors, at least two possible implementations exist in step S202, which are described below one by one:
in a first possible implementation manner, as shown in fig. 3, step S202 may specifically include:
S301, obtaining recommendation parameters according to the distance between the user characteristics of the target user and the information characteristics of the information to be recommended.
In the embodiment of the application, the distance between the recommendation parameter and the user characteristic of the target user and the information characteristic of the information to be recommended is in a negative correlation relationship, and the distance between the recommendation parameter and the user characteristic of the target user and the information characteristic of the information to be recommended can be represented by cosine similarity or Euclidean distance between feature vectors. It can be understood that the smaller the distance between the user feature of the target user and the information feature of the information to be recommended is, the larger the recommendation parameter is, and the smaller the euclidean distance between the user feature of the target user and the information feature of the information to be recommended is, the larger the recommendation parameter is. The specific form of the recommended parameters can be set according to actual needs, and are not listed here.
S302, judging whether the recommended parameter is larger than a third preset threshold value, and executing step S303 when the recommended parameter is larger than the third preset threshold value.
And S303, recommending the information to be recommended to the target user.
It is understood that in a first possible implementation, information that the recommended parameter is greater than the third preset threshold is recommended to the target user. In practical applications, the third preset threshold and the specific recommendation mode may be set according to actual needs, which is not limited herein.
In a second possible implementation manner, as shown in fig. 4, step S202 may specifically include:
S401, obtaining recommendation parameters of each piece of information to be recommended according to the user characteristics of the target user and the distance between the information characteristics of each piece of information to be recommended in the information set to be recommended.
In a second possible implementation, similar to the first possible implementation, the recommendation parameter and the distance between the user feature of the target user and the information feature of the information to be recommended are in a negative correlation, and the distance between the recommendation parameter and the user feature of the target user and the information feature of the information to be recommended can be represented by cosine similarity between feature vectors or euclidean distance. It can be understood that the smaller the distance between the user feature of the target user and the information feature of the information to be recommended is, the larger the recommendation parameter is, and the smaller the euclidean distance between the user feature of the target user and the information feature of the information to be recommended is, the larger the recommendation parameter is. The specific form of the recommended parameters can be set according to actual needs, and are not listed here.
S402, sorting the recommendation parameters of each piece of information to be recommended, and recommending the information to be recommended corresponding to the largest first N recommendation parameters to the target user.
It is understood that N is a positive integer. In a second possible implementation, the first N pieces of information with the largest recommendation parameters are recommended to the target user. In practical application, the value of N and the specific recommended mode may be set according to the actual needs, which is not limited herein.
In the embodiment of the application, the information characteristics of each information can be determined based on the information browsing records of a plurality of users, and the user characteristics of the target user can be obtained based on the information characteristics of each information in the information browsing records of the target user in a first preset time period. And then, according to the similarity between the information characteristics of the information and the user characteristics of the target user, accurately recommending the information to the target user. Because the information features reflect the correlation of the information with other information in the information browsing records of a plurality of users, and the user features of the target users comprehensively reflect the information features of the information browsed by the target users and represent the information browsing tendency and trend of the target users, when the similarity of the information features and the user features is higher, the information corresponding to the information features is indicated to be consistent with the information browsing tendency and trend of the target users, and the information is recommended to the target users, so that the accuracy of information recommendation can be improved. In addition, the information recommending method and the information recommending device provided by the embodiment of the application are not limited by information types or labels, and the recommended information has diversity, so that the information recommending effect is improved.
The information characteristics of how the information is obtained will be described in detail with reference to a specific example.
Referring to fig. 5, the flow chart of a method for obtaining information features according to an embodiment of the present application is shown.
In some possible implementations of the embodiments of the present application, the information features may be obtained specifically by using the following steps:
s501, acquiring information browsing records of a plurality of users in a second preset time period.
In the embodiment of the present application, the duration of the second preset time period and the duration of the first preset time period may be equal or different, and may specifically be set according to actual needs, which is not limited herein. It may be appreciated that, in the embodiment of the present application, the information browsing records of multiple users in the second preset time period are used as references for information recommendation, so that in order to ensure the effect of information recommendation, it should be ensured that enough information browsing record data is obtained, for example, the number of users is enough, and/or the data in the information browsing records is enough, which are not listed one by one.
S502, constructing an information relation network based on information browsing records of a plurality of users.
In the embodiment of the application, the nodes in the information relation network are in one-to-one correspondence with the information in the information browsing records of a plurality of users, and the connecting lines between the two nodes in the information relation network represent the corresponding two pieces of information which are simultaneously present in the information browsing records of the same user. Each node in the information relation network represents one piece of information in the acquired information browsing records of a plurality of users, when two pieces of information appear in the information browsing records of the same user, a connecting line exists between the nodes corresponding to the two pieces of information, otherwise, when the pieces of information do not appear in the information browsing records of the same user, no connecting line exists between the nodes corresponding to the two pieces of information. It will be appreciated that links in the information-relation network represent correlations between information that the same user has browsed.
For example, user U1's information navigational record includes information A, B and C, user U2's information navigational record includes information C, B, F and D, and user U3's information navigational record includes information B and E, as shown in Table 1 below:
TABLE 1 information browsing records of multiple users in a first preset time period
Based on this, an information relationship network including nodes A, B, C, D, E and F can be constructed, with nodes A, B, C, D, E and F corresponding to information A, B, C, D, E and F, respectively.
Then, in some possible designs, because both information A, B and C appear in the information-handling record of user U1, there is a connection between the corresponding nodes A, B and C in the information-handling network, because both information C, B, F and D appear in the information-handling record of user U2, there is a connection between the corresponding nodes C, B, F and D in the information-handling network, and because both information B and E appear in the information-handling record of user U3, there is a connection between the corresponding nodes B and E in the information-handling network. The information-relation network constructed may be as shown in fig. 6 a.
In some possible implementation manners of the embodiments of the present application, the connection manner between nodes in the information relationship network may be adaptively adjusted according to the actual application scenario and the specific situation. In practical applications, there are at least two possible implementations:
In a first possible implementation manner, the development trend of the browsing information of the user, that is, the development of the information operation interest of the user, may be comprehensively considered, and the connection lines between the nodes in the constructed information relationship network may be determined according to the operation sequence of the user on the information in the information browsing record, where the determined connection lines may or may not have directions.
In one example, a connection between two nodes in the information-relation network represents that information corresponding to the two nodes appears in the information browsing record of the same user at the same time, and the information corresponding to the two nodes is sequentially operated by the same user. That is, when two pieces of information are sequentially operated by the same user, the nodes corresponding to the two pieces of information are connected.
Continuing with the above example, assuming that the order of the information in the information browsing record in the above example is the order of the user's operation on the information, there are no directional links between nodes a and B, nodes B and C, nodes B and E, nodes B and F, and nodes F and D in the constructed information relationship network, as shown in fig. 6B.
In another example, the connection line between two nodes in the information relation network represents that the information corresponding to the two nodes appears in the information browsing record of the same user at the same time, and the information corresponding to the two nodes is sequentially operated by the same user. When two pieces of information are sequentially operated by the same user, connecting lines are formed between the nodes corresponding to the two pieces of information, and the connecting line direction points from the node corresponding to the information operated by the same user to the node corresponding to the information operated by the user.
With continued reference to the above example, assuming that the order of the information in the information browsing record in the above example is the order in which the user operates the information, in the constructed information relationship network, there is a connection from node a to node B, a connection from node B to node C, a connection from node C to node B, a connection from node B to node F, a connection from node F to node D, and a connection from node B to node E, as shown in fig. 6C.
In a second possible implementation manner, the timeliness of the user on the information operation may also be considered, when determining the connection line between the nodes in the constructed information relationship network, two pieces of information in the information browsing record of the same user are disconnected from each other if the number of pieces of information of the time interval or the interval between the browsing times of the user on the two pieces of information exceeds a certain threshold, and the connection line exists between the nodes corresponding to the two pieces of information if the number of pieces of information of the time interval or the interval between the browsing times of the user on the two pieces of information does not exceed a certain threshold.
In one example, a connection line between two nodes in the information-relation network represents that information corresponding to the two nodes simultaneously appears in an information browsing record of the same user, and a time interval between operation times of the same user on the information corresponding to the two nodes is smaller than a first preset threshold. That is, when the time interval between the operation times of the same user on two pieces of information is smaller than the first preset threshold, the two pieces of information are connected between the corresponding nodes in the information relation network. The connection may be non-directional or may have a direction determined by the order of operation of the information by the same user.
Continuing with the above example, assuming that the order of the information in the information browsing record in the above example is the order in which the user operates the information, and only the time interval between the user U2's operation time on the information C and the information D exceeds the first preset threshold, in the constructed information relationship network, a connection exists between the nodes A, B and C, a connection exists between the nodes C, B and F, a connection exists between the nodes B, F and D, and a connection exists between the nodes B and E, as shown in fig. 6D.
In practical application, the first preset threshold may be set according to a specific application scenario and actual needs, which is not limited herein.
In some possible designs, the connection in the information-relation network may only exist between the nodes corresponding to two pieces of information sequentially operated by the same user, or may exist between the nodes corresponding to two pieces of information whose operation time of the same user is less than the first preset threshold, which may be specifically referred to the description in the first possible implementation manner and will not be repeated.
In another example, the connection line between two nodes in the information relationship network represents that the information corresponding to the two nodes appears in the information browsing record of the same user at the same time, and the number of information browsed by the same user between browsing the information corresponding to the two nodes is smaller than a second preset threshold. That is, when the same user has operated less than the second preset threshold between operations on two pieces of information, the two pieces of information are connected between corresponding nodes in the information relation network.
Continuing with the above example, assuming that the order of the information in the information browsing record in the above example is the operation order of the user on the information, and the second preset threshold is 1, in the constructed information relationship network, there is no connection between the nodes corresponding to the information C and the information D, there is a connection between the nodes A, B and C, there is a connection between the nodes C, B and F, there is a connection between the nodes B, F and D, and there is a connection between the nodes B and E, as shown in fig. 6D.
In practical application, the second preset threshold may be set according to a specific application scenario and actual needs, which is not limited herein.
In some possible designs, when building the relational network, not only the number of information operated between the operations of the two information by the same user may not exceed the second preset threshold, but also the time interval of the operation time of the two information by the same user may not exceed the first preset threshold at the same time, and the specific implementation may be referred to above related description. Similarly, the connection in the information relation network may only exist between the nodes corresponding to two pieces of information sequentially operated by the same user, or may also exist between the nodes corresponding to two pieces of information with the same user operation time less than the first preset threshold, and specifically, reference may be made to the description in the first possible implementation manner, which is not repeated.
It should be further noted that the foregoing is illustrative of various ways of constructing the information-relation network. Because the links between nodes in the information relation network represent the correlation between the corresponding information, the expression of the information characteristics is affected. Therefore, in practical application, in order to improve accuracy and effect of information recommendation, whether the correlation between the information has a logical relationship in browsing time and sequence can be determined according to a specific application scene, and whether the corresponding nodes are connected when the information relationship network is constructed can be determined based on the logical relationship.
For example, when the recommended information is news or commodity, the user has certain timeliness and order of browsing interests of the news or commodity, and when the information relation network is constructed, whether connection lines exist between nodes in the information relation network can be judged by considering the timeliness and order of browsing interests of the news or commodity of the user, for example, the information relation network is constructed by adopting the first implementation mode or the second implementation mode. For example, when the recommended information is music, the interests of the user are relatively fixed, and when the information relation network is constructed, two pieces of information which are simultaneously appeared in the information browsing record of the same user are directly connected with corresponding nodes in the information relation network, and timeliness and sequence of the interests of the user are ignored.
As can be seen from the above example, two of the information may appear in the information browsing records of a plurality of users, for example in the example of table 1, information B and information C are included in the information browsing records of user U1 and user U2. In some possible implementation manners, the weight can be set for the connection line between the two corresponding nodes based on the occurrence times of the two information in the information browsing records of the plurality of users, so that the correlation between the information is better reflected, and the accuracy of information recommendation is improved. It will be appreciated that the higher the weight of a link, the greater the correlation between the two information corresponding to that link. Fig. 6e illustrates an information-relation network comprising link weights, the numbers on the links representing their corresponding weights, taking the example shown in table 1 as an example.
S503, executing at least one random walk in the information relation network according to the preset walk parameters to obtain each walk path.
In the embodiment of the application, random walk takes any node in the information relation network as a starting point, and random walk traversal is carried out along the connecting line between the nodes. The walk parameter may set the longest number of steps of the walk (corresponding to the number of nodes passed), the number of times each node serves as a start point, the probability of the walk to each node, and the like. For example, the longest number of steps may be set to 4, and the number of times each node is set to 2 as the start point. For another example, the policy of information recommendation can be controlled by setting the probability of the node that walks again to the previous step, the probability of the node that walks to other nodes connected to the previous step, and the probability of the node that walks to other nodes, whether the random walk policy is biased to the depth-first search or the breadth-first search. It should be noted that, when a weight is set on a connection line of the information relationship network, the weight also affects the probability that the connection line is moved.
Taking the information relation network shown in fig. 6e, the longest walking step number is set to 4, and the number of times each node is set to 2 as a starting point as an example, table 2 below shows that each walking path is obtained by a plurality of random walks as an example.
TABLE 2 multiple travel paths
And S504, constructing a reference text based on each obtained walking path.
In the embodiment of the application, after each node in each obtained travelling path is arranged in sequence, each node is replaced by information parameters (such as identification, source, category and the like) of corresponding information, so as to obtain a reference text, thereby representing the operation condition of a user on the information.
In some possible designs, step S504 may specifically include:
and replacing the nodes in the first text with one information parameter of the corresponding information to obtain a reference text.
In the embodiment of the application, the information parameters can be identity, category, source, keywords (such as source website and author) or personalized tags. In practical application, the nodes in the first text can be replaced with information parameters of corresponding information to obtain a plurality of reference texts, the following operations are executed based on each reference text to obtain a plurality of context characteristics of the information, and each context characteristic is synthesized to obtain information characteristics of the corresponding information.
And S505, obtaining the context characteristics of the information corresponding to the nodes in the information relation network in the reference text.
In the embodiment of the present application, any contextual feature obtaining method may be used to obtain contextual features of information in a reference text, such as a skip-gram model of a word2vec frame, and the like, which will not be described herein. The obtained context features can be represented by feature vectors, and the dimension of the feature vectors can be set according to actual needs. In some possible designs, when multiple reference texts are obtained based on multiple information parameters, one context feature is obtained for each reference text corresponding to the information parameter. In one example, different dimensions may be set for the context features corresponding to the respective information parameters according to actual needs, and the dimensions of any two obtained context features may be the same or different. For example, when the context feature is a feature vector and the context feature is a plurality, the dimension of the feature vector may be set based on the contribution degree of the information parameter corresponding to each context feature at the time of information recommendation, the higher the contribution degree, the higher the dimension of the corresponding feature vector.
And S506, obtaining information characteristics of the corresponding information based on the context characteristics.
In the embodiment of the present application, the context feature may be directly used as the information feature of the corresponding information, or the context feature may be multiplied by a certain coefficient or calculated based on a preset formula to obtain the information feature of the corresponding information, which is not limited herein.
In some possible implementations of the embodiments of the present application, when the context feature is a feature vector and the context features are plural, step S506 may specifically include:
And splicing the context characteristics of each information parameter in the reference text to obtain the information characteristics of the corresponding information, or splicing the context characteristics of each information parameter in the reference text after multiplying the corresponding weight to obtain the information characteristics of the corresponding information.
It will be appreciated that assuming that 3 context features are obtained, (a, b, c), (d, e, f, g) and (h, i, j), respectively, the information features that are spliced may be (a, b, c, d, e, f, g, h, i, j). When the weights corresponding to the context features are α, β and γ, respectively, the information features obtained by stitching may be (αa, αb, αc, βd, βe, βf, βg, γh, γi, γj). In the embodiment of the present application, the tendency represented by the feature vector may be set by using the weight, and the weight corresponding to each context feature may be set according to actual needs, which is not listed here.
The above details how to obtain the information characteristics of each information in the information browsing records of multiple users, and in practical application, as the information increases, although the information characteristics of each information can be updated timely according to the browsing information of the user, there may still be information (i.e. cold information) not included in the information browsing records of the multiple users. In order to obtain information features of the cold information in order to recommend the cold information to a user, in some possible implementations of embodiments of the application, information features of the cold information may be obtained based on information features of other information similar to the cold information.
Specifically, as shown in fig. 7, for cold information not included in the information browsing records of a plurality of users, the information characteristics of the cold information are obtained by:
and S701, determining reference information with the same information parameters as the cold information in the information browsing records of a plurality of users.
In the embodiment of the application, at least one same information parameter exists between the reference information and the cold information, for example, the category is the same, the source is the same, and the like. It will be appreciated that the information parameter may be an identity, category, source, keyword or personalized tag, etc., and is not limited herein.
And S702, obtaining the information characteristic of the cold information according to the information characteristic of the reference information.
In an embodiment of the application, the information features of the cold information are identified with the information features of the reference information that are similar to the cold information. The method for obtaining the information features of the reference information may specifically be any one of the above descriptions, which is not repeated here. As an example, when the reference information is one, the information feature of the reference information may be regarded as the information feature of the cold information, and when the reference information is a plurality of and the information feature is a feature vector, the average value of the information features of the respective reference information may be regarded as the information feature of the cold information, which is not limited.
Based on the information recommendation method provided by the embodiment, the embodiment of the application also provides an information recommendation device.
Referring to fig. 8, the structure of an information recommendation device according to an embodiment of the present application is shown.
The information recommending device provided by the embodiment of the application comprises:
a feature acquisition module 801, configured to acquire user features of a target user and information features of information;
The information recommending module 802 is configured to recommend information to a target user according to the similarity between the information feature and the user feature;
the information characteristic is obtained based on the information characteristic of each information in the information browsing records of the corresponding users in a first preset time period.
In some possible implementation manners of the embodiment of the application, the information features can specifically include the context features of the corresponding information in the reference text, and the reference text can be obtained based on information browsing records of a plurality of users.
In some possible implementations of the embodiment of the present application, the feature obtaining module 801 may specifically include:
the record acquisition sub-module is used for acquiring information browsing records of a plurality of users in a second preset time period;
The network construction sub-module is used for constructing an information relation network based on information browsing records of a plurality of users, wherein nodes in the information relation network correspond to the information in the information browsing records one by one, and a connecting line between two nodes in the information relation network represents that two pieces of information corresponding to the two nodes are simultaneously present in the information browsing records of the same user;
the random walk sub-module is used for executing at least one random walk in the information relation network according to preset walk parameters, and each obtained walk path is obtained;
the text construction sub-module is used for constructing a reference text based on each obtained walking path;
The first acquisition sub-module is used for acquiring the contextual characteristics of the information corresponding to the nodes in the information relation network in the reference text;
And the second acquisition sub-module is used for obtaining the information characteristics of the corresponding information based on the context characteristics.
In some possible implementation manners of the embodiments of the present application, a connection line between two nodes in an information relationship network represents that information corresponding to the two nodes simultaneously appears in an information browsing record of the same user, and the information corresponding to the two nodes is sequentially operated by the same user;
Or the connection line between two nodes in the information relation network represents that the information corresponding to the two nodes simultaneously appears in the information browsing record of the same user, and the information corresponding to the two nodes is sequentially operated by the same user, wherein the connection line between the two nodes in the information relation network points to the direction of one node, and the direction of the connection line is determined by the operation sequence of the same user on the information corresponding to the two nodes connected with the connection line.
In some possible implementation manners of the embodiments of the present application, a connection line between two nodes in an information relationship network represents that information corresponding to the two nodes simultaneously appears in an information browsing record of the same user, and a time interval between operation times of the same user on the information corresponding to the two nodes is smaller than a first preset threshold;
and/or, the connection line between two nodes in the information relation network represents that the information corresponding to the two nodes simultaneously appears in the information browsing record of the same user, and the information quantity browsed by the same user between the information corresponding to the two nodes is smaller than a second preset threshold value.
In some possible implementations of the embodiments of the present application, the text construction sub-module may specifically include an arrangement sub-module and a replacement sub-module;
The arrangement sub-module is used for sequentially arranging the nodes in each travel path to obtain a first text;
and the replacing sub-module is used for replacing the node in the first text with one information parameter of the corresponding information to obtain the reference text.
Optionally, the information parameter is an identity, a category, a source, a keyword or a personalized tag.
In some possible implementation manners of the embodiment of the present application, when the context is characterized by a feature vector and is plural, the second obtaining sub-module may specifically include a first splicing sub-module or a second splicing sub-module;
the first splicing sub-module is used for splicing the context characteristics of each information parameter in the reference text to obtain the information characteristics of the corresponding information;
And the second splicing sub-module is used for multiplying the context characteristics of each information parameter in the reference text by the corresponding weight and then splicing the context characteristics to obtain the information characteristics of the corresponding information.
In some possible implementations of the embodiments of the present application, when the context feature is a feature vector and is plural, the dimensions of the context feature corresponding to different information parameters are not all the same.
In some possible implementations of the embodiments of the present application, the feature obtaining module 801 may further include:
The system comprises a determining submodule, a judging submodule and a judging submodule, wherein the determining submodule is used for determining reference information with the same information parameters as cold information in information browsing records of a plurality of users;
And the third acquisition sub-module is used for obtaining the information characteristic of the cold information according to the information characteristic of the reference information.
In some possible implementations of the embodiments of the present application, when the information feature is a feature vector, the feature obtaining module 801 may further include a first computing sub-module or a second computing sub-module;
the first computing sub-module is used for taking the average value of the information characteristics of each piece of information in the information browsing records of the corresponding user in a first preset time period as the user characteristic of the corresponding user;
And the second calculation sub-module is used for multiplying the information characteristic of each piece of information in the information browsing record of the corresponding user in the first preset time period by the corresponding coefficient to obtain an average value as the user characteristic of the corresponding user.
In some possible implementations of the embodiment of the present application, when the information feature and the user feature are feature vectors, the information recommendation module 802 may specifically include:
The parameter determination submodule is used for obtaining recommended parameters according to the distance between the user characteristics of the target user and the information characteristics of the information to be recommended;
The judging sub-module is used for judging whether the recommended parameter is larger than a third preset threshold value or not;
And the recommending sub-module is used for recommending the information to be recommended to the target user when the judging sub-module judges that the recommending parameter is larger than a third preset threshold value.
In some possible implementations of the embodiment of the present application, when the information feature and the user feature are feature vectors, the information recommendation module 802 may specifically include:
the parameter determination submodule is used for obtaining the recommendation parameters of each piece of information to be recommended according to the user characteristics of the target user and the distance between the information characteristics of each piece of information to be recommended in the information set to be recommended;
The recommendation sub-module is used for sequencing the recommendation parameters of each piece of information to be recommended, recommending the information to be recommended corresponding to the first N maximum recommendation parameters to the target user, wherein N is a positive integer.
In the embodiment of the application, the information characteristics of each information can be determined based on the information browsing records of a plurality of users, and the user characteristics of the target user can be obtained based on the information characteristics of each information in the information browsing records of the target user in a first preset time period. And then, according to the similarity between the information characteristics of the information and the user characteristics of the target user, accurately recommending the information to the target user. Because the information features reflect the correlation of the information with other information in the information browsing records of a plurality of users, and the user features of the target users comprehensively reflect the information features of the information browsed by the target users and represent the information browsing tendency and trend of the target users, when the similarity of the information features and the user features is higher, the information corresponding to the information features is indicated to be consistent with the information browsing tendency and trend of the users, and the information is recommended to the target users, so that the accuracy of information recommendation can be improved. In addition, the information recommending method and the information recommending device provided by the embodiment of the application are not limited by information types or labels, and the recommended information has diversity, so that the information recommending effect is improved.
FIG. 9 is a block diagram illustrating an apparatus 900 for information recommendation, according to an example embodiment. For example, apparatus 900 may be a mobile phone, computer, digital broadcast terminal, messaging device, game console, tablet device, medical device, exercise device, personal digital assistant, or the like.
Referring to FIG. 9, apparatus 900 may include one or more of a processing component 902, a memory 904, a power component 906, a multimedia component 908, an audio component 910, an input/output (I/O) interface 912, a sensor component 914, and a communication component 916.
The processing component 902 generally controls overall operations of the apparatus 900, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing component 902 may include one or more processors 920 to execute instructions to perform all or part of the steps of the methods described above. Further, the processing component 902 can include one or more modules that facilitate interaction between the processing component 902 and other components. For example, the processing component 902 can include a multimedia module to facilitate interaction between the multimedia component 908 and the processing component 902.
The memory 904 is configured to store various types of data to support operations at the device 900. Examples of such data include instructions for any application or method operating on the device 900, contact data, phonebook data, messages, pictures, videos, and the like. The memory 904 may be implemented by any type of volatile or nonvolatile memory device or combination thereof, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disk.
The power supply component 906 provides power to the various components of the device 900. Power supply components 906 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for device 900.
The multimedia component 908 comprises a screen between the device 900 and the user that provides an output interface. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensor may sense not only the boundary of a touch or sliding action, but also the duration and pressure associated with the touch or sliding operation. In some embodiments, the multimedia component 908 includes a front-facing camera and/or a rear-facing camera. The front-facing camera and/or the rear-facing camera may receive external multimedia data when the device 900 is in an operational mode, such as a shooting mode or a video mode. Each front camera and rear camera may be a fixed optical lens system or have focal length and optical zoom capabilities.
The audio component 910 is configured to output and/or input audio signals. For example, the audio component 910 includes a Microphone (MIC) configured to receive external audio signals when the device 900 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may be further stored in the memory 904 or transmitted via the communication component 916. In some embodiments, the audio component 910 further includes a speaker for outputting audio signals.
The I/O interface 912 provides an interface between the processing component 902 and peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, a home button, a volume button, an activate button, and a lock button.
The sensor assembly 914 includes one or more sensors for providing status assessment of various aspects of the apparatus 900. For example, the sensor assembly 914 may detect the on/off state of the device 900, the relative positioning of the components, such as the display and keypad of the apparatus 900, the sensor assembly 914 may also detect the change in position of the apparatus 900 or one component of the apparatus 900, the presence or absence of user contact with the apparatus 900, the orientation or acceleration/deceleration of the apparatus 900, and the change in temperature of the apparatus 900. The sensor assembly 914 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 914 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 914 may also include an acceleration sensor, a gyroscopic sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.
The communication component 916 is configured to facilitate communication between the apparatus 900 and other devices in a wired or wireless manner. The device 900 may access a wireless network based on a communication standard, such as WiFi,2G, or 3G, or a combination thereof. In one exemplary embodiment, the communication component 916 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, the communication component 916 further includes a Near Field Communication (NFC) module to facilitate short range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, ultra Wideband (UWB) technology, bluetooth (BT) technology, and other technologies.
In an exemplary embodiment, apparatus 900 may be implemented by one or more Application Specific Integrated Circuits (ASICs), digital Signal Processors (DSPs), digital Signal Processing Devices (DSPDs), programmable Logic Devices (PLDs), field Programmable Gate Arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components for executing the methods described above.
In an exemplary embodiment, a non-transitory computer readable storage medium is also provided, such as a memory 904 including instructions executable by the processor 920 of the apparatus 900 to perform the above-described method. For example, the non-transitory computer readable storage medium may be ROM, random Access Memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.
A non-transitory computer readable storage medium, which when executed by a processor of a mobile terminal, causes the mobile terminal to perform any one of the information recommendation methods provided by the above embodiments.
Fig. 10 is a schematic structural diagram of a server according to an embodiment of the present application. The server 1000 may vary considerably in configuration or performance and may include one or more central processing units (central processing units, CPUs) 1022 (e.g., one or more processors) and memory 1032, one or more storage mediums 1030 (e.g., one or more mass storage devices) storing applications 1042 or data 1044. Wherein memory 1032 and storage medium 1030 may be transitory or persistent. The program stored on the storage medium 1030 may include one or more modules (not shown), each of which may include a series of instruction operations on a server. Further, central processor 1022 may be configured to communicate with storage medium 1030 to perform a series of instruction operations in storage medium 1030 on server 1000.
The server 1000 may also include one or more power supplies 1026, one or more wired or wireless network interfaces 1050, one or more input/output interfaces 1058, one or more keyboards 1056, and/or one or more operating systems 1041, such as Windows Server, mac OS XTM, unixTM, linuxTM, freeBSDTM, and the like.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative elements and steps are described above generally in terms of functionality in order to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. The terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises an element.
The above description is only of the preferred embodiment of the present application, and is not intended to limit the present application in any way. While the application has been described with reference to preferred embodiments, it is not intended to be limiting. Any person skilled in the art can make many possible variations and modifications to the technical solution of the present application or modifications to equivalent embodiments using the methods and technical contents disclosed above, without departing from the scope of the technical solution of the present application. Therefore, any simple modification, equivalent variation and modification of the above embodiments according to the technical substance of the present application still fall within the scope of the technical solution of the present application.