User interest point estimation method and related equipment

Abstract

The embodiment of the invention provides a user interest point estimation method and related equipment. The user interest point estimation method comprises the steps of obtaining a target historical data text, wherein the target historical data text is a historical data text associated with a user; industry label extraction is conducted on the target historical data texts, user interest point labels are obtained, and each user interest point label at least corresponds to one target historical data text; according to the user interest point label, determining corresponding multi-dimensional label associated data; and obtaining a user interest point result according to the multi-dimensional label association data. According to the technical scheme provided by the embodiment of the invention, the interest degree of the user in each interest point label can be more accurately determined according to the multi-dimensional label association data so as to obtain the user interest point result, and meanwhile, the user interest point estimation can be carried out only by needing the historical data text of the related user, so that the user interest point estimation efficiency is improved. A large amount of sample data is not needed, and the cost of interest point prediction is reduced.

Description

Technical Field

[0001] The present application relates to the field of computer and communication technology, and in particular to a method for estimating user interest points and related equipment. Background Art

[0002] As the growth curve of PC-side Internet business becomes slower and slower, refined operations are becoming more and more important for business growth. However, the ultimate factor for refined operations is the user's interest points. Traditional user interest point prediction requires a lot of complex calculations, or directly uses a prediction model, but the prediction requires a lot of samples and resources for training, otherwise the predicted interest point results are not accurate enough. Summary of the invention

[0003] The embodiments of the present application provide a method for estimating user interest points and related devices, thereby overcoming the problem that the accuracy of the interest point prediction results in the prior art is not high enough, at least to a certain extent.

[0004] Other features and advantages of the present application will become apparent from the following detailed description, or may be learned in part by the practice of the present application.

[0005] According to one aspect of an embodiment of the present application, a method for estimating user interest points is provided, the method comprising: obtaining a target historical data text, the target historical data text being a historical data text associated with the user; performing industry label extraction on the target historical data text to obtain user interest point labels, each user interest point label corresponding to at least one target historical data text; determining corresponding multi-dimensional label association data based on the user interest point labels; and obtaining user interest point results based on the label association data.

[0006] In some embodiments of the present application, the industry label extraction of the target historical data text to obtain user interest point labels specifically includes: extracting industry labels from each target historical data text to obtain corresponding text industry labels; clustering the text industry labels to obtain user interest point labels.

[0007] In some embodiments of the present application, the multi-dimensional tag association data includes a user diversity parameter, and determining the corresponding multi-dimensional tag association data based on the user interest point tags specifically includes: determining the number of target historical data texts of different contents corresponding to the user interest point tags; and determining the user diversity parameter based on the number of target historical data texts of different contents corresponding to the user interest point tags.

[0008] In some embodiments of the present application, the multi-dimensional tag association data includes user frequency data, and determining the corresponding multi-dimensional tag association data based on the user interest point tag specifically includes: obtaining a tag text set, the tag text set including all target historical data texts corresponding to the user interest point tag; determining the frequency data corresponding to each target historical data text in the tag text set; and determining the user frequency data based on the frequency data corresponding to each target historical data text in the tag text set.

[0009] In some embodiments of the present application, the multi-dimensional tag association data includes a target weight, and determining the corresponding multi-dimensional tag association data based on the user interest point tag specifically includes: determining the inverse text frequency corresponding to each target historical data text in the tag text set; obtaining the text weight corresponding to each target historical data text based on the inverse text frequency and frequency data corresponding to each target historical data text in the tag text set; obtaining the target weight based on the text weight corresponding to each target historical data text in the tag text set.

[0010] In some embodiments of the present application, the multi-dimensional tag association data includes an industry time factor, and determining the corresponding multi-dimensional tag association data based on the user interest point tag specifically includes: determining the tag time, the tag time being the generation time of each target historical data text corresponding to the user interest point tag; determining the industry time factor based on the tag time.

[0011] In some embodiments of the present application, obtaining the user interest point results based on the multi-dimensional tag association data specifically includes: obtaining the interest point scores corresponding to the user interest point tags based on the tag association data; and sorting the user interest point tags based on the interest point scores to obtain the interest point results.

[0012] According to one aspect of an embodiment of the present application, a user interest point estimation device is provided, and the user interest point estimation device includes: a text acquisition module, used to acquire a target historical data text, wherein the target historical data text is a historical data text associated with the user; a label extraction module, used to extract industry labels from the target historical data text to obtain user interest point labels, each user interest point label corresponds to at least one target historical data text; a data analysis module, used to determine corresponding multi-dimensional label association data based on the user interest point labels; and a result determination module, used to obtain user interest point results based on the multi-dimensional label association data.

[0013] In some embodiments of the present application, the label extraction module specifically includes: a label extraction submodule, which is used to extract industry labels from each of the target historical data texts to obtain corresponding text industry labels; and a label clustering submodule, which is used to cluster the text industry labels to obtain user interest point labels.

[0014] In some embodiments of the present application, the multi-dimensional tag association data includes user diversity parameters, and the data analysis module specifically includes: a quantity determination submodule, used to determine the number of target historical data texts with different contents corresponding to the user interest point tags; a parameter determination submodule, used to determine the user diversity parameters according to the number of target historical data texts with different contents corresponding to the user interest point tags.

[0015] In some embodiments of the present application, the multi-dimensional tag association data includes user frequency data, and the data analysis module specifically includes: a text collection submodule, used to obtain a tag text collection, the tag text collection including all target historical data texts corresponding to the user interest point tag; a text frequency submodule, used to determine the frequency data corresponding to each target historical data text in the tag text collection; and a user frequency submodule, used to determine the user frequency data based on the frequency data corresponding to each target historical data text in the tag text collection.

[0016] In some embodiments of the present application, the multi-dimensional label association data includes target weights, and the data analysis module specifically includes: an inverse text frequency submodule, used to determine the inverse text frequency corresponding to each target historical data text in the label text set; a text weight submodule, used to obtain the text weight corresponding to each target historical data text according to the inverse text frequency and frequency data corresponding to each target historical data text in the label text set; a target weight submodule, used to obtain the target weight according to the text weight corresponding to each target historical data text in the label text set.

[0017] In some embodiments of the present application, the multi-dimensional tag association data includes industry time factors, and the data analysis module specifically includes: a tag time module, used to determine the tag time, the tag time is the generation time of each target historical data text corresponding to the user interest point tag; a time factor module, used to determine the industry time factor based on the tag time.

[0018] In some embodiments of the present application, the result determination module specifically includes: an interest point assignment submodule, which is used to obtain the interest point score corresponding to the user interest point label according to the label association data; and an interest point sorting submodule, which is used to sort the user interest point label according to the interest point score to obtain the interest point result.

[0019] According to one aspect of an embodiment of the present application, a computer-readable medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method for estimating user interest points as described in the above embodiment is implemented.

[0020] According to one aspect of an embodiment of the present application, an electronic device is provided, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the user interest point estimation method as described in the above embodiments.

[0021] In the technical solutions provided in some embodiments of the present application, various historical data texts associated with the user, i.e., the target historical data texts, are collected and then analyzed and extracted to obtain user interest point tags and multi-dimensional tag association data corresponding to the user interest point tags. Based on the multi-dimensional tag association data, the user's interest level in each interest point tag can be more accurately determined, thereby obtaining the user interest point results. At the same time, in this embodiment, only the historical data texts of the relevant users are needed to estimate the user interest points, without the need for a large amount of sample data, thereby reducing the cost of interest point prediction.

[0022] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0024] Figure 1 A schematic diagram of an exemplary system architecture to which the technical solution of the embodiments of the present application can be applied is shown.

[0025] Figure 2 A flow chart of a method for estimating user interest points provided in an embodiment of the present application is shown.

[0026] Figure 3 Shown according to Figure 2 A specific implementation flow chart of step S200 in the method for estimating user interest points shown in the corresponding embodiment.

[0027] Figure 4 Shown according to Figure 2A specific implementation flow chart of step S300 in the method for estimating user interest points shown in the corresponding embodiment.

[0028] Figure 5 Shown according to Figure 4 Another specific implementation flowchart of step S300 in the method for estimating user interest points shown in the corresponding embodiment.

[0029] Figure 6 Shown according to Figure 2 A specific implementation flowchart corresponding to the method for estimating user interest points shown in the corresponding embodiment.

[0030] Figure 7 A schematic diagram of the structure of a user interest point estimation device provided in an embodiment of the present application is shown.

[0031] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0032] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this application will be more comprehensive and complete and fully convey the concept of the example embodiments to those skilled in the art.

[0033] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, known methods, devices, realizations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0034] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0035] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0036] Figure 1 A schematic diagram of an exemplary system architecture to which the technical solution of the embodiments of the present application can be applied is shown.

[0037] like Figure 1 As shown, the system architecture may include terminal devices (such as Figure 1 The embodiment of the present invention is a schematic diagram of a mobile phone 101, a tablet computer 102, a portable computer 103, a desktop computer, etc., a network 104, and a server 105. The network 104 is used to provide a medium for a communication link between the terminal device and the server 105. The network 104 may include various connection types, such as a wired communication link, a wireless communication link, etc.

[0038] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. According to the implementation requirements, there may be any number of terminal devices, networks and servers. For example, the server 105 may be a server cluster composed of multiple servers.

[0039] The user can use the terminal device to interact with the server 105 through the network 104 to receive or send messages, etc. The server 105 can be a server that provides various services. For example, the user uses the terminal device 103 (or the terminal device 101 or 102) to upload the target historical data text to the server 105, and the server 105 can extract the industry label of the target historical data text to obtain the user interest point label, each user interest point label corresponds to at least one target historical data text; according to the user interest point label, determine the corresponding multi-dimensional label association data; according to the multi-dimensional label association data, obtain the user interest point result.

[0040] It should be noted that the user interest point estimation method provided in the embodiment of the present application is generally executed by the server 105, and accordingly, the user interest point estimation device is generally set in the server 105. However, in other embodiments of the present application, the terminal device may also have similar functions as the server, so as to execute the user interest point estimation solution provided in the embodiment of the present application.

[0041] At the same time, it should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and signals involved in the embodiments of this specification are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions. For example, the object characteristics, interactive behavior characteristics and user information involved in this specification are all obtained with full authorization.

[0042] The implementation details of the technical solution of the embodiment of the present application are described in detail below:

[0043] Figure 2 A flowchart of a method for estimating user interest points according to an embodiment of the present application is shown. The method for estimating user interest points can be executed by a server, which can be Figure 1 Refer to the server shown in Figure 2 As shown, the method for estimating user interest points at least includes:

[0044] Step S100, obtaining target historical data text, wherein the target historical data text is historical data text associated with the user.

[0045] Step S200, extracting industry labels from the target historical data text to obtain user interest point labels, each user interest point label corresponds to at least one target historical data text.

[0046] Step S300: determining corresponding multi-dimensional tag association data according to the user's point of interest tag.

[0047] Step S400, obtaining user interest point results according to the multi-dimensional tag association data.

[0048] In an embodiment of the present application, various historical data texts associated with the user, i.e., the target historical data texts, are collected and then analyzed and extracted to obtain user interest point tags and multi-dimensional tag association data corresponding to the user interest point tags. Based on the multi-dimensional tag association data, the user's interest level in each interest point tag can be more accurately determined, thereby obtaining the user interest point results. At the same time, in this embodiment, only the historical data texts of the relevant users are needed to estimate the user interest points, without the need for a large amount of sample data, thereby reducing the cost of interest prediction.

[0049] In step S100, if the user's points of interest need to be estimated, a large amount of historical data text related to the user, ie, target historical data text, needs to be obtained first.

[0050] The target history data text may include user browsing data text and user search data text. The user browsing data text refers to the web page data browsed by the user. In an embodiment of the present application, it may be the title of the web page browsed by the user; the user search data text includes the search terms, search statements, etc. used by the user.

[0051] In step S200, industry tags are extracted from the target historical data text, and each extracted industry tag is used as a user interest point to form a user interest point tag. Based on the user interest point tag, the historical data text related to each user interest point tag is analyzed to obtain the analysis result.

[0052] Specifically, in some embodiments, the specific implementation of step S200 can be found in Figure 3 . Figure 3 is based on Figure 2 The detailed description of step S200 in the method for estimating user interest points shown in the corresponding embodiment, in the method for estimating user interest points, step S200 may include the following steps:

[0053] Step S210, extracting industry labels from each of the target historical data texts to obtain corresponding text industry labels.

[0054] Step S220: clustering the text industry tags to obtain user interest point tags.

[0055] In an embodiment of the present application, industry labels are first extracted for each target historical data text, and the corresponding text industry labels are extracted for each target historical data text. Then, the text industry labels are classified and clustered to obtain user interest point labels. In this way, one user interest point label can correspond to multiple target historical data texts.

[0056] Specifically, in some embodiments, the specific implementation of step S210 can refer to the following embodiments. Figure 2 The detailed description of step S210 in the method for estimating user interest points shown in the corresponding embodiment, in the method for estimating user interest points, step S210 may include the following steps:

[0057] Preprocess each of the target historical data texts to obtain a preprocessed text.

[0058] Industry labels are extracted from each of the preprocessed texts to obtain corresponding text industry labels.

[0059] In the above embodiment, the preprocessing of the target historical data text includes data cleaning and data aggregation, in which target historical data text with omissions is eliminated and similar target historical data texts are aggregated, wherein similar target historical data texts refer to target historical data texts with similar text content and similar generation time, so as to avoid repeated extraction of the same historical data text, which affects the determination of label-related data in step S300, and further affects the judgment of user interest points.

[0060] Whether two historical data texts are similar can be determined by comparing the similarity of the text content and the time difference between the generation times. When the similarity is higher than the similarity threshold and the time difference between the generation times is within the predetermined time difference threshold, the two historical data texts can be directly determined to be similar.

[0061] After the preprocessed text is obtained, the industry label corresponding to the preprocessed text, that is, the text industry label, is determined by comparing the preprocessed text with a pre-configured industry label library.

[0062] For example, for a user browsing data text, whose title is "AI mobile phone upgrade changes at a glance", the corresponding text industry tag can be extracted from the industry tag library as digital products, or further subdivided into smartphones.

[0063] For another example, for a user search data text, which is "transfer of a certain player", the corresponding text industry tag can be extracted as football according to the industry tag library, or further subdivided into a certain player and the corresponding club.

[0064] In some embodiments, a target historical data text may correspond to multiple industry tags. Taking "transfer of a certain player" as an example, it may correspond to multiple text industry tags at the same time, namely football, a certain player, a certain player's current club, and a certain player's target club for transfer.

[0065] In step S220, after obtaining the text industry tags, the same text industry tags from different target historical data texts are clustered and integrated to obtain user interest point tags, each of which corresponds to at least one target historical data text.

[0066] In step S300, based on the user's point of interest tags, relevant data is mined from multiple dimensions to obtain corresponding tag-related data.

[0067] In an embodiment of the present application, the tag-associated data has multiple dimensions, including user diversity parameters, user frequency data, target weights, industry time factors, etc.

[0068] Specifically, in some embodiments, the specific implementation of step S300 can refer to the following embodiments. Figure 2 The detailed description of step S300 in the method for estimating user interest points shown in the corresponding embodiment, in the method for estimating user interest points, the multi-dimensional tag association data includes a user diversity parameter, and step S300 may include the following steps:

[0069] Determine the number of target historical data texts with different contents corresponding to the user interest point tags.

[0070] A user diversity parameter is determined according to the number of target historical data texts of different contents corresponding to the user interest point tags.

[0071] In an embodiment of the present application, after extracting the user interest point tags corresponding to all target historical data texts in step S200, the total number of extracted user interest point tags is counted, and the user diversity parameter is determined based on the total number of user interest point tags. The more target historical data texts with different contents under a user interest point tag, the higher the diversity of user interests under the tag, and the larger the user diversity parameter. The higher the diversity of user interests under the tag, the more content that the user is interested in, and the deeper the interest in the user interest tag may be.

[0072] In some embodiments, the user diversity parameter can be obtained by querying the text number-diversity parameter association table to obtain the user diversity parameters corresponding to different text numbers. The above text number-diversity parameter association table is pre-configured, and the number of target historical data texts with the same content is positively correlated with the user diversity parameter.

[0073] In some other embodiments, the user diversity parameter may be calculated by the following formula:

[0074] D = log2(1+n)

[0075] Wherein, D is a user diversity parameter, and n is the number of target historical data texts of different contents corresponding to the user's interest point tags.

[0076] Specifically, in some embodiments, the specific implementation of step S300 can refer to the following embodiments. Figure 2 The detailed description of step S300 in the method for estimating user interest points shown in the corresponding embodiment, in the method for estimating user interest points, the multi-dimensional tag association data includes industry time factors, and step S300 may include the following steps:

[0077] Determine the label time, where the label time is the generation time of each of the target historical data texts corresponding to the user interest point label.

[0078] The industry time factor is determined according to the label time.

[0079] In the embodiment of the present application, according to the user's interest point tag, the generation time of each corresponding target historical data text is confirmed one by one, the tag time corresponding to the user's interest point tag is obtained, and the time factor is determined according to the tag time. The time factor can also be called the time decay factor, which reflects the process that the popularity of the tag gradually cools down over time, making the estimation of the user's interest point timely and improving the accuracy of the estimation of the user's interest point at the current time.

[0080] Specifically, after obtaining the generation time corresponding to each target historical data text corresponding to the user's interest point tag, the time factor corresponding to each target historical data text can be determined, and then the time factors corresponding to each target historical data text corresponding to the user's interest point tag are added together to obtain the corresponding industry time factor.

[0081] In some embodiments, the time factor corresponding to each target historical data text can be obtained by querying the time and decay factor association table according to the generation time. The time and decay factor association table is pre-configured, and the longer the generation time of the historical data text, the smaller the time factor.

[0082] In other embodiments, the time factor corresponding to each target historical data text may be obtained according to the following formula:

[0083]

[0084] Among them, T(t) is the time factor corresponding to the target historical data text; t is the generation time corresponding to the target historical data text; t0 is the current time, which is generally taken as 0; T(t0) is the time factor corresponding to the current time, which is generally taken as 1; k is the constant term coefficient.

[0085] The method for determining the constant term coefficient k may refer to the following embodiment.

[0086] In one embodiment, T(t) is set to 1 or any constant, and then it is only necessary to determine the time factor corresponding to any generation time to determine the constant term coefficient k. For example, if T(t) is 1, and the time factor corresponding to the generation time 10 days before is 0.2, then the constant term coefficient k can be determined to be approximately 0.16.

[0087] In another embodiment, T(t) may also be uncertain, and it is only necessary to determine the time factor corresponding to any two different generation times, that is, to determine the constant term coefficient k.

[0088] In some other embodiments, time factors corresponding to multiple different generation times may be taken, and the constant term coefficient k may be determined by fitting to obtain a more accurate formula.

[0089] Specifically, in some embodiments, the specific implementation of step S300 can be found in Figure 4 . Figure 4 is based on Figure 2 The detailed description of step S300 in the method for estimating user interest points shown in the corresponding embodiment, in the method for estimating user interest points, the multi-dimensional tag association data includes user frequency data, and step S300 may include the following steps:

[0090] Step S302: Obtain a tag text set, where the tag text set includes all target historical data texts corresponding to the user's interest point tags.

[0091] Step S304, determining the frequency data corresponding to each target historical data text in the label text set.

[0092] Step S306: Determine the user frequency data according to the frequency data corresponding to each target historical data text in the label text set.

[0093] In this embodiment, taking the user interest point label as the anchor point, first obtain all the target historical data texts corresponding to each user interest point label, and form their own sets, namely label text sets, one user interest point label corresponds to one label text set.

[0094] In the tag text set, the frequency data of each target historical data text is determined respectively. According to the frequency data of each target historical data text, the frequency data corresponding to the user's point of interest tag is finally obtained, that is, the user frequency data. The user frequency data is the frequency of the user searching or browsing the content related to the point of interest tag. The higher the user frequency data, the more interested the user is in the user's point of interest tag. The user frequency data can more accurately confirm the user's interest in each point of interest tag.

[0095] The frequency data of a target historical data text is determined by the proportion of a target historical text data and target historical text data with the same content in all target historical text data.

[0096] That is, user frequency data can be obtained according to the following formula:

[0097]

[0098]

[0099] Among them, P v For user frequency data, is the frequency data corresponding to the target historical data text of the label text set, N i is the target historical text data and the number of target historical text data with the same content, N t is the total number of all target historical text data.

[0100] Specifically, in some embodiments, the specific implementation of step S300 can be found in Figure 5 . Figure 5 is based on Figure 4The detailed description of step S300 in the method for estimating user interest points shown in the corresponding embodiment, in the method for estimating user interest points, step S300 may further include the following steps:

[0101] Step S307, determining the inverse text frequency corresponding to each target historical data text in the label text set.

[0102] Step S305, obtaining the text weight corresponding to each target historical data text according to the inverse text frequency and frequency data corresponding to each target historical data text in the label text set.

[0103] Step S303, obtaining the target weight according to the text weight corresponding to each target historical data text in the label text set.

[0104] In the label text set, in addition to determining the frequency data of each target historical data text, it is also necessary to determine the inverse text frequency of each target historical data text, and then obtain the text weight of each historical data text based on the frequency data and the inverse text frequency. Finally, according to the text weight corresponding to each target historical data text in the label text set, the content weight corresponding to the user's interest point label is obtained, that is, the target weight.

[0105] The inverse text frequency can be obtained from the total amount of historical data texts of all users contained in the text library and the number of the target historical text data and the historical text data with the same content contained in the text library. The above text library contains all historical data texts corresponding to all users.

[0106] That is, the inverse text frequency of the historical text can be obtained according to the following formula:

[0107]

[0108] Among them, F i is the inverse text frequency corresponding to the target historical data text, W is the total amount of historical data texts of all users contained in the text library, and W i The target historical text data and the number of historical text data with the same content contained in the text library.

[0109] The text weight and target weight can be obtained according to the following formula:

[0110]

[0111]

[0112] Among them, T is the target weight, t(i) is the text weight, is the frequency data corresponding to the target historical data text of the label text set, Ni is the target historical text data and the number of target historical text data with the same content, N t is the total number of all target historical text data, F i is the inverse text frequency corresponding to the target historical data text, W is the total amount of historical data texts of all users contained in the text library, and W i The target historical text data and the number of historical text data with the same content contained in the text library.

[0113] In step S400, based on the multi-dimensional tag association data, one point of interest tag is regarded as one user point of interest, and prediction fusion of the user point of interest is performed to obtain a user point of interest result.

[0114] Specifically, in some embodiments, the specific implementation of step S400 can refer to the following embodiments. Figure 2 The detailed description of step S400 in the method for estimating user interest points shown in the corresponding embodiment, in the method for estimating user interest points, step S400 may include the following steps:

[0115] According to the tag association data, the point of interest score corresponding to the user point of interest tag is obtained.

[0116] The user's point of interest tags are sorted according to the point of interest scores to obtain point of interest results.

[0117] In the embodiment of the present application, based on the above multi-dimensional tag association data, a point of interest tag is used as a user point of interest, and the score of the point of interest is evaluated. The higher the score of the point of interest, the more interested the user is. After obtaining the scores of all user point of interest tags, the point of interest tags are sorted according to the scores to obtain the point of interest results, which show the user's interest in each point of interest, which helps to achieve personalized services for customers.

[0118] Specifically, the interest point score can be obtained by the following formula:

[0119] score interest =P v ·T(t)·D·T

[0120] Among them, score interest is the interest point score, D is the user diversity parameter, T(t) is the time factor corresponding to the target historical data text, P v is the user frequency data, and T is the target weight.

[0121] In some embodiments, before sorting the user's interest point tags, the interest point scores need to be normalized to obtain a normalized result so that all interest point scores can be distributed within a threshold range of 0 to 100.

[0122] The following formula can be used to normalize the interest point scores:

[0123] Y=log(X+e)·Weight

[0124]

[0125]

[0126] Among them, Y is the normalized interest point score, X is the interest point score before normalization, and e is the natural base.

[0127] After obtaining the interest point score, the user's various interest points (i.e., user interest point tags) are sorted according to the interest point score. The higher the interest point score, the more interested the user is in the interest point, and the user's interest point tag is ranked relatively higher. After the sorting is completed, the interest point result is obtained.

[0128] In some embodiments, in addition to the ranking of the user's interest point tags, the interest point results will also be associated with the output interest point scores to better provide personalized services based on the user's interest points.

[0129] like Figure 6 As shown, in the above embodiment, labels are extracted based on the historical data text associated with the user (i.e., the target historical data text) to obtain user interest point labels, and then analysis is performed based on the user interest point labels to obtain multi-dimensional label association data - user diversity parameters, user frequency data, target weight, industry time factors, etc. Finally, based on the multi-dimensional label association data, the score of each interest point is calculated to obtain the user interest point result.

[0130] After obtaining the points of interest results, the user's interest content can be analyzed based on the points of interest results, which greatly improves the accuracy of personalized recommendations for content on the site. Especially for non-behaving users on the site, the content classification that the user is interested in is captured through points of interest, which can more accurately predict the user's content list of interest, thereby improving the precise distribution and conversion of content.

[0131] At the same time, after obtaining the points of interest results, we can explore the user's potential needs based on the points of interest results and find the user's potential points of interest, which provides direction for the functional optimization of current products and the iteration of subsequent products, and also provides data support for the introduction of subsequent content sources.

[0132] The above-mentioned interest point results have also brought about a qualitative improvement in refined operations. Users can be stratified by their interest points, and different targeted marketing activities and operation strategies can be carried out for users with different interests. Different types of marketing activities can be actively pushed to different interest groups, which not only increases the participation rate of push activities, but also avoids disturbing irrelevant users and reduces negative feedback from users. The accurate prediction of interest points has greatly improved the operational efficiency of operators, and can quickly capture the interests of various users, and various conversion indicators on the site have been greatly improved.

[0133] The following describes an apparatus embodiment of the present application, which can be used to execute the user interest point estimation method in the above embodiment of the present application. For details not disclosed in the apparatus embodiment of the present application, please refer to the embodiment of the user interest point estimation method in the above embodiment of the present application.

[0134] Figure 7 A block diagram of a device for estimating user interest points according to an embodiment of the present application is shown.

[0135] Reference Figure 7 As shown, according to an embodiment of the present application, a user interest point estimation device 700 includes: a text acquisition module 710, a label extraction module 720, a data analysis module 730, and a result determination module 740.

[0136] Among them, the text acquisition module 710 is used to obtain the target historical data text, and the target historical data text is the historical data text associated with the user; the label extraction module 720 is used to extract the industry label of the target historical data text to obtain the user interest point label, and each user interest point label corresponds to at least one target historical data text; the data analysis module 730 is used to determine the corresponding multi-dimensional label association data according to the user interest point label; the result determination module 740 is used to obtain the user interest point result according to the multi-dimensional label association data.

[0137] In some embodiments of the present application, the label extraction module specifically includes: a label extraction submodule, which is used to extract industry labels from each of the target historical data texts to obtain corresponding text industry labels; and a label clustering submodule, which is used to cluster the text industry labels to obtain user interest point labels.

[0138] In some embodiments of the present application, the multi-dimensional tag association data includes user diversity parameters, and the data analysis module specifically includes: a quantity determination submodule, used to determine the number of target historical data texts with different contents corresponding to the user interest point tags; a parameter determination submodule, used to determine the user diversity parameters according to the number of target historical data texts with different contents corresponding to the user interest point tags.

[0139] In some embodiments of the present application, the multi-dimensional tag association data includes user frequency data, and the data analysis module specifically includes: a text collection submodule, used to obtain a tag text collection, the tag text collection including all target historical data texts corresponding to the user interest point tag; a text frequency submodule, used to determine the frequency data corresponding to each target historical data text in the tag text collection; and a user frequency submodule, used to determine the user frequency data based on the frequency data corresponding to each target historical data text in the tag text collection.

[0140] In some embodiments of the present application, the multi-dimensional label association data includes target weights, and the data analysis module specifically includes: an inverse text frequency submodule, used to determine the inverse text frequency corresponding to each target historical data text in the label text set; a text weight submodule, used to obtain the text weight corresponding to each target historical data text according to the inverse text frequency and frequency data corresponding to each target historical data text in the label text set; a target weight submodule, used to obtain the target weight according to the text weight corresponding to each target historical data text in the label text set.

[0141] In some embodiments of the present application, the multi-dimensional tag association data includes industry time factors, and the data analysis module specifically includes: a tag time module, used to determine the tag time, the tag time is the generation time of each target historical data text corresponding to the user interest point tag; a time factor module, used to determine the industry time factor based on the tag time.

[0142] In some embodiments of the present application, the result determination module specifically includes: an interest point assignment submodule, which is used to obtain the interest point score corresponding to the user interest point label according to the label association data; and an interest point sorting submodule, which is used to sort the user interest point label according to the interest point score to obtain the interest point result.

[0143] In an embodiment of the present application, various historical data texts associated with the user, i.e., the target historical data texts, are collected and then analyzed and extracted to obtain user interest point tags and multi-dimensional tag association data corresponding to the user interest point tags. Based on the multi-dimensional tag association data, the user's interest level in each interest point tag can be more accurately determined, thereby obtaining the user interest point results. At the same time, in this embodiment, only the historical data texts of the relevant users are needed to estimate the user interest points, without the need for a large amount of sample data, thereby reducing the cost of interest prediction.

[0144] Figure 8A schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present application is shown.

[0145] It should be noted that Figure 8 The computer system of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0146] like Figure 8 As shown, the computer system includes a central processing unit (CPU) 1801, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 1802 or the program loaded from the storage part 1808 to the random access memory (RAM) 1803, such as executing the method described in the above embodiment. In RAM 1803, various programs and data required for system operation are also stored. CPU 1801, ROM 1802 and RAM 1803 are connected to each other through bus 1804. Input / output (I / O) interface 1805 is also connected to bus 1804.

[0147] The following components are connected to the I / O interface 1805: an input section 1806 including a keyboard, a mouse, etc.; an output section 1807 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1808 including a hard disk, etc.; and a communication section 1809 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 1809 performs communication processing via a network such as the Internet. A drive 1810 is also connected to the I / O interface 1805 as needed. A removable medium 1811, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1810 as needed so that a computer program read therefrom is installed into the storage section 1808 as needed.

[0148] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication section 1809, and / or installed from a removable medium 1811. When the computer program is executed by a central processing unit (CPU) 1801, various functions defined in the system of the present application are executed.

[0149] It should be noted that the computer-readable medium shown in the embodiment of the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, - but not limited to - an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by an instruction execution system, device or device or used in combination with it. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate, or transmit programs for use by or in conjunction with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0150] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0151] The units involved in the embodiments described in this application may be implemented by software or hardware, and the units described may also be set in a processor. The names of these units do not, in some cases, constitute limitations on the units themselves.

[0152] As another aspect, the present application also provides a computer-readable medium, which may be included in the electronic device described in the above embodiment; or may exist independently without being assembled into the electronic device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements the method described in the above embodiment.

[0153] It should be noted that, although several modules or units of the equipment for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.

[0154] Through the description of the above implementation methods, it is easy for those skilled in the art to understand that the example implementation methods described here can be implemented by software or by combining software with necessary hardware. Therefore, the technical solution according to the implementation methods of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the implementation methods of the present application.

[0155] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the embodiments disclosed herein. The present application is intended to cover any variations, uses or adaptations of the present application, which follow the general principles of the present application and include common knowledge or customary technical means in the art that are not disclosed in the present application.

[0156] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for estimating user interest points, characterized in that: The method for estimating user interest points comprises: Acquire target historical data text, where the target historical data text is historical data text associated with the user; Extracting industry labels from the target historical data text to obtain user interest point labels, each user interest point label corresponds to at least one target historical data text; Determine corresponding multi-dimensional tag association data according to the user's point of interest tag; According to the multi-dimensional tag association data, the user interest point results are obtained.

2. The method for estimating user interest points according to claim 1, characterized in that: The industry label extraction of the target historical data text to obtain the user interest point label specifically includes: Extracting industry labels from each of the target historical data texts to obtain corresponding text industry labels; The text industry tags are clustered to obtain user interest point tags.

3. The method for estimating user interest points according to claim 1, characterized in that: The multi-dimensional tag association data includes a user diversity parameter, and the determining of the corresponding multi-dimensional tag association data according to the user interest point tag specifically includes: Determine the number of target historical data texts with different contents corresponding to the user's interest point tags; A user diversity parameter is determined according to the number of target historical data texts of different contents corresponding to the user interest point tags.

4. The method for estimating user interest points according to claim 1, wherein: The multi-dimensional tag association data includes user frequency data, and the determining of the corresponding multi-dimensional tag association data according to the user interest point tag specifically includes: Acquire a label text set, wherein the label text set includes all target historical data texts corresponding to the user's point of interest label; Determine the frequency data corresponding to each target historical data text in the label text set; The user frequency data is determined according to the frequency data corresponding to each target historical data text in the label text set.

5. The method for estimating user interest points according to claim 4, characterized in that: The multi-dimensional tag association data includes a target weight, and the corresponding multi-dimensional tag association data is determined according to the user interest point tag, and specifically includes: Determine the inverse text frequency corresponding to each target historical data text in the label text set; According to the inverse text frequency and frequency data corresponding to each target historical data text in the label text set, the text weight corresponding to each target historical data text is obtained; The target weight is obtained according to the text weight corresponding to each target historical data text in the label text set.

6. The method for estimating user interest points according to claim 1, wherein: The multi-dimensional tag association data includes industry time factors, and the corresponding multi-dimensional tag association data is determined according to the user interest point tag, specifically including: Determine the label time, where the label time is the generation time of each of the target historical data texts corresponding to the user interest point label; The industry time factor is determined according to the label time.

7. The method for estimating user interest points according to claim 1, characterized in that: The step of obtaining the user interest point result according to the multi-dimensional tag association data specifically includes: Obtaining, according to the tag association data, a point of interest score corresponding to the user point of interest tag; The user's point of interest tags are sorted according to the point of interest scores to obtain point of interest results.

8. A user interest point estimation device, characterized in that: The user interest point estimation device comprises: A text acquisition module, used to acquire a target historical data text, wherein the target historical data text is a historical data text associated with the user; A label extraction module is used to extract industry labels from the target historical data text to obtain user interest point labels, each user interest point label corresponds to at least one target historical data text; A data analysis module, used to determine corresponding multi-dimensional tag-related data according to the user's point of interest tags; The result determination module is used to obtain the user interest point result according to the multi-dimensional tag association data.

9. A computer readable medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for estimating user interest points according to any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, enables the one or more processors to implement the user interest point estimation method as described in any one of claims 1 to 7.

Images