Cross-platform product push method, device, equipment and storage medium based on tags

Abstract

The present invention relates to intelligent decision-making technology, and discloses a label-based cross-platform product push method, including: extracting a product user group according to a first product label set of a first platform, classifying and labeling the product user group to obtain a first user group and a second user group, obtaining a second product label set from a second platform, screening out a first product label to be recommended from the first product label set, pushing the corresponding product to the first user group in the second platform, screening out a second product label to be recommended from the second product label set, and pushing the corresponding product to the second user group in the second platform. In addition, the present invention also relates to blockchain technology, and the product user group can be obtained from a node of a blockchain. The present invention also proposes a label-based cross-platform product push method device, an electronic device, and a computer-readable storage medium. The present invention can solve the problem of low accuracy of cross-platform product push.

Description

Technical Field

[0001] The present invention relates to the field of intelligent decision-making technology, and in particular to a tag-based cross-platform product push method, device, electronic device and computer-readable storage medium. Background Art

[0002] Selecting different groups of people (label groups) based on user tags for marketing is a common method of smart marketing. The current leading Internet giants are all selecting groups of people based on user portraits for smart marketing. Different business scenarios and marketing activities will have different target groups.

[0003] Traditional tag crowd management is to conduct crowd selection marketing based on the company's user portrait tags. This method is often useful on the company's marketing platform, but if you want to carry out marketing delivery across company platforms, you cannot deliver to the same group of people quickly and effectively. Since the businesses of different professional companies within the group are often highly complementary, there will be quite a number of users who are also common users of two or more companies within the group. However, the existing technology only pushes products indiscriminately to common users across platforms. It often happens that users have participated in activities on platform A, but still receive push notifications for the same product on platform B. It is impossible to accurately push products to users across platforms. Summary of the invention

[0004] The present invention provides a tag-based cross-platform product push method, device, equipment and storage medium, the main purpose of which is to solve the problem of low accuracy of cross-platform product push.

[0005] To achieve the above purpose, the present invention provides a tag-based cross-platform product push method, comprising:

[0006] Acquire a first product tag set from a first platform, and extract a product user group according to the first product tag set;

[0007] Classify and label the product user groups to obtain a first user group and a second user group;

[0008] Acquire a second product tag set from the second platform, filter out a first product tag to be recommended from the first product tag set based on the annotations of the first user group, and push products corresponding to the first product tag to be recommended to users corresponding to the first user group on the second platform;

[0009] Based on the annotations of the second user group, the tag weights in the first product tag set are calculated, the tag weights are used to filter out the second product tag to be recommended from the second product tag set, and the products corresponding to the second product tag to be recommended are pushed to the users corresponding to the second user group in the second platform.

[0010] Optionally, extracting a product user group according to the first product tag set includes:

[0011] Extracting target users corresponding to product tags in the first product tag set from the underlying user table of the first platform;

[0012] All extracted target users are aggregated to obtain the product user group.

[0013] Optionally, the classifying and labeling the product user groups to obtain the first user group and the second user group includes:

[0014] Determine whether users in the product user group are interested in the product corresponding to the first product tag set;

[0015] If the status of the target customer is not interested, a first mark is made for the target customer, and the mark is added as a first identifier;

[0016] If the target customer's status is "interested", a second mark is made for the target user, and the mark is added as a second identifier;

[0017] All target users marked with the first identifier are aggregated as the first user group, and all target users marked with the second identifier are aggregated as the second user group.

[0018] Optionally, after aggregating all target users marked with the first identifier as the first user group and aggregating all target users marked with the second identifier as the second user group, the method further includes:

[0019] Using the product tags in the first product tag set as tag fields, and constructing a tag user table according to the tag fields, the first user group, and the second user group;

[0020] Building a key-value pair template based on the tag field, the first annotation and the second annotation;

[0021] Newly added product tags are obtained at preset time intervals, and the newly added product tags and target update users of the newly added product tags are classified and labeled using the key-value pair template, and the classified target update users and the newly added product tags are added to the tag user table.

[0022] Optionally, the filtering out a first product tag to be recommended from the first product tag set based on the annotations of the first user group, and pushing products corresponding to the first product tag to be recommended to users corresponding to the first user group on the second platform includes:

[0023] For a target user corresponding to any product label in the label user table, use the any product label as the first product label to be recommended for all first category user groups among the target users;

[0024] All first product labels to be recommended in the label user table are summarized, and products corresponding to the first product labels to be recommended are pushed to users corresponding to the first user group on the second platform.

[0025] Optionally, the calculating the tag weights in the first product tag set based on the annotations of the second user group includes:

[0026] Count the number of times the first platform pushes products to users in the second user group;

[0027] The reciprocal of the number of times the product is pushed is taken as the user participation rate;

[0028] Arrange the user participation rates of all users in the second user group into a matrix to obtain a user participation matrix;

[0029] According to the first annotation and the second annotation, the similarity of user participation rates of different users in the user participation matrix for the same product tag is calculated to obtain the similarity of the same category, and the similarity of user participation rates of different users in the user participation matrix for different product tags is calculated to obtain the similarity of different categories;

[0030] The similarity of the same category and the similarity of the different categories are calculated using a preset weight formula to obtain the label weight.

[0031] Optionally, the preset weight formula is as follows:

[0032]

[0033] Among them, δ j is the weight of the j-th product label, is the participation rate of the i-th user in the j-th product tag, For Recent user engagement rate under the same product tag, For Recent user engagement rates under different product tags, is the similarity of the same category, is the different class similarities.

[0034] In order to solve the above problems, the present invention also provides a tag-based cross-platform product push method and device, the device comprising:

[0035] A user extraction module, configured to obtain a first product tag set from a first platform, and extract a product user group according to the first product tag set;

[0036] A user labeling module, used to classify and label the product user groups to obtain a first category user group and a second category user group;

[0037] A first product push module is used to obtain a second product tag set from a second platform, filter out a first to-be-recommended product tag from the first product tag set based on the annotations of the first user group, and push products corresponding to the first to-be-recommended product tag to users corresponding to the first user group on the second platform;

[0038] The second product push module is used to calculate the label weights in the first product label set based on the annotations of the second user group, use the label weights to filter out second product labels to be recommended from the second product label set, and push products corresponding to the second product labels to be recommended to users corresponding to the second user group in the second platform.

[0039] In order to solve the above problem, the present invention further provides an electronic device, the electronic device comprising:

[0040] a memory storing at least one instruction; and

[0041] The processor executes the instructions stored in the memory to implement the above-mentioned tag-based cross-platform product push method.

[0042] In order to solve the above problem, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned tag-based cross-platform product push method.

[0043] The present invention can accurately determine the user group by extracting the product user group through the first product tag set, and classify and label the product user group according to the product tags in the first product tag set to obtain the first user group and the second user group. Different users will have different labels, and push analysis can be performed on users in a more fine-grained manner, further improving the accuracy of product recommendations to users. At the same time, different product tags to be recommended are screened out for different user groups, and products are pushed to different user groups according to the product tags to be recommended, making cross-platform product push more accurate. Therefore, the tag-based cross-platform product push method, device, electronic device and computer-readable storage medium proposed in the present invention can solve the problem of low accuracy of cross-platform product push. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1A schematic diagram of a flow chart of a tag-based cross-platform product push method provided in one embodiment of the present invention;

[0045] Figure 2 A functional module diagram of a tag-based cross-platform product push method and device provided in one embodiment of the present invention;

[0046] Figure 3 A schematic diagram of the structure of an electronic device for implementing the tag-based cross-platform product push method provided by an embodiment of the present invention.

[0047] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0048] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0049] The embodiment of the present application provides a cross-platform product push method based on tags. The execution subject of the cross-platform product push method based on tags includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the cross-platform product push method based on tags can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0050] Reference Figure 1 FIG. 1 is a flow chart of a method for pushing products across platforms based on tags according to an embodiment of the present invention. In this embodiment, the method for pushing products across platforms based on tags includes:

[0051] S1. Obtain a first product tag set from a first platform, and extract a product user group according to the first product tag set.

[0052] In the embodiment of the present invention, the first platform refers to any professional subsidiary of the group in each field, and the product label set refers to the label set of various products and promotional activities recently released or to be released. For example, in the financial field, the first platform of the AA group can be: AA Bank, AA Property Insurance and AA Trust, etc. The product label set includes: "BB Credit Card", "BB Financial Products", etc.

[0053] Specifically, extracting a product user group according to the first product tag set includes:

[0054] Extracting target users corresponding to product tags in the first product tag set from the underlying user table of the first platform;

[0055] All extracted target users are aggregated to obtain the product user group.

[0056] In the embodiment of the present invention, the underlying user table refers to an information table storing user information in the first platform database. For example, for the product label "Credit Card A", all users who have browsed "Credit Card A" are extracted from the underlying user table as target users. At the same time, for the upcoming activities, a preset number of target users can be directly circled from the underlying user table as the product user group, for example, 1 million users are selected as the product user group.

[0057] S2. Classify and label the product user groups to obtain a first user group and a second user group.

[0058] Specifically, the product user groups are classified and labeled to obtain the first user group and the second user group, including:

[0059] Determine whether users in the product user group are interested in the product corresponding to the first product tag set;

[0060] If the status of the target customer is not interested, a first mark is made for the target customer, and the mark is added as a first identifier;

[0061] If the target customer's status is "interested", a second mark is made for the target user, and the mark is added as a second identifier;

[0062] All target users marked with the first identifier are aggregated as the first user group, and all target users marked with the second identifier are aggregated as the second user group.

[0063] In an embodiment of the present invention, each product tag in the first product tag set refers to a product or activity in the first platform, and whether the user is interested in the product or activity is determined by marking the users in the product user group. For example, if the user has not participated in the product or activity corresponding to the product tag (including not browsing the product page or browsing the product page but not participating), it is considered that the user is not interested in the product or activity corresponding to the product tag, and a first marking is performed; if the user has participated in the product or activity corresponding to the product tag, it is considered that the user is interested in the product or activity corresponding to the product tag, and a second marking is performed, wherein the first marking can be 0 and the second marking can be 1.

[0064] In another optional embodiment of the present invention, after aggregating all target users marked with the first identifier as the first user group and aggregating all target users marked with the second identifier as the second user group, the method further includes:

[0065] Using the product tags in the first product tag set as tag fields, and constructing a tag user table according to the tag fields, the first user group, and the second user group;

[0066] Building a key-value pair template based on the tag field, the first annotation and the second annotation;

[0067] Newly added product tags are obtained at preset time intervals, and the newly added product tags and target update users of the newly added product tags are classified and labeled using the key-value pair template, and the classified target update users and the newly added product tags are added to the tag user table.

[0068] In the embodiment of the present invention, a field refers to a column name storing data in a table, for example, the fields of product table A may include product ID, product name, product barcode, etc. The tag user table may be a Hive table, which stores data in string format, including product tags and corresponding target users.

[0069] In an optional embodiment of the present invention, the key-value pair template can be a key-value pair, wherein the product tag is the value of the key, and the value is Boolean type data 0 or 1. The key-value pair template is used to classify and label the target update users of the newly added product tags and add them to the tag user table. Without manual configuration, automatic user classification labeling can be achieved, thereby improving product push efficiency.

[0070] In an embodiment of the present invention, the label user table includes labels corresponding to all products or activities in the first platform, and the labeling of users under each label. For example, the label user table includes two product labels: A and B, and three target users: a, b, and c. The classification labels of users under each product label are: A: 0, 0, 1; B: 1, 1, 1.

[0071] S3. Obtain a second product tag set from the second platform, filter out a first product tag to be recommended from the first product tag set based on the annotations of the first user group, and push products corresponding to the first product tag to be recommended to users corresponding to the first user group on the second platform.

[0072] In the embodiment of the present invention, the second product tag set includes products or activities that have been released or are about to be released on the second platform.

[0073] Specifically, the filtering out a first product tag to be recommended from the first product tag set based on the annotations of the first user group, and pushing the product corresponding to the first product tag to be recommended to the user corresponding to the first user group on the second platform includes:

[0074] For a target user corresponding to any product label in the label user table, use the any product label as the first product label to be recommended for all first category user groups among the target users;

[0075] All first product labels to be recommended in the label user table are summarized, and products corresponding to the first product labels to be recommended are pushed to users corresponding to the first user group on the second platform.

[0076] In an optional embodiment of the present invention, for example, taking two professional companies within the group as an example, the label user table includes the product label "AA Credit Card" of Company A, and 1 million people are circled as target customers for this product label, of which 700,000 are the first user group, that is, users who have not participated in the activity. The product label "AA Credit Card" is used as the first product label to be recommended for the first user group, and products or activities corresponding to the product label "AA Credit Card" are pushed to users in Company B who belong to the first user group.

[0077] S4. Based on the annotations of the second user group, calculate the tag weights in the first product tag set, use the tag weights to filter out second product tags to be recommended from the second product tag set, and push products corresponding to the second product tags to be recommended to users corresponding to the second user group in the second platform.

[0078] In the embodiment of the present invention, since the label user table includes multiple product labels, there will be a second user group under each product label. Each user has different preferences for different product labels. By calculating the user's weight for each product label, more accurate product push can be performed in the second platform.

[0079] Specifically, the calculating the tag weights in the first product tag set based on the annotations of the second user group includes:

[0080] Count the number of times the first platform pushes products to users in the second user group;

[0081] The reciprocal of the number of times the product is pushed is taken as the user participation rate;

[0082] Arrange the user participation rates of all users in the second user group into a matrix to obtain a user participation matrix;

[0083] According to the first annotation and the second annotation, the similarity of user participation rates of different users in the user participation matrix for the same product tag is calculated to obtain the similarity of the same category, and the similarity of user participation rates of different users in the user participation matrix for different product tags is calculated to obtain the similarity of different categories;

[0084] The similarity of the same category and the similarity of the different categories are calculated using a preset weight formula to obtain the label weight.

[0085] In an optional embodiment of the present invention, the Relief feature selection algorithm can be used to calculate the weight of each user in the second user group for each product label. For example, for each user x in the second user group i , using the Relief feature selection algorithm, we can get i The nearest neighbor x with the same product label i,nh (the recent user participation rate under the same tag), and x i Wrong nearest neighbors x for different product labels i,nm (the most recent user participation rate under different tags), the similarity may be the Euclidean distance.

[0086] In the embodiment of the present invention, the preset weight formula is as follows:

[0087]

[0088] Among them, δ j is the weight of the j-th product label, is the participation rate of the i-th user in the j-th product tag, For Recent user engagement rates under the same product tag, For Recent user engagement rates under different product tags, is the similarity of the same category, is the different class similarities.

[0089] In an optional embodiment of the present invention, taking product labels A, B, and C as an example, if the distance between the user and his / her nearest neighbor who guessed correctly on product label A is smaller than the distance between the user and his / her nearest neighbor who guessed incorrectly on this dimension, it indicates that the user prefers this type of label (i.e., product or activity), and thus the efficiency of product push can be improved.

[0090] In an embodiment of the present invention, products can be pushed to the second platform based on label weights. For example, the label weights calculated by user a for product labels A, B, and C are [10, 4, 6], and product label A is "CC Financial Management". Then, labels corresponding to financial products are selected from the second product label set as the second product labels to be recommended, and products corresponding to such product labels are pushed to user a on the second platform.

[0091] The present invention can accurately determine the user group by extracting the product user group through the first product tag set, and classify and label the product user group according to the product tags in the first product tag set to obtain the first user group and the second user group. Different users will have different labels, and the push analysis can be performed on users in a more fine-grained manner, further improving the accuracy of product recommendations for users. At the same time, different product tags to be recommended are screened out for different user groups, and products are pushed to different user groups according to the product tags to be recommended, making cross-platform product push more accurate. Therefore, the tag-based cross-platform product push method proposed in the present invention can solve the problem of low accuracy of cross-platform product push.

[0092] like Figure 2 , which is a functional module diagram of a tag-based cross-platform product push device provided by an embodiment of the present invention.

[0093] The tag-based cross-platform product push device 100 of the present invention can be installed in an electronic device. According to the functions implemented, the tag-based cross-platform product push device 100 can include a user extraction module 101, a user annotation module 102, a first product push module 103 and a second product push module 104. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0094] In this embodiment, the functions of each module / unit are as follows:

[0095] The user extraction module 101 is used to obtain a first product tag set from a first platform, and extract a product user group according to the first product tag set.

[0096] In the embodiment of the present invention, the first platform refers to any professional subsidiary of the group in each field, and the product label set refers to the label set of various products and promotional activities recently released or to be released. For example, in the financial field, the first platform of the AA group can be: AA Bank, AA Property Insurance and AA Trust, etc. The product label set includes: "BB Credit Card", "BB Financial Products", etc.

[0097] Specifically, the user extraction module 101 extracts the product user group by performing the following operations:

[0098] Extracting target users corresponding to product tags in the first product tag set from the underlying user table of the first platform;

[0099] All extracted target users are aggregated to obtain the product user group.

[0100] In the embodiment of the present invention, the underlying user table refers to an information table storing user information in the first platform database. For example, for the product label "Credit Card A", all users who have browsed "Credit Card A" are extracted from the underlying user table as target users. At the same time, for the upcoming activities, a preset number of target users can be directly circled from the underlying user table as the product user group, for example, 1 million users are selected as the product user group.

[0101] The user labeling module 102 is used to classify and label the product user groups to obtain a first category of user groups and a second category of user groups.

[0102] Specifically, the user labeling module 102 classifies and labels the product user groups by performing the following operations to obtain the first user group and the second user group:

[0103] Determine whether users in the product user group are interested in the product corresponding to the first product tag set;

[0104] If the status of the target customer is not interested, a first mark is made for the target customer, and the mark is added as a first identifier;

[0105] If the target customer's status is "interested", a second mark is made for the target user, and the mark is added as a second identifier;

[0106] All target users marked with the first identifier are aggregated as the first user group, and all target users marked with the second identifier are aggregated as the second user group.

[0107] In an embodiment of the present invention, each product tag in the first product tag set refers to a product or activity in the first platform, and whether the user is interested in the product or activity is determined by marking the users in the product user group. For example, if the user has not participated in the product or activity corresponding to the product tag (including not browsing the product page or browsing the product page but not participating), it is considered that the user is not interested in the product or activity corresponding to the product tag, and a first marking is performed; if the user has participated in the product or activity corresponding to the product tag, it is considered that the user is interested in the product or activity corresponding to the product tag, and a second marking is performed, wherein the first marking can be 0 and the second marking can be 1.

[0108] In another optional embodiment of the present invention, the user marking module 102 further includes:

[0109] Using the product tags in the first product tag set as tag fields, and constructing a tag user table according to the tag fields, the first user group, and the second user group;

[0110] Building a key-value pair template based on the tag field, the first annotation and the second annotation;

[0111] Newly added product tags are obtained at preset time intervals, and the newly added product tags and target update users of the newly added product tags are classified and labeled using the key-value pair template, and the classified target update users and the newly added product tags are added to the tag user table.

[0112] In the embodiment of the present invention, a field refers to a column name storing data in a table, for example, the fields of product table A may include product ID, product name, product barcode, etc. The tag user table may be a Hive table, which stores data in string format, including product tags and corresponding target users.

[0113] In an optional embodiment of the present invention, the key-value pair template can be a key-value pair, wherein the product tag is the value of the key, and the value is Boolean type data 0 or 1. The key-value pair template is used to classify and label the target update users of the newly added product tags and add them to the tag user table. Without manual configuration, automatic user classification labeling can be achieved, thereby improving product push efficiency.

[0114] In an embodiment of the present invention, the label user table includes labels corresponding to all products or activities in the first platform, and the labeling of users under each label. For example, the label user table includes two product labels: A and B, and three target users: a, b, and c. The classification labels of users under each product label are: A: 0, 0, 1; B: 1, 1, 1.

[0115] The first product push module 103 is used to obtain a second product tag set from the second platform, filter out a first product tag to be recommended from the first product tag set based on the annotations of the first user group, and push products corresponding to the first product tag to be recommended to users corresponding to the first user group on the second platform.

[0116] In the embodiment of the present invention, the second product tag set includes products or activities that have been released or are about to be released on the second platform.

[0117] Specifically, the first product push module 103 pushes the product corresponding to the first to-be-recommended product tag to the users corresponding to the first user group on the second platform by performing the following operations:

[0118] For a target user corresponding to any product label in the label user table, use the any product label as the first product label to be recommended for all first category user groups among the target users;

[0119] All first product labels to be recommended in the label user table are summarized, and products corresponding to the first product labels to be recommended are pushed to users corresponding to the first user group on the second platform.

[0120] In an optional embodiment of the present invention, for example, taking two professional companies within the group as an example, the label user table includes the product label "AA Credit Card" of Company A, and 1 million people are circled as target customers for this product label, of which 700,000 are the first user group, that is, users who have not participated in the activity. The product label "AA Credit Card" is used as the first product label to be recommended for the first user group, and products or activities corresponding to the product label "AA Credit Card" are pushed to users in Company B who belong to the first user group.

[0121] The second product push module 104 is used to calculate the label weights in the first product label set based on the annotations of the second user group, use the label weights to filter out second product labels to be recommended from the second product label set, and push products corresponding to the second product labels to be recommended to users corresponding to the second user group in the second platform.

[0122] In the embodiment of the present invention, since the label user table includes multiple product labels, there will be a second user group under each product label. Each user has different preferences for different product labels. By calculating the user's weight for each product label, more accurate product push can be performed in the second platform.

[0123] Specifically, the second product push module 104 calculates the tag weights in the first product tag set by performing the following operations:

[0124] Count the number of times the first platform pushes products to users in the second user group;

[0125] The reciprocal of the number of times the product is pushed is taken as the user participation rate;

[0126] Arrange the user participation rates of all users in the second user group into a matrix to obtain a user participation matrix;

[0127] According to the first annotation and the second annotation, the similarity of user participation rates of different users in the user participation matrix for the same product tag is calculated to obtain the similarity of the same category, and the similarity of user participation rates of different users in the user participation matrix for different product tags is calculated to obtain the similarity of different categories;

[0128] The similarity of the same category and the similarity of the different categories are calculated using a preset weight formula to obtain the label weight.

[0129] In an optional embodiment of the present invention, the Relief feature selection algorithm can be used to calculate the weight of each user in the second user group for each product label. For example, for each user x in the second user group i , using the Relief feature selection algorithm, we can get i The nearest neighbor x with the same product label i,nh (the recent user participation rate under the same tag), and x i Wrong nearest neighbors x for different product labels i,nm (the most recent user participation rate under different tags), the similarity may be the Euclidean distance.

[0130] In the embodiment of the present invention, the preset weight formula is as follows:

[0131]

[0132] Among them, δ j is the weight of the j-th product label, is the participation rate of the i-th user in the j-th product tag, For Recent user engagement rates under the same product tag, For Recent user engagement rates under different product tags, is the similarity of the same category, is the different class similarities.

[0133] In an optional embodiment of the present invention, taking product labels A, B, and C as an example, if the distance between the user and his / her nearest neighbor who guessed correctly on product label A is smaller than the distance between the user and his / her nearest neighbor who guessed incorrectly on this dimension, it indicates that the user prefers this type of label (i.e., product or activity), and thus the efficiency of product push can be improved.

[0134] In an embodiment of the present invention, products can be pushed to the second platform based on label weights. For example, the label weights calculated by user a for product labels A, B, and C are [10, 4, 6], and product label A is "CC Financial Management". Then, labels corresponding to financial products are selected from the second product label set as the second product labels to be recommended, and products corresponding to such product labels are pushed to user a on the second platform.

[0135] The present invention can accurately determine the user group by extracting the product user group through the first product tag set, and classify and label the product user group according to the product tags in the first product tag set to obtain the first user group and the second user group. Different users will have different labels, and push analysis can be performed on users in a more fine-grained manner, further improving the accuracy of product recommendations to users. At the same time, different product tags to be recommended are screened out for different user groups, and products are pushed to different user groups according to the product tags to be recommended, making cross-platform product push more accurate. Therefore, the label-based cross-platform product push device proposed in the present invention can solve the problem of low accuracy of cross-platform product push.

[0136] like Figure 3 , is a schematic diagram of the structure of an electronic device for implementing a tag-based cross-platform product push method provided by an embodiment of the present invention, including a processor 111, a communication interface 112, a memory 113 and a communication bus 114, wherein the processor 111, the communication interface 112, and the memory 113 communicate with each other through the communication bus 114.

[0137] The memory 113 is used to store computer programs, such as a tag-based cross-platform product push method program;

[0138] In one embodiment of the present application, the processor 111 is configured to execute a program stored in the memory 113 to implement a tag-based cross-platform product push method provided by any of the aforementioned method embodiments, including:

[0139] Acquire a first product tag set from a first platform, and extract a product user group according to the first product tag set;

[0140] Classify and label the product user groups to obtain a first user group and a second user group;

[0141] Acquire a second product tag set from the second platform, filter out a first product tag to be recommended from the first product tag set based on the annotations of the first user group, and push products corresponding to the first product tag to be recommended to users corresponding to the first user group on the second platform;

[0142] Based on the annotations of the second user group, the tag weights in the first product tag set are calculated, the tag weights are used to filter out the second product tag to be recommended from the second product tag set, and the products corresponding to the second product tag to be recommended are pushed to the users corresponding to the second user group in the second platform.

[0143] The communication bus 114 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus 114 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0144] The communication interface 112 is used for communication between the above electronic device and other devices.

[0145] The memory 113 may include a random access memory (RAM) or a non-volatile memory, such as at least one disk memory. Optionally, the memory 113 may also be at least one storage device located away from the processor 111 .

[0146] The above-mentioned processor 111 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0147] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, the computer program can implement:

[0148] Acquire a first product tag set from a first platform, and extract a product user group according to the first product tag set;

[0149] Classify and label the product user groups to obtain a first user group and a second user group;

[0150] Acquire a second product tag set from the second platform, filter out a first product tag to be recommended from the first product tag set based on the annotations of the first user group, and push products corresponding to the first product tag to be recommended to users corresponding to the first user group on the second platform;

[0151] Based on the annotations of the second user group, the tag weights in the first product tag set are calculated, the tag weights are used to filter out the second product tag to be recommended from the second product tag set, and the products corresponding to the second product tag to be recommended are pushed to the users corresponding to the second user group in the second platform.

[0152] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0153] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0154] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0155] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0156] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0157] The blockchain referred to in this invention is a new application mode of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0158] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in the system claim can also be implemented by one unit or device through software or hardware. The second and other words are used to indicate names, but not to indicate any particular order.

[0159] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A tag-based cross-platform product push method, characterized in that: The method comprises: Acquire a first product tag set from a first platform, and extract a product user group according to the first product tag set; Determine whether the user in the product user group is interested in the product corresponding to the first product tag set; if the target user is not interested, perform a first tag on the target user, adding the tag as a first identifier; if the target user is interested, perform a second tag on the target user, adding the tag as a second identifier; aggregate all target users marked with the first identifier as a first user group, and aggregate all target users marked with the second identifier as a second user group; Acquire a second product tag set from the second platform, filter out a first product tag to be recommended from the first product tag set based on the annotations of the first user group, and push products corresponding to the first product tag to be recommended to users corresponding to the first user group on the second platform; Based on the annotations of the second user group, calculate the tag weights in the first product tag set, filter out second to-be-recommended product tags from the second product tag set using the tag weights, and push products corresponding to the second to-be-recommended product tags to users corresponding to the second user group on the second platform; The calculating of the tag weights in the first product tag set based on the annotations of the second user group includes: counting the number of product pushes by the first platform to users in the second user group; taking the inverse of the number of product pushes as the user participation rate; arranging the user participation rates of all users in the second user group in a matrix to obtain a user participation matrix; calculating the similarity of user participation rates of different users in the user participation matrix for the same product tag according to the first annotation and the second annotation to obtain the similarity of the same category, and calculating the similarity of user participation rates of different users in the user participation matrix for different product tags to obtain the similarity of different categories; calculating the similarity of the same category and the similarity of different categories using a preset weight formula to obtain the tag weight; The preset weight formula is as follows: in, is the weight of the j-th product label, is the participation rate of the i-th user in the j-th product tag, For Recent user engagement rates under the same product tag, For Recent user engagement rates under different product tags, is the similarity of the same category, is the different class similarities.

2. The tag-based cross-platform product push method according to claim 1, characterized in that: The extracting a product user group according to the first product tag set includes: Extracting target users corresponding to product tags in the first product tag set from the underlying user table of the first platform; All extracted target users are aggregated to obtain the product user group.

3. The tag-based cross-platform product push method according to claim 1, characterized in that: After aggregating all target users marked with the first identifier as the first user group and aggregating all target users marked with the second identifier as the second user group, the method further includes: Using the product tags in the first product tag set as tag fields, and constructing a tag user table according to the tag fields, the first user group, and the second user group; Building a key-value pair template based on the tag field, the first annotation and the second annotation; Newly added product tags are obtained at preset time intervals, and the newly added product tags and target update users of the newly added product tags are classified and labeled using the key-value pair template, and the classified target update users and the newly added product tags are added to the tag user table.

4. The tag-based cross-platform product push method as claimed in claim 1, characterized in that: The step of selecting a first product tag to be recommended from the first product tag set based on the annotations of the first user group, and pushing products corresponding to the first product tag to be recommended to users corresponding to the first user group on the second platform includes: For a target user corresponding to any product label in the label user table, use the any product label as the first product label to be recommended for all first category user groups among the target users; All first product labels to be recommended in the label user table are summarized, and products corresponding to the first product labels to be recommended are pushed to users corresponding to the first user group on the second platform.

5. A tag-based cross-platform product push device, used to implement the tag-based cross-platform product push method according to any one of claims 1 to 4, characterized in that: The device comprises: A user extraction module, configured to obtain a first product tag set from a first platform, and extract a product user group according to the first product tag set; A user labeling module, used to classify and label the product user groups to obtain a first category user group and a second category user group; A first product push module is used to obtain a second product tag set from a second platform, filter out a first to-be-recommended product tag from the first product tag set based on the annotations of the first user group, and push products corresponding to the first to-be-recommended product tag to users corresponding to the first user group on the second platform; The second product push module is used to calculate the label weights in the first product label set based on the annotations of the second user group, use the label weights to filter out second product labels to be recommended from the second product label set, and push products corresponding to the second product labels to be recommended to users corresponding to the second user group in the second platform.

6. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the tag-based cross-platform product push method as described in any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the tag-based cross-platform product push method according to any one of claims 1 to 4 is implemented.

Images