User recommendation method and device and readable storage medium

Abstract

The invention discloses a user recommendation method and device and a readable storage medium, and the method comprises the steps: determining a geographic region corresponding to an enterprise user based on the enterprise related information of the enterprise user accessing a target application; determining a geographic area corresponding to the terminal user based on a location based service (LBS) of the terminal user accessing the target application; obtaining each target enterprise user and each target terminal user in the same geographic area; and based on the user behavior data and the enterprise behavior data, recommending a matched target terminal user to each target enterprise user.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet, and particularly relates to a user recommendation method and device and a readable storage medium. BACKGROUND

[0002] With the rapid development of network communication technology, the popularity of 5G and its wide application in the Internet, the coverage of users is significantly improved, and the network use habits of users have gradually developed. The rapid growth of various Internet applications has injected strong impetus into the Internet economy. At the same time, the C-end user group of the Internet application is continuously expanding, and it is crucial for enterprise users of the Internet application to improve their service promotion and marketing recommendation optimization. How to accurately locate the C-end users related to the enterprise users and improve the accuracy of the C-end user recommendation corresponding to the enterprise users is a technical problem to be solved at present. SUMMARY

[0003] The purpose of the embodiments of the present application is to provide a user recommendation method and device and a readable storage medium to solve the problem of inaccurate user recommendation.

[0004] In order to solve the above technical problems, the present specification is implemented as follows: In a first aspect, a user recommendation method is provided, comprising: determining a geographic area corresponding to an enterprise user based on enterprise-related information of the enterprise user accessing a target application; determining a geographic area corresponding to a terminal user based on location-based services (LBS) of the terminal user accessing the target application; obtaining each target enterprise user and each target terminal user located in the same geographic area; recommending a matched target terminal user for each target enterprise user based on user behavior data and enterprise behavior data.

[0005] Optionally, the step of determining the geographic area corresponding to the enterprise user based on the enterprise-related information of the enterprise user accessing the target application comprises: determining the geographic area corresponding to the enterprise user based on location-related data in the enterprise-related information, wherein the location-related data comprises a business license registration location of the enterprise user, an IP address and a port of the enterprise user currently logging in the target application.

[0006] Optionally, the step of recommending the matched target terminal user for the target enterprise user based on the user behavior data and the enterprise behavior data comprises: obtaining a user portrait of each target terminal user based on collected compliant user behavior data; obtaining enterprise behavior data of each target enterprise user, the enterprise behavior data including industry information, service information of the target enterprise user, historical publishing records and social interaction information of the target enterprise user in the target application, and the target enterprise user including at least one service; matching corresponding services for each target terminal user in each target enterprise user according to the user portrait of each target terminal user and the enterprise behavior data of each target enterprise user; displaying service information of the target enterprise user to which the corresponding target service belongs to each target terminal user who matches the target service.

[0007] Optionally, further comprising: displaying user information of each target terminal user who matches the target service to the target enterprise user to which the target service belongs.

[0008] Optionally, the user behavior data includes historical browsing records, consumption information, and social interaction information of the target terminal user in the target application. The matching of the corresponding services for each target terminal user in each target enterprise user according to the user portrait of each target terminal user and the enterprise behavior data of each target enterprise user comprises: extracting service features of the corresponding target terminal user based on the user portrait of each target terminal user; extracting service features of the corresponding target enterprise user based on the enterprise behavior data of each target enterprise user; matching the corresponding services for each target terminal user by calculating the similarity between the service features of each target terminal user and the service features of each target enterprise user.

[0009] Optionally, the extraction of the service features of the corresponding target terminal user based on the user portrait of each target terminal user comprises: extracting text data in the user portrait of any target terminal user; obtaining corresponding word vectors by segmenting the text data; comparing the word vectors with a special dictionary to determine word vectors matching special words in the special dictionary, the special dictionary being pre-constructed based on enterprise-related information and including industry classification and service classification; and determining the matching word vectors as the service features of the target terminal user.

[0010] Optionally, the displaying of the service information of the target enterprise user to which the corresponding target service belongs to each target terminal user who matches the target service further comprises: adjusting a frequency of displaying corresponding service information to each target terminal user matching the target service according to a level of a target enterprise user to which the target service belongs, the level being related to an activity of the target enterprise user in the target application; or in a case where any target terminal user matching the target service does not generate user behavior data within a preset time, displaying service information of a target enterprise user to which the target service belongs to the target terminal user at a random time.

[0011] Optionally, before determining the geographic area of the enterprise user based on the enterprise-related information of the enterprise user accessing the target application, the method further comprises: receiving authentication information of the target enterprise user, the authentication information comprising a registered place of a business license of the target enterprise user, industry information and service information of the target enterprise user; sending the authentication information to a corresponding service provider end through an HTTPS protocol for authentication; in a case where the enterprise-related information authenticated by the service provider end is received, obtaining corresponding industry classification and service classification by respectively marking the industry information and the service information in the enterprise-related information; adding the industry classification and the service classification of the target enterprise user to the special dictionary.

[0012] In a third aspect, a user recommendation apparatus is provided, comprising a processor and a memory, the memory storing programs or instructions executable on the processor, and the programs or instructions, when executed by the processor, implement the steps of the method according to the first aspect.

[0013] In a fourth aspect, a readable storage medium is provided, the readable storage medium storing programs or instructions, and the programs or instructions, when executed by a processor, implement the steps of the method according to the first aspect.

[0014] In a fifth aspect, a computer program product is provided, the computer program product comprising a non-transitory computer readable storage medium storing a computer program, and the computer program being operable to cause a computer to perform the steps of the method according to the first aspect.

[0015] In the embodiments of the present application, the geographic area corresponding to the enterprise user is determined based on the enterprise-related information of the enterprise user accessing the target application; the geographic area corresponding to the terminal user is determined based on the location-based service (LBS) of the terminal user accessing the target application; each target enterprise user and each target terminal user located in the same geographic area are obtained; and the matched target terminal user is recommended for each target enterprise user based on the user behavior data and the enterprise behavior data. Thus, the precise positioning of the terminal user (i.e., C-end user) and the enterprise user can be realized, the localized content and service of the local enterprise user are presented to the local C-end user, the personalized intelligent recommendation to the local C-end user is realized, the effect of the fine service recommendation and delivery of the enterprise is improved, the recommendation cost and customer maintenance cost are reduced, the needs of the enterprise user for personalized recommendation, the C-end user for social interaction, and the private domain traffic marketing conversion are met, the experience of the enterprise user is improved, and the diversified needs of the local C-end user are solved. The service information, activity notification, or other targeted content of the local enterprise user can be positioned according to the location of the C-end user, and the interaction between the C-end user and the local enterprise user is enhanced. BRIEF DESCRIPTION OF DRAWINGS

[0016] The accompanying drawings, which are included to provide a further understanding of the present application, constitute a part of the present application and illustrate the illustrative embodiments of the present application and their description serves to explain the present application, and do not constitute improper limitations on the present application. In the drawings: Figure 1 FIG. 1 is a flow diagram of a user recommendation method according to an embodiment of the present application.

[0017] Figure 2 FIG. 1 is a flow diagram of a user recommendation method according to an embodiment of the present application.

[0018] Figure 3 FIG. 1 is a flow diagram of a user recommendation method according to an embodiment of the present application. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application. The numbering of the drawings in the present application is only used to distinguish each step in the scheme, and is not used to limit the execution order of each step, and the specific execution order is subject to the description in the specification.

[0020] To solve the problems in the prior art, the embodiments of the present application provide a user recommendation method, which positions the local terminal user, i.e., C-end user, based on the location of the enterprise user. As shown in FIG. 1, the method comprises the following steps. Figure 1As shown, comprising the following steps 102 to step 108.

[0021] Step 102, based on the enterprise related information of the enterprise user accessing the target application, determining the geographical area corresponding to the enterprise user.

[0022] The enterprise related information is the information supplemented after the authentication information provided by the target enterprise is audited and passed when the target enterprise enters the target application, including enterprise ID, geographical area information, industry information, product information and marketing content, etc. The authentication information may include, for example, the registered place of the business license of the target enterprise, the authentication mobile phone number of the target enterprise, the industry information and product information of the target enterprise.

[0023] The authenticity of the authentication information needs to be audited and authenticated, and after passing the authentication, more information of the enterprise can be obtained and added to the authentication information, so as to obtain the enterprise related information.

[0024] Based on the scheme provided in the above embodiment, optionally, in the above step 102, based on the enterprise related information of the enterprise user accessing the target application, determining the geographical area corresponding to the enterprise user, comprising: based on the position related data in the enterprise related information, determining the geographical area corresponding to the enterprise user, the position related data including the registered place of the business license of the enterprise user, the IP address and port of the enterprise user currently logging in the target application.

[0025] First, the geographical area position of the enterprise user is located through the position related data, including the registered place of the business license of the enterprise user, the mobile phone number registered by the enterprise user in the target application and the IP address and port currently logging in the target application. Through these position related data, the geographical area to which the enterprise user belongs can be determined. Different types of position related data may locate different geographical positions of the enterprise user, including the geographical position corresponding to the original registered place of the business license of the enterprise user, or the geographical position corresponding to the mobile phone number or the IP address and port. Since the IP address and port may change, there may be differences between these geographical positions.

[0026] Step 104, based on the location service LBS of the terminal user accessing the target application, determining the geographical area corresponding to the terminal user.

[0027] Location Based Services (LBS) is to obtain the current location of the positioning device by using various types of positioning technology, and to provide information resources and basic services to the positioning device through mobile Internet. By accessing LBS service, after determining the geographical area of the enterprise user, the geographical area corresponding to the terminal user can be obtained through mobile Internet.

[0028] Step 106, obtaining each target enterprise user and each target terminal user located in the same geographic area.

[0029] After determining the geographic area corresponding to the enterprise user and the geographic area corresponding to the terminal user respectively through steps 102 and 104, the terminal user related to the location of the enterprise user, such as the terminal user located in the same geographic area, can be further determined. The obtained terminal user here is, for example, a local target terminal user registered in the same target application as the target enterprise user. Here, the local target terminal user refers to a terminal user located within a certain geographic area range of the location of the target enterprise user. If the location of the target enterprise user includes multiple different locations, then based on the corresponding geographic area range, the local target terminal user corresponding to each of the multiple different locations can be obtained respectively.

[0030] Step 108, recommending a matching target terminal user for each target enterprise user based on user behavior data and enterprise behavior data.

[0031] After determining the target terminal user located in the same geographic area as the target enterprise in step 106, the user behavior data of the corresponding target terminal user and the enterprise behavior data of the target enterprise can be collected to recommend and match the target enterprise user and the target terminal user based on the collected behavior data.

[0032] Here, the collected user behavior data is compliant user behavior data, including non-private user behavior data allowed to be collected by laws and regulations, and private user behavior data authorized to be collected by the user.

[0033] Based on the scheme provided in the above embodiment, optionally, in step 108, the method for recommending a matching target terminal user for a target enterprise user based on user behavior data and enterprise behavior data includes: obtaining a user portrait of each target terminal user based on the collected compliant user behavior data; obtaining enterprise behavior data of each target enterprise user, the enterprise behavior data including industry information, service information of the target enterprise user, historical publication records and social interaction information of the target enterprise user in the target application, and the target enterprise user including at least one service; matching a corresponding service for each target terminal user in each target enterprise user according to the user portrait of each target terminal user and the enterprise behavior data of each target enterprise user; and displaying service information of the target enterprise user to which the corresponding target service belongs to each target terminal user matched with the target service.

[0034] In the above steps, based on the collected compliant user behavior data, the user portrait of each target terminal user is obtained, including: collecting the compliant user behavior data of each target terminal user, including the historical browsing records, consumption information and social interaction information of each target terminal user in the target application, and determining the user portrait of the target terminal user by analyzing the user behavior data.

[0035] The target enterprise user and the local target terminal user correspond to the same application, the target enterprise user is stationed in the application, and the local target terminal user is also registered in the application. The local target terminal user can browse pages, order goods and / or perform social interactions such as commenting, liking, sharing, collecting, scoring, etc. with other enterprises or users in the application, thereby corresponding to generate historical browsing records, consumption information, social interaction information, etc.

[0036] Based on these user behavior data, the corresponding user portrait can be analyzed, such as the user type to which each local target terminal user belongs, the corresponding consumption level, the user age and gender, etc. The local target terminal user has browsed which goods, ordered which goods, and interacted with which goods of which enterprise user. Thus, the user portrait of the local target terminal user can be obtained.

[0037] According to the user portrait, the target enterprise user is recommended to match the target terminal user, and the matched target terminal user can include multiple target terminal users.

[0038] Based on the scheme provided in the above embodiment, optionally, the user behavior data includes historical browsing records, consumption information, and social interaction information of the target terminal user in the target application; and the matching of the corresponding service for each target terminal user among each target enterprise user based on the user portrait of each target terminal user and the enterprise behavior data of each target enterprise user includes: extracting the service features of the corresponding target terminal user based on the user portrait of each target terminal user; extracting the service features of the corresponding target enterprise user based on the enterprise behavior data of each target enterprise user; and matching the corresponding service for each target terminal user by calculating the similarity between the service features of each target terminal user and the service features of each target enterprise user.

[0039] Based on the user portrait, the service features of the target terminal user can be extracted, which can also include the age features, gender features, and occupation features of the user.

[0040] Specifically, the service feature of the target terminal user is extracted based on the user portrait of each target terminal user, including: for any target terminal user, extracting text data in the user portrait of the target terminal user; obtaining corresponding word vectors by performing word segmentation on the text data; comparing each word vector with a special dictionary to determine a word vector matching a special word in the special dictionary, the special dictionary being pre-constructed based on enterprise-related information and including industry classification and service classification; and determining the matching word vector as the service feature of the target terminal user.

[0041] First, the text data in the user portrait is pre-processed, and the word vectors in the text data can be segmented by a word segmentation model to construct word vectors. At least one word vector obtained by word segmentation is compared with a pre-constructed special dictionary to determine a special word vector matching the target terminal user in the special dictionary. The special dictionary is pre-constructed based on enterprise-related information and includes industry classification and service classification. According to the industry classification and service classification corresponding to the special word vector, the service feature matching the target terminal user can be determined, and thus the preliminary semantic classification of the text data can be obtained to match and predict the local terminal user and the local enterprise user.

[0042] In one embodiment, before determining the geographic area of the enterprise user based on the enterprise-related information of the target application, the method further includes: receiving authentication information of the target enterprise user, the authentication information including the registered place of the business license of the target enterprise user, industry information and service information of the target enterprise user; sending the authentication information to the corresponding service provider side for authentication through the HTTPS protocol; in the case of receiving the enterprise-related information fed back by the service provider side after passing the authentication, obtaining the corresponding industry classification and service classification by respectively labeling the industry information and service information in the enterprise-related information; and adding the industry classification and service classification of the target enterprise user to the special dictionary.

[0043] When the enterprise user first enters the target application, it is necessary to verify whether the enterprise user can pass the authentication. The authentication verification is based on the authentication information provided by the enterprise user. In the case of a large number of enterprise authentications, the authentication information of the enterprise user can be sent to the service provider side of the third party for assistance in authentication. The authentication information of the enterprise user includes the registered place of the business license of the enterprise user, the authentication mobile phone number of the enterprise user, the industry information and service information of the enterprise user.

[0044] The authentication information is sent to the corresponding service provider end through the HTTPS protocol negotiated with the service provider end and is authenticated. The service provider end can further communicate and mine with the corresponding enterprise user based on the authentication information of the enterprise user to supplement complete enterprise-related information and audit whether it meets the authentication requirements. If the authentication is passed, the enterprise-related information that passes the authentication is fed back. When building a special dictionary, the industry information and service information in the enterprise-related information that passes the authentication can be respectively marked to obtain the real and effective industry classification and service classification of the enterprise user. The industry classification and service classification are added to the special dictionary to enrich the vocabulary of the special dictionary and facilitate to improve the accuracy of subsequent terminal user matching.

[0045] Figure 2 Different embodiments of the authentication information audit of the enterprise user of the present application are given, wherein (a) in the corresponding embodiment, the corresponding service provider end is selected through the agent_uid of the service provider end to realize automatic issuance of authentication information (step 202), and the corresponding issued list is obtained after issuance (204) to facilitate knowing the issued authentication information and the corresponding service provider end.

[0046] In the (b) corresponding embodiment, the corresponding service provider end is selected by recording the pre-allocated list of agent_uid of the service provider end (step 302) to realize automatic issuance of authentication information (step 304), and the corresponding issued list is obtained after issuance (step 306). After automatic issuance, it can be automatically / judicially determined whether the issued service provider end has an error (step 310). If so, the error service provider end is changed (step 312), and the step 304 is continued to execute the issuance.

[0047] In the (c) corresponding embodiment, the corresponding service provider end can be manually selected according to the agent_uid of the service provider end to realize manual issuance of authentication information (step 402), and the corresponding issued list is obtained after issuance (404).

[0048] The service characteristics of the target enterprise user can be extracted from the enterprise behavior data of the target enterprise user. The enterprise behavior data of the target enterprise user includes industry information and service information in the enterprise-related information of the target enterprise user, and can also include historical records such as articles, posts, videos published by the target enterprise user in using the target application to publicize, promote, and market enterprise services, and social interaction information such as comments, likes, and shares with other terminal users on the platform of the target application, mainly the service-related social interaction information of the target enterprise user.

[0049] Specifically, the service feature of the target enterprise user is extracted based on the enterprise behavior data of the target enterprise user, including: extracting corresponding text features from the text data in the enterprise behavior data of the target enterprise user; extracting corresponding image features from the image data in the enterprise behavior data of the target enterprise user; extracting corresponding video features from the video data in the enterprise behavior data of the target enterprise user; determining the service feature of the target enterprise user by fusing the text features, the image features and the video features, the service feature including industry classification and service classification.

[0050] The service feature of the target enterprise user can be extracted based on multi-modal features, such as features representing text, image and video modalities. Based on the text data, image data and video features in the enterprise behavior data of the target enterprise user, text features, image features and video features can be extracted one by one. By fusing the above different types of extracted features, fusion features are obtained, and based on the fusion features, the service feature of the target enterprise user can be determined more accurately.

[0051] Then, the service feature of the target terminal user and the service feature of the target enterprise user are calculated respectively, so that the intersection of the service user feature of the target terminal user and the service feature of the target enterprise user can be found, which is the service matched by each target terminal user.

[0052] Specifically, by calculating the similarity between the service feature of each target terminal user and the service feature of the target enterprise user, the corresponding service of each target terminal user is matched, including: calculating the similarity between the service feature of the target terminal user and the service feature of the target enterprise user by Sim algorithm; based on the service feature of the target terminal user and the target enterprise user whose similarity is higher than the preset similarity threshold, the service matched by each target terminal user is determined.

[0053] Next, the extraction of the service feature of the target terminal user and the service feature of the target enterprise user and the similarity calculation are described in combination with specific formulas.

[0054] Let U represent the set of target terminal users, and V represent the set of enterprise behavior data of target enterprise users.

[0055] The feature of the target terminal user u ∈ U is f u .

[0056] The multi-modal feature f v of the enterprise behavior data v ∈ V of the target enterprise user includes text features , image features , and video features .

[0057] That is, Formula (1) Text features Extraction: Formula (2) Where T v is the text description in the enterprise behavior data v, and the text features are extracted by a pre-trained model (such as BERT).

[0058] Image features Extraction: Formula (3) Where I v is the image in the enterprise behavior data v, and the image features are extracted by a convolutional neural network (CNN).

[0059] Video features Extraction: Formula (4) Where V v is the video content in the enterprise behavior data v, and the video features are extracted by a 3D-CNN model.

[0060] The text features , image features , and video features are fused to obtain the fused features .

[0061] When matching the target end user and the target enterprise user, the Sim algorithm can be used: Formula (5) Where may be dot product, cosine similarity, or multilayer perceptron (MLP), and W1, W2, b1, b2, and sigma are coefficients.

[0062] According to formula (5), the similarity score between the service features of the target end user and the service features of the target enterprise user is calculated. If there are multiple service features, the similarity scores of the corresponding services can be obtained. The similarity score is compared with the preset similarity threshold, and the service feature with a similarity score higher than the preset similarity threshold is the service recommended by the target end user to the target enterprise user, so that the service category matched by the target end user can be obtained.

[0063] Further, according to the matched target enterprise user service, the target service information of the target enterprise user to which the target service belongs can be displayed to the target end user.

[0064] In the case where the target enterprise user has multiple services, the marketing content corresponding to different services is different. In this regard, the corresponding service information is selected according to the matched service, and is pushed to the matched target terminal user.

[0065] In one embodiment, the display of the service information of the target enterprise user to which the corresponding target service belongs to each target terminal user matched with the target service further comprises: adjusting the frequency of displaying the corresponding service information to each target terminal user matched with the target service according to the level of the target enterprise user to which the target service belongs, the level being related to the activity of the target enterprise user in the target application; or in the case where any target terminal user matched with the target service does not generate user behavior data within a preset time, displaying the service information of the target enterprise user to which the corresponding target service belongs to the target terminal user at a random time.

[0066] Optionally, the level of the enterprise user to which the service belongs is related to the activity of the enterprise user in the target application, for example, the level of the enterprise user who has been active for nearly one year is determined as S level, which is a high-quality customer, and the frequency of pushing the service information of the S level enterprise user to the matched target terminal user can be increased. The lower the level, the lower the frequency of pushing.

[0067] In addition, the user behavior data of the target terminal user corresponding to the target enterprise user can be queried to determine whether the corresponding target terminal user has not generated user behavior data or has generated very little user behavior data within a preset time, and the generation of user behavior data reflects the activity of the corresponding terminal user in the target application. If any target terminal user does not generate user behavior data within a preset time, the service information of the target enterprise user is randomly displayed to the target terminal user at a random time with a non-fixed frequency.

[0068] Thus, for part of the target terminal users with no user behavior or weak user behavior, the time is randomly issued by dynamically allocating the pushing or display proportion, which can realize the cultivation of the customer group of the private domain terminal user.

[0069] In the embodiments of the present application, the local target users corresponding to the target enterprises are obtained based on the location-based service (LBS) and the enterprise-related information; the geographic areas corresponding to the enterprise users accessing the target application are determined based on the enterprise-related information of the enterprise users; the geographic areas corresponding to the terminal users accessing the target application are determined based on the location-based service (LBS) of the terminal users; the target enterprise users and the target terminal users located in the same geographic area are obtained; and the matched target terminal users are recommended to each target enterprise user based on the user behavior data and the enterprise behavior data, so as to realize the accurate positioning of the terminal users (i.e., C-end users) and the enterprise users, present the localized content and services of the local enterprise users to the local C-end users, realize the personalized intelligent recommendation to the local C-end users, improve the effect of the fine service recommendation and delivery of the enterprises, reduce the recommendation cost and the customer maintenance cost, meet the needs of the personalized recommendation of the enterprise users, the social interaction of the C-end users, the marketing conversion of the private domain traffic, and the like, and both improve the experience of the enterprise users and solve the diversified needs of the local C-end users. The service information, activity notification or other targeted content of the local enterprise users can be positioned according to the location of the C-end users, and the interaction between the C-end users and the local enterprise users is enhanced.

[0070] In addition, the social functions such as comments, likes, sharing, scoring and the like can be integrated on the platform of the target application, the interaction between the terminal users and the local enterprise users and other terminal users is encouraged, and these behavior interactions are further fed back to the system to optimize the recommendation algorithm.

[0071] Optionally, the method further includes displaying the user information of the target terminal users matched with the target service to the target enterprise users to which the target service belongs.

[0072] In this embodiment, after the target enterprise users are recommended to the matched target terminal users, the user information of the matched target terminal users can be displayed to the target enterprise users, so that the target enterprise users can know the related information of the matched target terminal users, the performance of the enterprise users in the fine service recommendation and delivery is further improved, and the enterprise users can more effectively meet the needs of the terminal users through the optimization of the service pushing process and the operation efficiency.

[0073] Optionally, as shown in Figure 3 The embodiments of the present application also provide a user recommendation device 2000, which includes a processor 2400 and a memory 2200, the memory 2200 stores programs or instructions executable on the processor 2400, the programs or instructions are executed by the processor 2400 to realize each step of the above-mentioned user recommendation method embodiments and achieve the same technical effects, and details are not repeated here.

[0074] The embodiment of the present application further provides a readable storage medium, which stores programs or instructions, and the programs or instructions are executed by a processor to realize the processes of any one of the user recommendation method embodiments and achieve the same technical effects. To avoid repetition, details are not described herein. The readable storage medium includes a computer readable storage medium, such as a read-only memory (ROM), a random access memory (RAM), a magnetic disc or an optical disc, and the like.

[0075] The embodiment of the present application further provides a computer program product, which includes a non-transitory computer readable storage medium storing a computer program, and the computer program is operable to make a computer execute the processes of any one of the user recommendation method embodiments and achieve the same technical effects. To avoid repetition, details are not described herein.

[0076] It should be noted that in this document, the term “comprising” or “including” or any other variant thereof is intended to cover non-exclusive inclusion, so that processes, methods, articles or devices including a series of elements not only include those elements, but also include other elements not explicitly listed or inherent to such processes, methods, articles or devices. Without more limitations, the element defined by the statement “including a…” does not exclude the presence of another identical element in the process, method, article or device including the element.

[0077] From the above description of the embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be realized by means of software and necessary general hardware platforms, and of course, they can also be realized by hardware, but in many cases, the former is a better embodiment. Based on this understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as a ROM / RAM, a magnetic disc, an optical disc), and includes a plurality of instructions for making a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) execute the methods described in the embodiments of the present application.

[0078] The embodiments of the present application are described above in combination with the drawings, but the present application is not limited to the above-mentioned specific embodiments, and the above-mentioned specific embodiments are only illustrative, not restrictive. Those skilled in the art can make many forms under the inspiration of the present application without departing from the scope of the present application and the protection scope of the claims, and all of them belong to the protection scope of the present application.

Claims

1. A user recommendation method, characterized in that: include: Determine the geographical area corresponding to the enterprise user based on the enterprise-related information of the enterprise user accessing the target application; Determine the geographical area corresponding to the terminal user based on the location service (LBS) of the terminal user accessing the target application; Acquire target enterprise users and target terminal users located in the same geographical area; Based on user behavior data and enterprise behavior data, matching target end users are recommended for each target enterprise user.

2. The method according to claim 1, characterized in that Determining the geographical area corresponding to the enterprise user based on the enterprise-related information of the enterprise user accessing the target application includes: Based on the location-related data in the enterprise-related information, the geographical area corresponding to the enterprise user is determined, where the location-related data includes the business license registration place of the enterprise user, and the IP address and port where the enterprise user currently logs into the target application.

3. The method according to claim 1, characterized in that The method of recommending matching target end users to target enterprise users based on user behavior data and enterprise behavior data includes: Based on the collected compliant user behavior data, obtain the user profile of each target terminal user; Acquire enterprise behavior data of each target enterprise user, wherein the enterprise behavior data includes industry information, service information, historical publishing records of the target enterprise user in the target application, and social interaction information. The target enterprise user includes at least one service; Match corresponding services among target enterprise users for each target terminal user based on the user profile of each target terminal user and the enterprise behavior data of each target enterprise user; The service information of the target enterprise user to which the corresponding target service belongs is displayed to each target terminal user matched with the target service.

4. The method according to claim 3, characterized in that Also includes: The user information of each target terminal user matching the target service is displayed to the target enterprise user to which the target service belongs.

5. The method according to claim 3, characterized in that The user behavior data includes the target terminal user's historical browsing records, consumption information, and social interaction information on the target application; The matching of corresponding services among target enterprise users for each target terminal user based on the user profile of each target terminal user and the enterprise behavior data of each target enterprise user includes: Extracting service features of corresponding target terminal users based on the user profile of each target terminal user; Extract the service characteristics of each target enterprise user based on the enterprise behavior data of each target enterprise user; By calculating the similarity between the service characteristics of each target terminal user and the service characteristics of each target enterprise user, the corresponding service is matched for each target terminal user.

6. The method according to claim 5, characterized in that The extracting the service features of the corresponding target terminal user based on the user profile of each target terminal user includes: For any target terminal user, extract text data from the user portrait of the target terminal user; By segmenting the text data, corresponding word vectors are obtained; Comparing each word vector with a dedicated dictionary to determine a word vector that matches a dedicated word in the dedicated dictionary, wherein the dedicated dictionary is pre-built based on enterprise-related information and includes industry classification and service classification; The matched word vector is determined as the service feature of the target terminal user.

7. The method according to claim 3, wherein The displaying of service information of the target enterprise user to which the corresponding target service belongs to each target terminal user matched with the target service also includes: Adjusting the frequency of displaying corresponding service information to each target terminal user matching the target service according to the level of the target enterprise user to which the target service belongs, wherein the level is related to the activity of the target enterprise user in the target application; or In the case that any target terminal user matched with the target service does not generate user behavior data within a preset time, service information of a target enterprise user to which the corresponding target service belongs is randomly displayed to the target terminal user.

8. The method according to claim 6, characterized in that Before determining the geographical area of ​​the enterprise user based on the enterprise-related information of the enterprise user accessing the target application, the method further includes: Receive authentication information of the target enterprise user, the authentication information including the business license registration location of the target enterprise user, the industry information and service information of the target enterprise user; Send the authentication information to the corresponding service provider via HTTPS protocol for authentication; Upon receiving the authenticated enterprise-related information fed back by the service provider, the corresponding industry classification and service classification are obtained by respectively marking the industry information and service information in the enterprise-related information; The industry classification and service classification of the target enterprise users are added to the dedicated dictionary.

9. A user recommendation device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, the steps of the method according to any one of claims 1 to 8 are implemented.

10. A readable storage medium, characterized in that: The readable storage medium stores a program or instruction, and when the program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images