A method and device for constructing a category association graph

Abstract

The present invention discloses a method and device for constructing a category association map, and relates to the field of computer technology. A specific implementation of the method includes: determining a population classification set associated with the physical category set based on a physical category set, and determining a virtual category set associated with the population classification set based on the population classification set; constructing a category association map based on a pre-set virtual category weight value of the virtual category set, and the association relationship between the population classification set and the physical category set and the virtual category set; screening out virtual categories to be recommended that are associated with target users from the virtual category set; and modifying the virtual category weight value corresponding to the virtual category to be recommended based on the conversion result of the virtual category to be recommended, thereby updating the category association map. This implementation can solve the technical problem that association between categories with different attributes cannot be achieved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and device for constructing a category association graph. Background Art

[0002] The knowledge graph can complete tree associations and use visual graphs to vividly display the association structure, so that users can find more accurate information, make more comprehensive summaries, and provide more in-depth and relevant information. Based on the knowledge graph technology, graph algorithms for various scenarios can be derived, such as object knowledge graphs, data knowledge graphs, etc. Using knowledge graph link analysis and cluster analysis, the correlation between item categories can be expressed by the weight values ​​of category terms and the feature vectors they form.

[0003] In the process of implementing the present invention, the inventors found that there are at least the following problems in the prior art:

[0004] In the item knowledge graph, only items with the same category attributes can be associated, and it is impossible to associate categories with different attributes. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides a method and device for constructing a category association map to solve the technical problem of being unable to achieve association between categories with different attributes.

[0006] To achieve the above object, according to one aspect of an embodiment of the present invention, a method for constructing a category association graph is provided, comprising:

[0007] Determine a population classification set associated with the physical category set according to the physical category set, and determine a virtual category set associated with the population classification set according to the population classification set;

[0008] Constructing a category association map according to the preset virtual category weight values ​​of the virtual category set and the association relationship between the population classification set and the physical category set and the virtual category set;

[0009] Filtering virtual categories to be recommended that are associated with a target user from the virtual category set;

[0010] According to the conversion result of the virtual category to be recommended, the virtual category weight value corresponding to the virtual category to be recommended is modified, so as to update the category association map.

[0011] Optionally, determining a population category set associated with the physical category set according to the physical category set, and determining a virtual category set associated with the population category set according to the population category set includes:

[0012] According to the identifiers of each physical item category in the physical item category set, a crowd label associated with the identifiers of each physical item category is determined, thereby obtaining a crowd classification set associated with the physical item category set; wherein the crowd classification set includes at least one of the crowd labels;

[0013] Based on the crowd portrait, filter out the user identifiers associated with the crowd tag;

[0014] According to the user identifiers, the identifiers of the virtual categories associated with the respective user identifiers are determined respectively, so as to obtain a virtual category set associated with the population classification set; wherein the virtual category set includes multiple virtual category identifiers.

[0015] Optionally, constructing a category association map according to a preset virtual category weight value of the virtual category set and an association relationship between the population classification set and the physical category set and the virtual category set includes:

[0016] According to the physical category set, the crowd classification set and the virtual category set, configuring in the graph construction tool an association relationship between the crowd classification set and each physical category in the physical category set and each virtual category in the virtual category set;

[0017] According to the preset virtual category weight values ​​corresponding to each virtual category in the virtual category set, and the association relationship between the population classification set and each physical category in the physical category set and each virtual category in the virtual category set, a category association map is constructed by the map construction tool.

[0018] Optionally, selecting a virtual category to be recommended that is associated with a target user from the virtual category set includes:

[0019] Determine whether the target user's crowd label belongs to any crowd label in the crowd classification set;

[0020] If yes, continue to determine whether the physical item category corresponding to the target user belongs to the physical item category set associated with the target user's population tag;

[0021] If so, the virtual categories to be recommended that are associated with the target user are screened out from the virtual category set according to the population tag of the target user.

[0022] Optionally, according to the population tag of the target user, selecting the virtual category to be recommended associated with the target user from the virtual category set includes:

[0023] According to the target user's population tag, a target user virtual category set associated with the target user is filtered out from the virtual category set;

[0024] According to the virtual category weight values ​​corresponding to the virtual categories in the target user virtual category set, the top N virtual categories are selected as the virtual categories to be recommended;

[0025] Wherein, N is an integer greater than zero.

[0026] Optionally, modifying the virtual category weight value corresponding to the virtual category to be recommended according to the conversion result of the virtual category to be recommended includes:

[0027] If the conversion result of the virtual category to be recommended is conversion, the virtual category weight value corresponding to the virtual category to be recommended is increased based on a preset weight enhancement coefficient.

[0028] Optionally, increasing the virtual category weight value corresponding to the virtual category to be recommended based on a preset weight enhancement coefficient includes:

[0029] The sum of the virtual category weight value corresponding to the virtual category to be recommended and the preset weight enhancement coefficient is used as the modified virtual category weight value corresponding to the virtual category to be recommended.

[0030] In addition, according to another aspect of an embodiment of the present invention, there is provided a device for constructing a category association map, comprising:

[0031] An association module, configured to determine a population classification set associated with the physical object classification set based on the physical object classification set, and determine a virtual category set associated with the population classification set based on the population classification set;

[0032] A construction module, configured to construct a category association map according to a preset virtual category weight value of the virtual category set and an association relationship between the population classification set and the physical category set and the virtual category set;

[0033] A screening module, used to screen out virtual categories to be recommended that are associated with a target user from the virtual category set;

[0034] The updating module is used to modify the virtual category weight value corresponding to the virtual category to be recommended according to the conversion result of the virtual category to be recommended, so as to update the category association map.

[0035] Optionally, the association module is further used for:

[0036] According to the identifiers of each physical item category in the physical item category set, a crowd label associated with the identifiers of each physical item category is determined, thereby obtaining a crowd classification set associated with the physical item category set; wherein the crowd classification set includes at least one of the crowd labels;

[0037] Based on the crowd portrait, filter out the user identifiers associated with the crowd tag;

[0038] According to the user identifiers, the identifiers of the virtual categories associated with the respective user identifiers are determined respectively, so as to obtain a virtual category set associated with the population classification set; wherein the virtual category set includes multiple virtual category identifiers.

[0039] Optionally, the building block is further used to:

[0040] According to the physical category set, the crowd classification set and the virtual category set, configuring in the graph construction tool an association relationship between the crowd classification set and each physical category in the physical category set and each virtual category in the virtual category set;

[0041] According to the preset virtual category weight values ​​corresponding to each virtual category in the virtual category set, and the association relationship between the population classification set and each physical category in the physical category set and each virtual category in the virtual category set, a category association map is constructed by the map construction tool.

[0042] Optionally, the screening module is further used for:

[0043] Determine whether the target user's crowd label belongs to any crowd label in the crowd classification set;

[0044] If yes, continue to determine whether the physical item category corresponding to the target user belongs to the physical item category set associated with the target user's population tag;

[0045] If so, the virtual categories to be recommended that are associated with the target user are screened out from the virtual category set according to the population tag of the target user.

[0046] Optionally, the screening module is further used for:

[0047] According to the target user's population tag, a target user virtual category set associated with the target user is filtered out from the virtual category set;

[0048] According to the virtual category weight values ​​corresponding to the virtual categories in the target user virtual category set, the top N virtual categories are selected as the virtual categories to be recommended;

[0049] Wherein, N is an integer greater than zero.

[0050] Optionally, the update module is further used for:

[0051] If the conversion result of the virtual category to be recommended is conversion, the virtual category weight value corresponding to the virtual category to be recommended is increased based on a preset weight enhancement coefficient.

[0052] Optionally, the update module is further used for:

[0053] The sum of the virtual category weight value corresponding to the virtual category to be recommended and the preset weight enhancement coefficient is used as the modified virtual category weight value corresponding to the virtual category to be recommended.

[0054] According to another aspect of an embodiment of the present invention, there is further provided an electronic device, including:

[0055] one or more processors;

[0056] a storage device for storing one or more programs,

[0057] When the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any of the above embodiments.

[0058] According to another aspect of an embodiment of the present invention, a computer-readable medium is provided, on which a computer program is stored. When the program is executed by a processor, the method described in any one of the above embodiments is implemented.

[0059] One embodiment of the above invention has the following advantages or beneficial effects: because it adopts a technical means of constructing a category association map according to the virtual category weight value and the association relationship between the crowd classification set and the physical category set and the virtual category set, and modifies the virtual category weight value corresponding to the virtual category to be recommended according to the conversion result of the virtual category to be recommended, thereby updating the category association map, it overcomes the technical problem that the association between categories with different attributes cannot be achieved in the prior art. The embodiment of the present invention constructs a category relationship map through the association relationship between crowd labels and physical categories and virtual categories to achieve visual associations between categories with different attributes, which helps to explore the deep relationship between categories with different attributes. Moreover, the embodiment of the present invention also performs weight intervention based on the conversion result of the virtual category to be recommended, and continuously optimizes the virtual category weight value corresponding to the virtual category to be recommended to improve the accuracy of the virtual category weight value.

[0060] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0062] Figure 1 is a schematic diagram of the main process of the method for constructing a category association map according to an embodiment of the present invention;

[0063] Figure 2 is a schematic diagram of a category association map according to an embodiment of the present invention;

[0064] Figure 3 It is a schematic diagram of the main process of a method for constructing a category association map according to a reference embodiment of the present invention;

[0065] Figure 4 is a schematic diagram of main modules of an apparatus for constructing a category association map according to an embodiment of the present invention;

[0066] Figure 5 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0067] Figure 6 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0068] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0069] Figure 1 Schematic diagram of the main process of the method for constructing a category association map according to an embodiment of the present invention. As an embodiment of the present invention, Figure 1 As shown, the method for constructing a category association map may include:

[0070] Step 101: determine a population category set associated with the physical category set based on the physical category set, and determine a virtual category set associated with the population category set based on the population category set.

[0071] In step 101, sample data retrieval and data preprocessing are completed to screen out mutually related physical item category sets, population classification sets, and virtual item category sets. In the embodiment of the present invention, the physical item category may be an item category that has a physical item delivery process, and the virtual item category may be an item category that has no physical delivery process, such as a user going to a store, a service person coming to the home, direct online recharge, or obtaining digital services.

[0072] Optionally, step 101 may include: according to the identifiers of each physical category in the physical category set, determining the crowd tags associated with the identifiers of each physical category, thereby obtaining a crowd classification set associated with the physical category set; wherein the crowd classification set includes at least one of the crowd tags; based on the crowd portrait, filtering out the user identifiers associated with the crowd tags; according to the user identifiers, respectively determining the identifiers of the virtual categories associated with each of the user identifiers, thereby obtaining a virtual category set associated with the crowd classification set; wherein the virtual category set includes multiple identifiers of the virtual categories. For example, if "baby strollers", "diapers", and "infant milk powder" are pre-selected as physical categories, then crowd tags that purchase these physical categories can be filtered out from the database, such as "parenting families", "mothers", etc.; then, according to these crowd tags, users marked with these tags can be filtered out from the database; finally, according to the virtual categories purchased by these users, such as "baby swimming", "audio books", "month-old nanny", etc., a virtual category set is obtained. For another example, if "cleaning supplies", "grains", and "kitchen condiments" are selected as physical categories in advance, then the crowd labels that purchase these physical categories, such as "housewives", "grandparents", etc., can be screened from the database; then users marked with these labels can be screened from the database based on these crowd labels; finally, a virtual category set can be obtained based on the virtual categories purchased by these users, such as "cleaning", "hourly workers", etc. Therefore, through step 101, the association between the physical category set, the crowd classification set, and the virtual category set can be achieved.

[0073] For example, assuming that there is a physical category set D, a crowd classification set P is obtained by associating the crowd portrait with the physical category set D, and then the crowd classification set P is associated with the virtual category set S. It should be pointed out that the crowd classification set P can include multiple crowd labels, that is, users are divided into multiple groups of people, and one crowd label represents one group of people.

[0074] Assume that the set of physical items of the w-th group of people can be expressed as PD w =[d1,d2,…,d n ];

[0075] Assume that the virtual category set of the w-th group of people can be expressed as PS w =[s1,s2,…,s n ].

[0076] Where d1 is the first physical item category purchased by the wth group in the physical item category set D, and d1 represents the physical item category ID (for example, when ID = 3001, it represents kitchen cleaning products). nRepresents the virtual category ID. Therefore, all types of people constitute the crowd classification set P, the physical category set of all types of people constitutes the physical category set D, and the virtual category set of all types of people constitutes the virtual category set P.

[0077] Step 102, constructing a category association map according to the preset virtual category weight values ​​of the virtual category set, and the association relationship between the population classification set and the physical category set and the virtual category set.

[0078] In this step, a category association map can be constructed using a map construction tool (such as Ucinet NetDraw, VOSviewer, etc.) based on the preset virtual category weight values ​​and the association relationship between the physical category set, the population classification set and the virtual category set obtained in step 101 to achieve visualization.

[0079] Optionally, step 102 may include: configuring, in a graph construction tool, an association relationship between the population classification set and each physical category in the physical category set, and each virtual category in the virtual category set, according to the physical category set, the population classification set and the virtual category set; and constructing a category association graph through the graph construction tool according to a pre-set virtual category weight value corresponding to each virtual category in the virtual category set, and an association relationship between the population classification set and each physical category in the physical category set, and each virtual category in the virtual category set.

[0080] It should be noted that, in the embodiment of the present invention, the virtual category weight value corresponding to each virtual category in the virtual category set may be preset based on experience, and then the virtual category weight value may be updated based on step 104 .

[0081] For example, suppose: the virtual category weight value of the crowd

[0082] Then, in the population classification set P, the virtual category weight value of the w-th group of people is:

[0083]

[0084] In the embodiment of the present invention, for the wth group of people, based on the physical item category set PD associated with the wth group of people w , Crowd Classification Set PS w , we can get the category association map, that is, the category association map of the w-th group of people is G w .

[0085] For example, if Figure 2In the category relationship graph shown, the physical categories include complementary food, children's beds, infant milk powder, diapers, baby strollers, medicines, daily necessities, skin care products, clothing, shoes and hats, baby carriages, books, fresh milk products, teaching materials, and sports equipment; the virtual categories include healthy meals, home haircuts, nannies, confinement nannies, audio books, education and training, grade examinations, real estate, tutors, studying abroad, glasses, baby swimming, and cleaning. These physical categories and virtual categories are associated with crowd labels, and the crowd labels include parenting families (infants and young children), parenting families (preschool), and parenting families (primary and secondary schools); since the physical categories and virtual categories associated with these crowd labels have certain similarities, similar crowd labels can be put together in the crowd classification set P, so that the physical categories and virtual categories corresponding to these crowd labels are constructed in the same graph, so as to analyze the category relationship and mine deep information. Of course, it is also possible to construct separate category association maps for labels such as child-rearing families (infants and young children), child-rearing families (preschool), and child-rearing families (primary and secondary schools), and the embodiment of the present invention does not limit this. Figure 2 The numbers shown in represent the virtual category weight values ​​corresponding to each virtual category, so that users can intuitively understand the weight information of each virtual category.

[0086] Step 103: Filter out virtual categories to be recommended that are associated with the target user from the virtual category set.

[0087] In this step, according to the physical category to which the physical item purchased by the target user belongs, the related virtual categories are screened out from the virtual category set, and the virtual items under these virtual categories are recommended to the user.

[0088] Optionally, determine whether the target user's crowd label belongs to any one of the crowd labels in the crowd classification set; if so, continue to determine whether the physical item category corresponding to the target user belongs to the physical item category set associated with the target user's crowd label; if so, filter out the virtual category to be recommended associated with the target user from the virtual category set based on the target user's crowd label. For example, assuming that the physical item purchased by the target user pin belongs to the physical item category SKU.cid, then determine A: the pin's crowd portrait is w-type crowd in P, and determine B: the physical item category SKU.cid purchased by pin is in the physical item category set PD associated with w-type crowd. w Inside. Right now:

[0089] Judgment A: P w <>null;

[0090] Judgment B: PD w {SKU.cid}<>null;

[0091] When ((Pw <>null)∩(PD w {SKU.cid}<>null))=true, then in the virtual category set PS associated with the w group of people w Filter out the virtual categories to be recommended and make virtual item recommendations.

[0092] Optionally, according to the target user's crowd tag, selecting the to-be-recommended virtual category associated with the target user from the virtual category set includes: selecting, according to the target user's crowd tag, a target user virtual category set PS associated with the target user from the virtual category set P w ; According to the descending sorting of the virtual category weight values ​​corresponding to each virtual category in the target user virtual category set, the first N virtual categories are selected as the virtual categories to be recommended; wherein N is an integer greater than zero. For example, the target user virtual category set PS associated with w groups of people can be selected. w The top 10 virtual categories with the highest weights are selected, and then virtual items under the top 10 virtual categories are recommended to the user. Optionally, N virtual categories may also be randomly selected for recommendation, which is not limited in the embodiment of the present invention.

[0093] Optionally, when recommending virtual items to users, M virtual items with high popularity can be selected from N virtual categories to be recommended for recommendation, so as to improve the conversion rate of the virtual category, thereby accurately updating the virtual category weight value in step 104. Wherein, M is an integer greater than zero. It should be noted that virtual items can be recommended to target users or to other users, and the embodiment of the present invention does not limit this.

[0094] Step 104, modifying the virtual category weight value corresponding to the virtual category to be recommended according to the conversion result of the virtual category to be recommended, thereby updating the category association map.

[0095] Since virtual items are recommended to the user in step 103, the user may purchase or browse these recommended virtual items, or may not purchase or browse them. The embodiment of the present invention uses these results as the conversion results of the virtual category to be recommended, and based on this, modifies the virtual category weight value corresponding to the virtual category to be recommended, thereby updating the category association map.

[0096] Optionally, based on the conversion result of the virtual category to be recommended, the virtual category weight value corresponding to the virtual category to be recommended is modified, including: if the conversion result of the virtual category to be recommended is a conversion, then the virtual category weight value corresponding to the virtual category to be recommended is increased based on a preset weight enhancement coefficient. The conversion may be a purchase behavior, a browsing behavior, or a purchase rate, a browsing rate, etc., which can be set as needed. The importance of the virtual category to be recommended can be accurately reflected by the conversion result of the virtual category to be recommended, so modifying the virtual category weight value based on the conversion result can effectively improve the accuracy of the virtual category weight value, thereby obtaining an accurate category association map.

[0097] Optionally, increasing the virtual category weight value corresponding to the virtual category to be recommended based on a preset weight enhancement coefficient includes: taking the sum of the virtual category weight value corresponding to the virtual category to be recommended and the preset weight enhancement coefficient as the modified virtual category weight value corresponding to the virtual category to be recommended. w The weight of the corresponding category in (PS) is increased, and the weight is updated recursively in this mode.

[0098] For example, suppose s n For the recommended virtual category of this conversion, based on the preset weight enhancement coefficient (the weight enhancement coefficient can be set as needed), T w (PS) Weight increase:

[0099]

[0100] Finally, according to To update the category association map.

[0101] According to the various embodiments described above, it can be seen that the present invention constructs a category association map based on the virtual category weight values ​​and the association relationship between the crowd classification set and the physical category set and the virtual category set, and modifies the virtual category weight values ​​corresponding to the virtual category to be recommended according to the conversion results of the virtual category to be recommended, thereby updating the category association map. The technical problem that the association between categories with different attributes cannot be achieved in the prior art. The embodiment of the present invention constructs a category relationship map through the association relationship between crowd labels and physical categories and virtual categories to achieve visual associations between categories with different attributes, which helps to explore the deep relationship between categories with different attributes. Moreover, the embodiment of the present invention also performs weight intervention based on the conversion results of the virtual category to be recommended, and continuously optimizes the virtual category weight values ​​corresponding to the virtual category to be recommended to improve the accuracy of the virtual category weight values.

[0102] Figure 3It is a schematic diagram of the main process of a method for constructing a category association map according to a reference embodiment of the present invention.

[0103] Step 301, according to the identification of each physical object category in the physical object category set, determine the crowd label associated with the identification of each physical object category, so as to obtain the crowd classification set associated with the physical object category set.

[0104] Specifically, firstly, some physical categories are selected according to actual needs to obtain a physical category set D, and then the labels of the people who purchase these physical categories are screened out from the database to obtain a crowd classification set P. Generally speaking, the physical category set D includes multiple similar physical categories, such as "baby strollers", "diapers", "infant milk powder", etc., so as to dig deep into virtual categories with higher correlation, thereby improving the correlation between entities in the category association map. Moreover, the crowd classification set P includes multiple crowd labels (that is, multiple categories of people), and the crowd labels associated with these physical categories also have a certain degree of similarity, such as "parenting families", "moms", etc., to facilitate the analysis of the deep relationship between similar crowd labels. That is, the physical category set of the wth group of people can be expressed as PD w =[d1,d2,…,d n ].

[0105] Step 302, based on the crowd portrait, filter out the user identifiers associated with the crowd tag, and determine the identifiers of the virtual categories associated with each of the user identifiers according to the user identifiers, so as to obtain a virtual category set associated with the crowd classification set.

[0106] In this step, the virtual categories to which these virtual items belong are screened out according to the records of users marked with these population tags who purchased virtual items, thereby obtaining a virtual category set S. That is, the virtual category set of the wth population can be expressed as PS w =[s1,s2,…,s n ].

[0107] Step 303: construct a category association map according to the preset virtual category weight values ​​of the virtual category set and the association relationship between the population classification set and the physical category set and the virtual category set.

[0108] The virtual category weight values ​​corresponding to each virtual category in the virtual category set can be preset based on experience. Assume that:

[0109] Virtual category weight value of the crowd

[0110] Then, in the population classification set P, the virtual category weight value of the w-th group of people is:

[0111]

[0112] Then, the association relationship between the population classification set obtained in step 301 and step 302 and the physical category set and the virtual category set is constructed as follows: Figure 2 The category association map shown.

[0113] Step 304, determine whether the target user's crowd label belongs to any crowd label in the crowd classification set; if so, execute step 305; if not, end.

[0114] If a user pins and purchases a physical item, determine whether the user's crowd label is in the crowd classification set P.

[0115] Step 305, determine whether the physical item category corresponding to the target user belongs to the physical item category set associated with the target user's population tag; if so, execute step 306; if not, end.

[0116] If P w =P{pin}, indicating that the user's crowd label belongs to a crowd label in the crowd classification set P, that is, the user is the w-th crowd. Then according to the physical category set PD associated with the w-th crowd w , determine whether the physical item category SKU.cid of the physical item purchased by the user is in PD w middle.

[0117] Step 306 , based on the target user's population tag, filter out a target user virtual category set associated with the target user from the virtual category set.

[0118] If the physical item purchased by the user belongs to the physical item category SKU.cid in PD w According to the user's crowd label w, the target user virtual category set PS is selected from the virtual category set S. w .

[0119] Step 307 , based on the descending sorting of the virtual category weight values ​​corresponding to the virtual categories in the target user virtual category set, the first N virtual categories are selected as virtual categories to be recommended.

[0120] Specifically, for the target user virtual category set PS w The virtual category weight values ​​corresponding to each virtual category in the list are sorted from large to small, thereby selecting the top N virtual categories as virtual categories to be recommended. Virtual items (or popular virtual items) under the N virtual categories to be recommended can be pushed to the user or other users.

[0121] Step 308: modify the virtual category weight value corresponding to the virtual category to be recommended according to the conversion results of the N virtual categories to be recommended.

[0122] After recommending N virtual items under the virtual categories to be recommended to the user, the user's purchase record is used to query whether the user has purchased or browsed these virtual items, thereby obtaining the conversion result of the virtual category to be recommended. Then, according to the conversion result of the virtual category to be recommended, the virtual category weight value corresponding to the virtual category to be recommended is modified, for example, based on a preset weight enhancement coefficient (the weight enhancement coefficient can be set as needed), T w (PS) Weight increase:

[0123]

[0124] Step 309: updating the category association map according to the modified virtual category weight value corresponding to the virtual category to be recommended.

[0125] In addition, the specific implementation content of the method for constructing a category association map in a reference embodiment of the present invention has been described in detail in the method for constructing a category association map described above, so the repeated content will not be described here.

[0126] Figure 4 is a schematic diagram of the main modules of the device for constructing a category association map according to an embodiment of the present invention, such as Figure 4 As shown, the device 400 for constructing a category association map includes an association module 401, a construction module 402, a screening module 403 and an updating module 404. The association module 401 is used to determine a population classification set associated with the physical category set according to a physical category set, and to determine a virtual category set associated with the population classification set according to the population classification set; the construction module 402 is used to construct a category association map according to a preset virtual category weight value of the virtual category set, and the association relationship between the population classification set and the physical category set and the virtual category set; the screening module 403 is used to screen out virtual categories to be recommended that are associated with target users from the virtual category set; and the updating module 404 is used to modify the virtual category weight value corresponding to the virtual category to be recommended according to the conversion result of the virtual category to be recommended, so as to update the category association map.

[0127] Optionally, the association module 401 is further used for:

[0128] According to the identifiers of each physical item category in the physical item category set, a crowd label associated with the identifiers of each physical item category is determined, thereby obtaining a crowd classification set associated with the physical item category set; wherein the crowd classification set includes at least one of the crowd labels;

[0129] Based on the crowd portrait, filter out the user identifiers associated with the crowd tag;

[0130] According to the user identifiers, the identifiers of the virtual categories associated with the respective user identifiers are determined respectively, so as to obtain a virtual category set associated with the population classification set; wherein the virtual category set includes multiple virtual category identifiers.

[0131] Optionally, the construction module 402 is further configured to:

[0132] According to the physical category set, the crowd classification set and the virtual category set, configuring in the graph construction tool an association relationship between the crowd classification set and each physical category in the physical category set and each virtual category in the virtual category set;

[0133] According to the preset virtual category weight values ​​corresponding to each virtual category in the virtual category set, and the association relationship between the population classification set and each physical category in the physical category set and each virtual category in the virtual category set, a category association map is constructed by the map construction tool.

[0134] Optionally, the screening module 403 is further used for:

[0135] Determine whether the target user's crowd label belongs to any crowd label in the crowd classification set;

[0136] If yes, continue to determine whether the physical item category corresponding to the target user belongs to the physical item category set associated with the target user's population tag;

[0137] If so, the virtual categories to be recommended that are associated with the target user are screened out from the virtual category set according to the population tag of the target user.

[0138] Optionally, the screening module 403 is further used for:

[0139] According to the target user's population tag, a target user virtual category set associated with the target user is filtered out from the virtual category set;

[0140] According to the virtual category weight values ​​corresponding to the virtual categories in the target user virtual category set, the top N virtual categories are selected as the virtual categories to be recommended;

[0141] Wherein, N is an integer greater than zero.

[0142] Optionally, the updating module 404 is further configured to:

[0143] If the conversion result of the virtual category to be recommended is conversion, the virtual category weight value corresponding to the virtual category to be recommended is increased based on a preset weight enhancement coefficient.

[0144] Optionally, the updating module 404 is further configured to:

[0145] The sum of the virtual category weight value corresponding to the virtual category to be recommended and the preset weight enhancement coefficient is used as the modified virtual category weight value corresponding to the virtual category to be recommended.

[0146] According to the various embodiments described above, it can be seen that the present invention constructs a category association map based on the virtual category weight values ​​and the association relationship between the crowd classification set and the physical category set and the virtual category set, and modifies the virtual category weight values ​​corresponding to the virtual category to be recommended according to the conversion results of the virtual category to be recommended, thereby updating the category association map. The technical problem that the association between categories with different attributes cannot be achieved in the prior art. The embodiment of the present invention constructs a category relationship map through the association relationship between crowd labels and physical categories and virtual categories to achieve visual associations between categories with different attributes, which helps to explore the deep relationship between categories with different attributes. Moreover, the embodiment of the present invention also performs weight intervention based on the conversion results of the virtual category to be recommended, and continuously optimizes the virtual category weight values ​​corresponding to the virtual category to be recommended to improve the accuracy of the virtual category weight values.

[0147] It should be noted that the specific implementation content of the device for constructing a category association map described in the present invention has been described in detail in the method for constructing a category association map described above, so the repeated content will not be described here.

[0148] Figure 5 An exemplary system architecture 500 is shown in which the method for constructing a category association map or the apparatus for constructing a category association map according to an embodiment of the present invention can be applied.

[0149] like Figure 5 As shown, system architecture 500 may include terminal devices 501, 502, 503, a network 504 and a server 505. Network 504 is used to provide a medium for communication links between terminal devices 501, 502, 503 and server 505. Network 504 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0150] Users can use terminal devices 501, 502, 503 to interact with server 505 through network 504 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 501, 502, 503, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only as examples).

[0151] The terminal devices 501 , 502 , and 503 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0152] The server 505 may be a server that provides various services, such as a backend management server (only an example) that provides support for shopping websites browsed by users using the terminal devices 501, 502, and 503. The backend management server may analyze and process the received data such as the item information query request, and feed back the processing results (such as target push information, item information - only an example) to the terminal device.

[0153] It should be noted that the method for constructing a category association map provided in the embodiment of the present invention is generally executed by the server 505, and accordingly, the device for constructing a category association map is generally set in the server 505. The method for constructing a category association map provided in the embodiment of the present invention can also be executed by the terminal devices 501, 502, and 503, and accordingly, the device for constructing a category association map can be set in the terminal devices 501, 502, and 503.

[0154] It should be understood that Figure 5 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0155] Reference below Figure 6 , which shows a schematic diagram of the structure of a computer system 600 of a terminal device suitable for implementing an embodiment of the present invention. Figure 6 The terminal 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 invention.

[0156] like Figure 6As shown, the computer system 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the system 600 are also stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0157] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage section 608 as needed.

[0158] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program article, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the central processing unit (CPU) 601, the above-mentioned functions defined in the system of the present invention are executed.

[0159] It should be noted that the computer-readable medium shown in the present invention 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 or 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 invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal 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 a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0160] The flowchart and block diagram in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the system, method and computer program article according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a 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.

[0161] The modules involved in the embodiments of the present invention may be implemented by software or hardware. The modules described may also be set in a processor, for example, it may be described as follows: a processor includes an association module, a construction module, a screening module, and an update module, wherein the names of these modules do not constitute limitations on the modules themselves in some cases.

[0162] As another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes: determining a population classification set associated with the physical category set according to a physical category set, and determining a virtual category set associated with the population classification set according to the population classification set; constructing a category association map according to a pre-set virtual category weight value of the virtual category set, and the association relationship between the population classification set and the physical category set and the virtual category set; filtering out virtual categories to be recommended associated with the target user from the virtual category set; modifying the virtual category weight value corresponding to the virtual category to be recommended according to the conversion result of the virtual category to be recommended, thereby updating the category association map.

[0163] According to the technical solution of the embodiment of the present invention, because a category association map is constructed according to the virtual category weight value and the association relationship between the crowd classification set and the physical category set and the virtual category set, and the virtual category weight value corresponding to the virtual category to be recommended is modified according to the conversion result of the virtual category to be recommended, thereby updating the category association map, the technical problem that the association between different attribute categories cannot be achieved in the prior art is overcome. The embodiment of the present invention constructs a category relationship map through the association relationship between crowd labels and physical categories and virtual categories to achieve visual associations between categories with different attributes, which helps to explore the deep relationship between categories with different attributes. Moreover, the embodiment of the present invention also performs weight intervention based on the conversion result of the virtual category to be recommended, and continuously optimizes the virtual category weight value corresponding to the virtual category to be recommended to improve the accuracy of the virtual category weight value.

[0164] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for constructing a category association graph, characterized in that: include: Determine a population classification set associated with the physical category set according to the physical category set, and determine a virtual category set associated with the population classification set according to the population classification set; Constructing a category association map according to the preset virtual category weight values ​​of the virtual category set and the association relationship between the population classification set and the physical category set and the virtual category set; Filtering virtual categories to be recommended that are associated with a target user from the virtual category set; According to the conversion result of the virtual category to be recommended, the virtual category weight value corresponding to the virtual category to be recommended is modified, so as to update the category association map.

2. The method according to claim 1, characterized in that: Determining a population classification set associated with the physical category set according to the physical category set, and determining a virtual category set associated with the population classification set according to the population classification set, including: According to the identifiers of each physical item category in the physical item category set, a crowd label associated with the identifiers of each physical item category is determined, thereby obtaining a crowd classification set associated with the physical item category set; wherein the crowd classification set includes at least one of the crowd labels; Based on the crowd portrait, filter out the user identifiers associated with the crowd tag; According to the user identifiers, the identifiers of the virtual categories associated with the respective user identifiers are determined respectively, so as to obtain a virtual category set associated with the population classification set; wherein the virtual category set includes multiple virtual category identifiers.

3. The method according to claim 1, characterized in that According to the preset virtual category weight values ​​of the virtual category set, and the association relationship between the population classification set and the physical category set and the virtual category set, a category association map is constructed, including: According to the physical category set, the crowd classification set and the virtual category set, configuring in the graph construction tool an association relationship between the crowd classification set and each physical category in the physical category set and each virtual category in the virtual category set; According to the preset virtual category weight values ​​corresponding to each virtual category in the virtual category set, and the association relationship between the population classification set and each physical category in the physical category set and each virtual category in the virtual category set, a category association map is constructed by the map construction tool.

4. The method according to claim 1, characterized in that: Filtering out virtual categories to be recommended that are associated with the target user from the virtual category set includes: Determine whether the target user's crowd label belongs to any crowd label in the crowd classification set; If yes, continue to determine whether the physical item category corresponding to the target user belongs to the physical item category set associated with the target user's population tag; If so, the virtual categories to be recommended that are associated with the target user are screened out from the virtual category set according to the population tag of the target user.

5. The method according to claim 4, characterized in that According to the population tag of the target user, selecting the virtual category to be recommended associated with the target user from the virtual category set includes: According to the target user's population tag, a target user virtual category set associated with the target user is filtered out from the virtual category set; According to the virtual category weight values ​​corresponding to the virtual categories in the target user virtual category set, the top N virtual categories are selected as the virtual categories to be recommended; Wherein, N is an integer greater than zero.

6. The method according to claim 1, characterized in that According to the conversion result of the virtual category to be recommended, modifying the virtual category weight value corresponding to the virtual category to be recommended includes: If the conversion result of the virtual category to be recommended is conversion, the virtual category weight value corresponding to the virtual category to be recommended is increased based on a preset weight enhancement coefficient.

7. The method according to claim 6, characterized in that Increasing the virtual category weight value corresponding to the virtual category to be recommended based on a preset weight enhancement coefficient includes: The sum of the virtual category weight value corresponding to the virtual category to be recommended and the preset weight enhancement coefficient is used as the modified virtual category weight value corresponding to the virtual category to be recommended.

8. A device for constructing a category association map, characterized in that: include: An association module, configured to determine a population classification set associated with the physical object classification set based on the physical object classification set, and determine a virtual category set associated with the population classification set based on the population classification set; A construction module, configured to construct a category association map according to a preset virtual category weight value of the virtual category set and an association relationship between the population classification set and the physical category set and the virtual category set; A screening module, used to screen out virtual categories to be recommended that are associated with a target user from the virtual category set; The updating module is used to modify the virtual category weight value corresponding to the virtual category to be recommended according to the conversion result of the virtual category to be recommended, so as to update the category association map.

9. 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, the one or more processors implement the method according to any one of claims 1 to 7.

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

Images