A task recommendation method, device, storage medium and terminal based on user portrait

Abstract

The present invention discloses a method, device, storage medium and terminal for recommending tasks based on user portraits. The method includes: collecting and preprocessing truck data and user data; associating the preprocessed truck data and user data to generate a pulled-through data set; wherein the pulled-through data set includes multiple successfully associated truck data and user data; setting a label theme set of the user portrait according to a target application scenario; mapping multiple successfully associated truck data and user data one by one to each label theme in the label theme set of the user portrait to generate a label theme set of associated data; filtering the label theme set of the associated data according to a preset filtering rule to generate a target user group; and recommending at least one task corresponding to the target application scenario to the target user group. Therefore, by adopting the embodiments of the present application, the dimensions and quality of constructing user portraits can be enriched, and the accuracy of service recommendation for users can be further improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a task recommendation method, device, storage medium and terminal based on user portrait. Background Art

[0002] With the development of information technology, user portraits and recommendation systems are often used in various industries. They can be labeled according to user characteristics and then build user portraits. When user portraits that meet certain goals are generated, some label data is used for recommendation, association analysis and collaborative filtering.

[0003] At present, user profiling and recommendation technology has become increasingly mature, and the industries to which it is applicable has gradually expanded. Applying user profiling and recommendation technology to the emerging smart freight industry is becoming a popular research direction.

[0004] Currently, the initial network access information and daily operations of trucks across the country will generate a large amount of operation, parking, consumption and other information. The APP used by truck users will also generate a large amount of basic user information and user behavior, consumption and other information. However, the current user portrait system fails to effectively integrate the information generated by the two different entities of trucks and users, resulting in the isolation of information of two strongly related entities, making it impossible to analyze and think about problems from a higher-level perspective of freight participants, and to analyze and mine more potential, multi-dimensional, and more accurately describe the relationship between truck entities and user entities. Labels, and thus unable to effectively, completely and accurately portray the user portrait of the relevant user groups, resulting in precision marketing and recommendations for relevant user groups. It is difficult to implement and the effect is poor.

[0005] Therefore, how to find an effective method to achieve precise marketing and recommendations for relevant user groups is an urgent problem to be solved. Summary of the invention

[0006] The embodiments of the present application provide a method, device, storage medium and terminal for recommending tasks based on user portraits. In order to have a basic understanding of some aspects of the disclosed embodiments, a simple summary is given below. This summary is not a general review, nor is it intended to identify key / important components or describe the scope of protection of these embodiments. Its sole purpose is to present some concepts in a simple form as a preface to the detailed description that follows.

[0007] In a first aspect, an embodiment of the present application provides a task recommendation method based on a user profile, the method comprising:

[0008] Collect and pre-process truck data and user data;

[0009] Associating the pre-processed truck data with the user data to generate a connected data set; wherein the connected data set includes a plurality of successfully associated truck data and user data;

[0010] Set the tag theme set of the user portrait according to the target application scenario;

[0011] Map multiple successfully associated truck data and user data one by one to each label topic in the label topic set of the user portrait, and generate a label topic set of associated data;

[0012] Filter the tag topic set of the associated data according to the preset filtering rules to generate the target user group;

[0013] Recommend at least one task corresponding to the target application scenario to the target user group.

[0014] Optionally, after recommending at least one task corresponding to the target application scenario to the target user group, the method further includes:

[0015] Analyze operation information of a user group for at least one task and generate analysis results;

[0016] Based on the analysis results, the tag topic set of the user portrait is updated or deleted to generate the tag topic set of the updated or deleted user portrait;

[0017] Mapping multiple successfully associated truck data and user data one by one to each label topic in the label topic set of the updated or deleted user portrait, generating a label topic set of associated data;

[0018] and / or updating or deleting the user's processing information for the task in the task recommendation module based on the analysis result to generate updated or deleted user processing information;

[0019] Mapping the plurality of successfully associated truck data and user data one by one to the updated or deleted user processing information;

[0020] Continue to perform the step of filtering the tag subject set of the associated data and / or the user processing information according to the preset filtering rule.

[0021] Optionally, the tag topic set of the associated data is filtered according to a preset filtering rule to generate a target user group, including:

[0022] Initialize the preset label theme parameter value;

[0023] Get the parameter value of each tag topic in the tag topic set of the associated data;

[0024] Compare each tag topic parameter value in the tag topic set of the associated data with a preset tag topic parameter value to determine whether there is at least one tag with a value greater than the preset tag topic parameter value;

[0025] If so, obtain at least one user corresponding to a tag greater than a preset tag topic parameter value to generate a target user group.

[0026] Optionally, the tag topic set of the associated data is filtered according to a preset filtering rule to generate a target user group, including:

[0027] Initialize multiple preset label theme parameter values;

[0028] Get the parameter value of each tag topic in the tag topic set of the associated data;

[0029] Compare each tag theme parameter value in the tag theme set of the associated data with a plurality of preset tag theme parameter values ​​to determine whether there are multiple tags with values ​​greater than the preset tag theme parameter values;

[0030] If so, obtain the users corresponding to multiple tags greater than the preset tag topic parameter value to generate a target user group.

[0031] Optionally, the tag topic set of the associated data is filtered according to a preset filtering rule to generate a target user group, including:

[0032] Divide multiple user groups into different types based on the business recommendation requirements, experience, and data of the target application scenarios;

[0033] Each tag data in the tag topic set of the associated data is filtered as a feature to generate multiple types of tag topics;

[0034] Preprocess various types of label topics, and convert the preprocessed various types of label topics into vectors to generate user sets;

[0035] Use Kmeans clustering algorithm to perform clustering calculation on user sets and generate clustered user groups;

[0036] Compare and analyze the clustered user group with the multiple user group types that have been divided, and output the analysis value;

[0037] When the analysis value is greater than a preset threshold, a target user group is generated.

[0038] Optionally, when the analysis value is greater than a preset threshold, a target user group is generated, including:

[0039] When the analysis value is less than the preset threshold, the algorithm parameters are adjusted;

[0040] The step of comparing and analyzing the clustered user group with the divided multiple user group types is continued based on the adjusted algorithm until the analysis value is greater than a preset threshold, thereby generating a target user group.

[0041] Optionally, collect and pre-process truck data and user data, including:

[0042] Collecting truck data and user data; wherein the truck data at least includes basic information data of the truck and operation information data of the truck, and the user data at least includes basic information data of the user and data generated by the user on the APP;

[0043] Clean the basic information data of the truck, the operation information data of the truck, the basic information data of the user, and the data generated by the user on the APP to generate pre-processed truck data and user data;

[0044] Among them, data cleaning includes at least removing redundant data, normalizing data formats, and unifying data standards.

[0045] In a second aspect, an embodiment of the present application provides a task recommendation device based on a user portrait, the device comprising:

[0046] Data processing module, used to collect and pre-process truck data and user data;

[0047] A data association module is used to associate the pre-processed truck data with the user data to generate a connected data set; wherein the connected data set includes a plurality of successfully associated truck data and user data;

[0048] The tag theme setting module is used to set the tag theme set of the user portrait according to the target application scenario;

[0049] A data mapping module is used to map multiple successfully associated truck data and user data one by one to each label topic in the label topic set of the user portrait, and generate a label topic set of associated data;

[0050] A user group generation module is used to filter the tag topic set of the associated data according to the preset filtering rules to generate the target user group;

[0051] The task recommendation module is used to recommend at least one task corresponding to the target application scenario to the target user group.

[0052] In a third aspect, an embodiment of the present application provides a computer storage medium, which stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the above-mentioned method steps.

[0053] In a fourth aspect, an embodiment of the present application provides a terminal, which may include: a processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the above-mentioned method steps.

[0054] The technical solution provided by the embodiments of the present application may have the following beneficial effects:

[0055] In an embodiment of the present application, a task recommendation device based on user portraits first collects and preprocesses truck data and user data, then associates the preprocessed truck data and user data to generate a pulled-together data set; wherein the pulled-together data set includes multiple successfully associated truck data and user data; then, a label topic set of the user portrait is set according to the target application scenario, and then multiple successfully associated truck data and user data are mapped one by one to each label topic in the label topic set of the user portrait to generate a label topic set of associated data, and then the label topic set of the associated data is filtered according to the preset filtering rules to generate a target user group, and finally, at least one task corresponding to the target application scenario is recommended to the target user group. Since the present application effectively integrates the data generated by trucks and users, and conducts joint analysis and mining based on the integrated information, the dimension and quality of constructing user portraits are enriched, and the accuracy of service recommendations for users is further improved.

[0056] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0058] Figure 1 It is a flowchart of a task recommendation method based on user portrait provided in an embodiment of the present application;

[0059] Figure 2 This is a schematic diagram of a process of task recommendation based on user portrait provided by an embodiment of the present application;

[0060] Figure 3 It is a flowchart of another method for recommending tasks based on user portraits provided in an embodiment of the present application;

[0061] Figure 4 is a schematic diagram of a task recommendation device based on user portrait provided in an embodiment of the present application;

[0062] Figure 5 is a schematic diagram of the structure of a terminal provided in an embodiment of the present application;

[0063] Figure 6 It is a schematic diagram of a process of optimizing a single / multiple preset tag topic parameters w provided in an embodiment of the present application; Figure 7 It is a schematic diagram of the process of optimizing parameter k in a Kmeans clustering algorithm provided in an embodiment of the present application. DETAILED DESCRIPTION

[0064] The following description and the drawings sufficiently illustrate specific embodiments of the invention to enable those skilled in the art to practice them.

[0065] It should be clear that the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are only examples of devices and methods consistent with some aspects of the present invention as detailed in the attached claims.

[0067] In the description of the present invention, it should be understood that the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. In addition, in the description of the present invention, unless otherwise specified, "plurality" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the previously associated objects are in an "or" relationship.

[0068] In the technical solution provided in the present application, the data generated by trucks and users are effectively integrated, and joint analysis and mining are performed based on the integrated information, thereby enriching the dimensions and quality of constructing user portraits, and further improving the accuracy of service recommendations for users. An exemplary embodiment is used below to explain in detail.

[0069] The following will be combined with the attached Figure 1 -Attached Figure 3, the task recommendation method based on user portrait provided in the embodiment of the present application is introduced in detail. The method can be implemented by a computer program and can be run on a task recommendation device based on user portrait based on the von Neumann system. The computer program can be integrated in an application or run as an independent tool application. Among them, the task recommendation device based on user portrait in the embodiment of the present application can be a user terminal, including but not limited to: a personal computer, a tablet computer, a handheld device, a vehicle-mounted device, a wearable device, a computing device or other processing device connected to a wireless modem. In different networks, the user terminal can be called by different names, such as: user equipment, access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device, cellular phone, cordless phone, personal digital assistant (PDA), 5G network or terminal equipment in future evolution network, etc.

[0070] See also Figure 1 , which is a flowchart of a task recommendation method based on user portraits provided in an embodiment of the present application. Figure 1 As shown, the method of the embodiment of the present application may include the following steps:

[0071] S101, collecting and preprocessing truck data and user data;

[0072] Among them, truck data mainly includes basic information data of trucks and operation information data of trucks, and user data mainly includes basic information data of users and data generated by users on the APP. The initial network access information and daily operation of national trucks will generate a large amount of operation, parking, consumption and other information, which will be saved in the truck data center. User data will be saved in the database of the user's APP.

[0073] Usually, after data collection is completed, the data is preprocessed and cleaned, including removing useless redundant data, standardizing the data format, unifying the data standards, etc., and then the data is saved.

[0074] In one possible implementation, when preprocessing data, truck data and user data are first collected, wherein the truck data includes at least the basic information data of the truck and the operation information data of the truck, and the user data includes at least the basic information data of the user and the data generated by the user on the APP. Finally, the basic information data of the truck, the operation information data of the truck, the basic information data of the user and the data generated by the user on the APP are cleaned to generate preprocessed truck data and user data; wherein data cleaning includes at least removing redundant data, normalizing data formats, and unifying data standards.

[0075] S102, associating the pre-processed truck data with the user data to generate a unified data set;

[0076] The integrated data set includes multiple successfully associated truck data and user data.

[0077] Usually, each user in the user data may correspond to one truck data, or may correspond to multiple truck data. Here, the truck data needs to be associated with the corresponding user data one by one, so that the truck data and the user data are connected.

[0078] In a possible implementation, when the truck data is connected with the user data, the basic information data of the truck is first classified from the truck data, and then the unique identification of each truck is located and identified from the basic information data, and the unique identification of the truck driver corresponding to each truck is obtained according to the unique identification of each truck, and a truck driver identification set is generated. Secondly, the basic information data of each user is extracted from the user data, and then the unique identification of each user is obtained from the basic information data of each user. A search algorithm is used to search for a truck driver identification that is the same as the unique identification of each user in the truck driver identification set, so as to obtain one or more truck data corresponding to each user identification, and finally each user is associated with one or more truck data corresponding to each user, and a connected data set is generated.

[0079] Furthermore, when using a search algorithm to search for a truck driver ID that is identical to each user's unique ID in a truck driver ID set, a search tree is first constructed based on the truck driver ID set, and each node on the search tree represents a truck driver ID. Using each user's unique ID as a parameter, a depth-first search algorithm is used to traverse the nodes of the tree along the depth of the tree, and the branches of the tree are searched as deeply as possible, thereby traversing and searching for multiple truck data corresponding to each user.

[0080] S103, setting a tag theme set of the user portrait according to the target application scenario;

[0081] The target application scenario is a scenario pre-set by the user according to business needs. For example, the scenario can be a task scenario of recommending a gas station to a user, or an application scenario of recommending a source of goods to a user. The label theme of the user portrait is a label that can describe the characteristics of each user.

[0082] Typically, the label topics of a user's portrait include: user basic information topic, user behavior information topic, user social information topic, user-associated vehicle basic information, user-associated vehicle operation information, user-associated vehicle parking information topic, user-associated vehicle safety information topic, and user-associated vehicle consumption information topic, etc.

[0083] Specifically, the main topics of user basic information include: user gender, age, active login location, registration time, installation channel, whether authenticated, user identity, etc., and the relevant tags are mainly obtained through statistics and machine learning; the main topics of user behavior information include: number of visits in the past xx days, number of active days in the past xx days, visit duration in the past xx days, date of the most recent visit, average visit depth, whether xx function is used in the past xx days, number of times xx function is used in the past xx days, etc., and the relevant tags are mainly obtained through statistics; user social information mainly includes: number of friends, number of likes in the past xx days, number of dynamic posts in the past xx days, number of comments in the past xx days, date of the last dynamic post, etc., and the relevant tags are mainly obtained through statistics; the basic information of user-associated vehicles includes: associated vehicle brand, associated vehicle model, associated vehicle age, associated vehicle fuel type, associated vehicle tire specifications, associated vehicle rated load capacity, associated vehicle purchase method, associated vehicle operation method, associated vehicle network access time, etc., and the relevant tags are mainly obtained through statistics; user-associated vehicle operation information mainly includes: associated vehicle The vehicle trajectory completeness rate, the number of days the associated vehicle has been online in the past xx days, the last time the associated vehicle was online, and the corresponding associated vehicle operating mileage, associated vehicle operating time, etc. are obtained through statistics of one or more dimensions such as time dimension: hour, day, day, month, season, year, and geographical dimension: province, city, district (county), and road type dimension: highway, national highway, provincial highway, county road, etc.; the user-associated vehicle stop information mainly includes: the number of times the associated vehicle has stopped at the logistics park in the past xx days, the number of times the associated vehicle has stopped at the gas station in the past xx days, the associated vehicle often stops at the logistics park, the associated vehicle often stops at the gas station, etc., and the relevant tags are mainly obtained through statistics; the user-associated vehicle safety information mainly includes: the fatigue driving time of the associated vehicle in the past xx days, the number of fatigue driving times of the associated vehicle in the past xx days, the speeding time of the associated vehicle in the past xx days, the number of speeding times of the associated vehicle in the past xx days, the average number of highway stops of the associated vehicle, etc., and the relevant tags are mainly obtained through statistics; the user-associated vehicle consumption information mainly includes: the consumption amount of the associated vehicle in the past xx days, the number of consumption times of the associated vehicle in the past xx days, etc., and the relevant tags are mainly obtained through statistics.

[0084] S104, mapping the multiple successfully associated truck data and user data one by one to each label topic in the label topic set of the user portrait, to generate a label topic set of associated data;

[0085] Among them, the label topic can be understood as multiple field names in the data dictionary, and the entity data corresponding to each field name is the successfully associated truck data and user data.

[0086] Usually, the user profiling program is executed through scheduling, and the tasks and data are monitored, and the generated data is saved in the data warehouse.

[0087] In one possible implementation, when user data and truck data are connected and a label theme set of the user portrait is set according to the scenario, the connected user data and truck data are classified one by one into labels consistent with their label theme types through data statistical analysis, and finally a label theme set of associated data is generated. An example of the label theme set of associated data is shown in Table 1.

[0088] Table 1

[0089]

[0090]

[0091] S105, filtering the tag topic set of the associated data according to a preset filtering rule to generate a target user group;

[0092] In one possible implementation, when generating a target user group, first initialize a preset label topic parameter value, then obtain each label topic parameter value in the label topic set of the associated data, and then compare each label topic parameter value in the label topic set of the associated data with the preset label topic parameter value to determine whether there is at least one label greater than the preset label topic parameter value. If so, finally obtain at least one user corresponding to a label greater than the preset label topic parameter w value to generate a target user group.

[0093] In another possible implementation, when generating a target user group, first initialize multiple preset label topic parameter values, then obtain each label topic parameter value in the label topic set of the associated data, and then compare each label topic parameter value in the label topic set of the associated data with multiple preset label topic parameter values ​​to determine whether there are multiple labels greater than the preset label topic parameter value. If so, finally obtain the users corresponding to the multiple labels greater than the preset label topic parameter w value to generate the target user group.

[0094] For example Figure 6 As shown, the optimization method of the single / multiple preset label topic parameter w value is as follows:

[0095] First, through the U w 'Then the personalized recommendation is performed for each user to obtain the recommendation result, and then the w value is evaluated based on the recommendation result to see whether it is necessary to optimize the w value. If not, the process ends. Otherwise, the w value is updated with the w' value, and the label is reset by the w' value.

[0096] In another possible implementation, when generating a target user group, first divide a plurality of user group types according to the business recommendation requirements, experience and data of the target application scenario, then screen each label data in the label topic set of the associated data as a feature to generate a plurality of types of label topics, then preprocess the plurality of types of label topics, and perform vector conversion on the preprocessed plurality of types of label topics to generate a user set, then use the Kmeans clustering algorithm to perform clustering calculation on the user set to generate a clustered user group, then compare and analyze the clustered user group with the plurality of divided user group types, output the analysis value, and finally generate the target user group when the analysis value is greater than a preset threshold.

[0097] Further, when the analysis value is less than the preset threshold, the parameter k of the Kmeans clustering algorithm is adjusted, and the step of comparing and analyzing the clustered user group with the multiple user group types divided is continued based on the adjusted algorithm until the analysis value is greater than the preset threshold, and the target user group is generated. Among them, the parameter k is the number of user groups, which refers to the number of centroids in the Kmeans clustering algorithm.

[0098] Specifically, the method of performing clustering calculation on the user set by the Kmeans clustering algorithm is as follows:

[0099] The first step is to select k representative user vectors: U1, U2, ...U k ;

[0100] The second step is to use the Kmeans algorithm to perform parameter k clustering. The clustering results are: U1', U2', ...U k ';

[0101] The third step is to make Similarity(U1,U1')

[0102] Similarity(U2,U2') <h2;

[0103] …

[0104] Similarity(U k ,U k ') <h k ;

[0105] Step 4: Determine U1', U2', ...U k 'User groups.

[0106] Specifically, when generating the final user group, there are recommendation methods for single-label or multi-label combination filtering and recommendation methods for clustering users with the same identity.

[0107] ​In the recommendation method of single-label or multi-label combination filtering, single-label recommendation refers to recommending to the user group filtered out by a certain label in the portrait that meets a certain threshold, and multi-label combination recommendation refers to recommending to the user group filtered out by multiple labels in the portrait that meet the threshold set for each label. Both the single-label threshold and the multi-label threshold are configurable and are set based on business experience or combined with the conclusions of data analysis, and are continuously optimized based on the evaluation of the recommendation effect.

[0108] In the clustering recommendation method for users with the same identity, users are first divided into several representative groups based on business recommendation requirements, experience and data analysis.

[0109] Secondly, the user portrait label data is used as a feature for screening, and labels with serious missing values ​​and no relevance to the clustering are deleted. Then each label is processed accordingly according to classification coding, sequence coding, continuous feature binning, etc., and converted into a vector representation, as follows:

[0110] User 1 = [0,0,1,0,0,1,2,...,1,3,0,...]

[0111] User 2 = [0,1,0,0,0,0,3,...,0,2,0,...]

[0112] User 3 = [0,0,1,0,0,1,1,...,0,1,0,...]

[0113] …

[0114] Finally, the Kmeans clustering algorithm is used to cluster the users. The user groups calculated by clustering are compared and analyzed with the representative user groups preset at the beginning, and the algorithm is adjusted and optimized until the similarity between the two reaches the threshold h. The corresponding user groups are obtained, and corresponding recommendations are made for each user group. Then, the whole process is continuously adjusted and optimized based on the evaluation of the recommendation effect. Figure 7 As shown in the figure, the optimization method of the parameter k value in the Kmeans clustering algorithm is as follows:

[0115] First, through U1', U2', ...U k 'Users are given personalized recommendations to obtain recommendation results, and then the k value needs to be optimized based on the recommendation results. If not, the process ends. Otherwise, the k value is updated with the k' value, and the clustering process is re-executed using the k' value.

[0116] S106: Recommend at least one task corresponding to the target application scenario to the target user group.

[0117] The task is a service item to be recommended according to the target application scenario. For example, when the target scenario is a cargo recommendation scenario, the task here can be a specific location where cargo can be transported and cargo information.

[0118] In an embodiment of the present application, after recommending at least one task corresponding to the target application scenario to the target user group, it is first necessary to analyze the operation information of the user group for at least one task and generate an analysis result; then, based on the analysis result, the label topic set of the user portrait is updated or deleted to generate an updated or deleted label topic set of the user portrait, and multiple successfully associated truck data and user data are mapped one by one to each label topic in the label topic set of the updated or deleted user portrait to generate a label topic set of associated data, and / or based on the analysis result, the user processing information for the task in the task recommendation module is updated or deleted to generate updated or deleted user processing information, and the multiple successfully associated truck data and user data are mapped one by one to the updated or deleted user processing information; finally, the step of filtering the label topic set of the associated data and / or the user processing information according to the preset filtering rules is continued to realize the closed loop of the solution so as to continuously optimize the label topics.

[0119] For example Figure 2 As shown, Figure 2 This is a flowchart of the task recommendation process based on user portrait provided by the present application. First, user APP data and truck big data are obtained, and then data preprocessing is performed on the user APP data and truck big data. Then the preprocessed user APP data and truck big data are connected, and the user portrait is characterized based on the connected data. Then, a specified group of user groups that meet the application scenario are found through the task recommendation module to recommend corresponding tasks, and then the analysis results are generated after analyzing the user's specific operations on the task. Finally, one of the three methods is selected to evaluate the analysis results: the first is a method of updating or deleting user portrait labels, the second is a method of updating or deleting user processing information in the task recommendation module, and the third is a method of updating or deleting user portrait labels and updating or deleting user processing information in the task recommendation module.

[0120] For example, in the target scenario, we first obtain the user app data and the truck data for preprocessing, then associate the data through the authentication information of the truck and the user, and build the following user portrait through label predefinition and statistical analysis and mining. Secondly, according to the business scenario requirements and data analysis, we use multi-label combinations to filter the user portrait labels. The number of active days in the past 30 days is greater than d, the amount of fuel consumption of the associated vehicle in the past 30 days is greater than m, and the user ID is the driver. The filtered results are provided to the business for corresponding recommendation services. Finally, the threshold is optimized according to the evaluation of the business recommendation effect.

[0121] In an embodiment of the present application, a task recommendation device based on user portraits first collects and preprocesses truck data and user data, then associates the preprocessed truck data and user data to generate a pulled-together data set; wherein the pulled-together data set includes multiple successfully associated truck data and user data; then, a label topic set of the user portrait is set according to the target application scenario, and then multiple successfully associated truck data and user data are mapped one by one to each label topic in the label topic set of the user portrait to generate a label topic set of associated data, and then the label topic set of the associated data is filtered according to the preset filtering rules to generate a target user group, and finally, at least one task corresponding to the target application scenario is recommended to the target user group. Since the present application effectively integrates the data generated by trucks and users, and conducts joint analysis and mining based on the integrated information, the dimension and quality of constructing user portraits are enriched, and the accuracy of service recommendations for users is further improved.

[0122] See also Figure 3 , provides a flowchart of another method for recommending tasks based on user portraits for the embodiment of the present application. Figure 3 As shown, the method of the embodiment of the present application may include the following steps:

[0123] The first step is to collect and preprocess truck data and user data;

[0124] The second step is to associate the pre-processed truck data with the user data to generate a connected data set; wherein the connected data set includes a plurality of successfully associated truck data and user data;

[0125] The third step is to set the tag theme set of the user portrait according to the target application scenario;

[0126] The fourth step is to map the multiple successfully associated truck data and user data to each label topic in the label topic set of the user portrait one by one, and generate a label topic set of associated data;

[0127] Step 5: Filter the tag topic set of the associated data according to the preset filtering rules to generate the target user group;

[0128] Step 6: Recommend at least one task corresponding to the target application scenario to the target user group;

[0129] Step 7: Analyze the operation information of the user group for at least one task and generate analysis results;

[0130] Step 8: Based on the analysis result, the label topic set of the user portrait is updated or deleted to generate the label topic set of the updated or deleted user portrait, and then the multiple successfully associated truck data and user data are mapped one by one to each label topic in the label topic set of the updated or deleted user portrait to generate a label topic set of associated data;

[0131] and / or based on the analysis result, updating or deleting the user processing information for the task in the task recommendation module, generating updated or deleted user processing information, and then mapping the multiple successfully associated truck data and user data one by one to the updated or deleted user processing information;

[0132] In the ninth step, continue to perform the step of filtering the tag topic set of the associated data and / or the user processing information according to the preset filtering rules to continuously cycle and optimize the recommendation system.

[0133] In an embodiment of the present application, a task recommendation device based on user portraits first collects and preprocesses truck data and user data, then associates the preprocessed truck data and user data to generate a pulled-together data set; wherein the pulled-together data set includes multiple successfully associated truck data and user data; then, a label topic set of the user portrait is set according to the target application scenario, and then multiple successfully associated truck data and user data are mapped one by one to each label topic in the label topic set of the user portrait to generate a label topic set of associated data, and then the label topic set of the associated data is filtered according to the preset filtering rules to generate a target user group, and finally, at least one task corresponding to the target application scenario is recommended to the target user group. Since the present application effectively integrates the data generated by trucks and users, and conducts joint analysis and mining based on the integrated information, the dimension and quality of constructing user portraits are enriched, and the accuracy of service recommendations for users is further improved.

[0134] The following are embodiments of the device of the present invention, which can be used to implement the method embodiments of the present invention. For details not disclosed in the device embodiments of the present invention, please refer to the method embodiments of the present invention.

[0135] See also Figure 4 , which shows a schematic diagram of the structure of a task recommendation device based on user portrait provided by an exemplary embodiment of the present invention. The task recommendation device based on user portrait can be implemented as all or part of a terminal through software, hardware or a combination of both. The device 1 includes a data processing module 10, a data association module 20, a label theme setting module 30, a data mapping module 40, a user group generation module 50, and a task recommendation module 60.

[0136] A data processing module 10, for collecting and preprocessing truck data and user data;

[0137] The data association module 20 is used to associate the pre-processed truck data with the user data to generate a connected data set; wherein the connected data set includes a plurality of successfully associated truck data and user data;

[0138] The label theme setting module 30 is used to set the label theme set of the user portrait according to the target application scenario;

[0139] The data mapping module 40 is used to map the multiple truck data and user data that are successfully associated one by one to each label topic in the label topic set of the user portrait, and generate a label topic set of associated data;

[0140] A user group generation module 50 is used to filter the tag topic set of the associated data according to a preset filtering rule to generate a target user group;

[0141] The task recommendation module 60 is used to recommend at least one task corresponding to the target application scenario to the target user group.

[0142] It should be noted that the task recommendation device based on user portrait provided in the above embodiment only uses the calculation of the above functional modules as an example when executing the task recommendation method based on user portrait. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is calculated into different functional modules to complete all or part of the functions described above. In addition, the task recommendation device based on user portrait provided in the above embodiment and the task recommendation method based on user portrait belong to the same concept. The implementation process is detailed in the method embodiment and will not be repeated here.

[0143] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0144] In an embodiment of the present application, a task recommendation device based on user portraits first collects and preprocesses truck data and user data, then associates the preprocessed truck data and user data to generate a pulled-together data set; wherein the pulled-together data set includes multiple successfully associated truck data and user data; then, a label topic set of the user portrait is set according to the target application scenario, and then multiple successfully associated truck data and user data are mapped one by one to each label topic in the label topic set of the user portrait to generate a label topic set of associated data, and then the label topic set of the associated data is filtered according to the preset filtering rules to generate a target user group, and finally, at least one task corresponding to the target application scenario is recommended to the target user group. Since the present application effectively integrates the data generated by trucks and users, and conducts joint analysis and mining based on the integrated information, the dimension and quality of constructing user portraits are enriched, and the accuracy of service recommendations for users is further improved.

[0145] The present invention also provides a computer-readable medium having program instructions stored thereon, which, when executed by a processor, implement the task recommendation method based on user profile provided in the above-mentioned various method embodiments. The present invention also provides a computer program product containing instructions, which, when executed on a computer, enables the computer to execute the task recommendation method based on user profile provided in the above-mentioned various method embodiments.

[0146] See also Figure 5 , which is a schematic diagram of the structure of a terminal provided in an embodiment of the present application. Figure 5 As shown, the terminal 1000 may include: at least one processor 1001 , at least one network interface 1004 , a user interface 1003 , a memory 1005 , and at least one communication bus 1002 .

[0147] The communication bus 1002 is used to realize the connection and communication between these components.

[0148] The user interface 1003 may include a display screen (Display) and a camera (Camera), and the optional user interface 1003 may also include a standard wired interface and a wireless interface.

[0149] The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0150] Among them, the processor 1001 may include one or more processing cores. The processor 1001 uses various interfaces and lines to connect various parts of the entire electronic device 1000, and executes various functions and processes data of the electronic device 1000 by running or executing instructions, programs, code sets or instruction sets stored in the memory 1005, and calling data stored in the memory 1005. Optionally, the processor 1001 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 1001 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 1001, and it can be implemented separately through a chip.

[0151] Among them, the memory 1005 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 1005 includes a non-transitory computer-readable storage medium. The memory 1005 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 1005 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 1005 may optionally be at least one storage device located away from the aforementioned processor 1001. As Figure 5 As shown, the memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a task recommendation application based on a user portrait.

[0152] exist Figure 5In the terminal 1000 shown, the user interface 1003 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 1001 can be used to call the task recommendation application based on the user portrait stored in the memory 1005, and specifically perform the following operations:

[0153] Collect and pre-process truck data and user data;

[0154] Associating the pre-processed truck data with the user data to generate a connected data set; wherein the connected data set includes a plurality of successfully associated truck data and user data;

[0155] Set the tag theme set of the user portrait according to the target application scenario;

[0156] Map multiple successfully associated truck data and user data one by one to each label topic in the label topic set of the user portrait, and generate a label topic set of associated data;

[0157] Filter the tag topic set of the associated data according to the preset filtering rules to generate the target user group;

[0158] Recommend at least one task corresponding to the target application scenario to the target user group.

[0159] In one embodiment, after recommending at least one task corresponding to the target application scenario to the target user group, the processor 1001 further performs the following operations:

[0160] Analyze operation information of a user group for at least one task and generate analysis results;

[0161] Based on the analysis results, the tag topic set of the user portrait is updated or deleted to generate the tag topic set of the updated or deleted user portrait;

[0162] Mapping multiple successfully associated truck data and user data one by one to each label topic in the label topic set of the updated or deleted user portrait, generating a label topic set of associated data;

[0163] and / or updating or deleting the user's processing information for the task in the task recommendation module based on the analysis result to generate updated or deleted user processing information;

[0164] Mapping the plurality of successfully associated truck data and user data one by one to the updated or deleted user processing information;

[0165] Continue to perform the step of filtering the tag subject set of the associated data and / or the user processing information according to the preset filtering rule.

[0166] In one embodiment, when the processor 1001 filters the tag topic set of the associated data according to the preset filtering rule to generate the target user group, the processor 1001 specifically performs the following operations:

[0167] Initialize the preset label theme parameter value;

[0168] Get the parameter value of each tag topic in the tag topic set of the associated data;

[0169] Compare each tag topic parameter value in the tag topic set of the associated data with a preset tag topic parameter value to determine whether there is at least one tag with a value greater than the preset tag topic parameter value;

[0170] If so, obtain at least one user corresponding to a tag greater than a preset tag topic parameter value to generate a target user group.

[0171] In one embodiment, when the processor 1001 filters the tag topic set of the associated data according to the preset filtering rule to generate the target user group, the processor 1001 specifically performs the following operations:

[0172] Initialize multiple preset label theme parameter values;

[0173] Get the parameter value of each tag topic in the tag topic set of the associated data;

[0174] Compare each tag theme parameter value in the tag theme set of the associated data with a plurality of preset tag theme parameter values ​​to determine whether there are multiple tags with values ​​greater than the preset tag theme parameter values;

[0175] If so, obtain the users corresponding to multiple tags greater than the preset tag topic parameter value to generate a target user group.

[0176] In one embodiment, when the processor 1001 filters the tag topic set of the associated data according to the preset filtering rule to generate the target user group, the processor 1001 specifically performs the following operations:

[0177] Divide multiple user groups into different types based on the business recommendation requirements, experience, and data of the target application scenarios;

[0178] Each tag data in the tag topic set of the associated data is filtered as a feature to generate multiple types of tag topics;

[0179] Preprocess various types of label topics, and convert the preprocessed various types of label topics into vectors to generate user sets;

[0180] Use Kmeans clustering algorithm to perform clustering calculation on user sets and generate clustered user groups;

[0181] Compare and analyze the clustered user group with the multiple user group types that have been divided, and output the analysis value;

[0182] When the analysis value is greater than a preset threshold, a target user group is generated.

[0183] In one embodiment, when the analysis value is greater than a preset threshold, the processor 1001 generates a target user group by specifically performing the following operations:

[0184] When the analysis value is less than the preset threshold, the algorithm parameters are adjusted;

[0185] The step of comparing and analyzing the clustered user group with the divided multiple user group types is continued based on the adjusted algorithm until the analysis value is greater than a preset threshold, thereby generating a target user group.

[0186] In one embodiment, the processor 1001 performs the following operations when collecting and preprocessing truck data and user data:

[0187] Collecting truck data and user data; wherein the truck data at least includes basic information data of the truck and operation information data of the truck, and the user data at least includes basic information data of the user and data generated by the user on the APP;

[0188] Clean the basic information data of the truck, the operation information data of the truck, the basic information data of the user, and the data generated by the user on the APP to generate pre-processed truck data and user data;

[0189] Among them, data cleaning includes at least removing redundant data, normalizing data formats, and unifying data standards.

[0190] In an embodiment of the present application, a task recommendation device based on user portraits first collects and preprocesses truck data and user data, then associates the preprocessed truck data and user data to generate a pulled-together data set; wherein the pulled-together data set includes multiple successfully associated truck data and user data; then, a label topic set of the user portrait is set according to the target application scenario, and then multiple successfully associated truck data and user data are mapped one by one to each label topic in the label topic set of the user portrait to generate a label topic set of associated data, and then the label topic set of the associated data is filtered according to the preset filtering rules to generate a target user group, and finally, at least one task corresponding to the target application scenario is recommended to the target user group. Since the present application effectively integrates the data generated by trucks and users, and conducts joint analysis and mining based on the integrated information, the dimension and quality of constructing user portraits are enriched, and the accuracy of service recommendations for users is further improved.

[0191] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program for recommending tasks based on user profiles can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes in the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0192] The above disclosure is only the preferred embodiment of the present application, which certainly cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made according to the claims of the present application are still within the scope covered by the present application.

Claims

1. A task recommendation method based on user portrait, characterized in that: The method comprises: Collect and pre-process truck data and user data; wherein the truck data at least includes basic information data of the truck and operation information data of the truck, and the user data at least includes basic information data of the user and data generated by the user on the APP; Associating the pre-processed truck data with the user data to generate a connected data set; wherein the connected data set includes a plurality of successfully associated truck data and user data; Set the tag theme set of the user portrait according to the target application scenario; Mapping the multiple successfully associated truck data and user data one by one to each label topic in the label topic set of the user portrait to generate a label topic set of associated data; The tag subject set of the associated data is filtered according to the preset filtering rules to generate a target user group; wherein, The step of filtering the tag subject set of the associated data according to the preset filtering rules to generate a target user group includes: Divide multiple user groups into types based on the business recommendation requirements, experience and data of the target application scenario; Screening each tag data in the tag topic set of the associated data as a feature to generate multiple types of tag topics; Preprocessing the multiple types of label topics, and performing vector conversion on the preprocessed multiple types of label topics to generate a user set; Using Kmeans clustering algorithm to perform clustering calculation on the user set to generate clustered user groups; Compare and analyze the clustered user group with the divided multiple user group types, and output an analysis value; When the analysis value is greater than a preset threshold, a target user group is generated; At least one task corresponding to the target application scenario is recommended to the target user group.

2. The method according to claim 1, characterized in that After recommending at least one task corresponding to the target application scenario to the target user group, the method further includes: analyzing operation information of the user group for the at least one task to generate an analysis result; Based on the analysis result, the tag topic set of the user portrait is updated or deleted to generate the tag topic set of the updated or deleted user portrait; Mapping the multiple successfully associated truck data and user data one by one to each label topic in the label topic set of the updated or deleted user portrait to generate a label topic set of associated data; and / or updating or deleting the user's processing information for the task in the task recommendation module based on the analysis result, and generating updated or deleted user processing information; Mapping the plurality of successfully associated truck data and user data one by one to the updated or deleted user processing information; Continue to perform the step of filtering the tag topic set and / or user processing information of the associated data according to the preset filtering rule.

3. The method according to claim 1, characterized in that The step of filtering the tag subject set of the associated data according to the preset filtering rules to generate a target user group includes: Initialize the preset label theme parameter value; Obtaining the parameter value of each tag topic in the tag topic set of the associated data; Compare each tag theme parameter value in the tag theme set of the associated data with the preset tag theme parameter value to determine whether there is at least one tag with a value greater than the preset tag theme parameter value; If so, obtaining the user corresponding to the at least one tag having a value greater than the preset tag topic parameter value generates a target user group.

4. The method according to claim 1, characterized in that The step of filtering the tag subject set of the associated data according to the preset filtering rules to generate a target user group includes: Initialize multiple preset label theme parameter values; Obtaining the parameter value of each tag topic in the tag topic set of the associated data; Compare each tag theme parameter value in the tag theme set of the associated data with the multiple preset tag theme parameter values ​​to determine whether there are multiple tags with values ​​greater than the preset tag theme parameter values; If so, the users corresponding to the multiple tags whose values ​​are greater than the preset tag topic parameter are obtained to generate a target user group.

5. The method according to claim 1, characterized in that When the analysis value is greater than a preset threshold, generating a target user group includes: When the analysis value is less than a preset threshold, adjusting the parameters of the algorithm; The step of comparing and analyzing the clustered user group with the divided multiple user group types is continued based on the adjusted algorithm until the analysis value is greater than a preset threshold, thereby generating a target user group.

6. The method according to claim 1, characterized in that The collecting and preprocessing of truck data and user data includes: Collect truck data and user data; Clean the basic information data of the truck, the operation information data of the truck, the basic information data of the user, and the data generated by the user on the APP to generate pre-processed truck data and user data; The data cleaning includes at least removing redundant data, normalizing data formats and unifying data standards.

7. A task recommendation device based on user portrait, characterized in that: The device comprises: A data processing module, used to collect and pre-process truck data and user data; wherein the truck data at least includes basic information data of the truck and operation information data of the truck, and the user data at least includes basic information data of the user and data generated by the user on the APP; A data association module, used to associate the pre-processed truck data with the user data to generate a connected data set; wherein the connected data set includes a plurality of successfully associated truck data and user data; The tag theme setting module is used to set the tag theme set of the user portrait according to the target application scenario; A data mapping module, used to map the multiple successfully associated truck data and user data one by one to each label topic in the label topic set of the user portrait, and generate a label topic set of associated data; The user group generation module is used to filter the tag subject set of the associated data according to the preset filtering rules to generate a target user group; wherein, The user group generation module is specifically used for: Divide multiple user groups into types based on the business recommendation requirements, experience and data of the target application scenario; Screening each tag data in the tag topic set of the associated data as a feature to generate multiple types of tag topics; Preprocessing the multiple types of label topics, and performing vector conversion on the preprocessed multiple types of label topics to generate a user set; Using Kmeans clustering algorithm to perform clustering calculation on the user set to generate clustered user groups; Compare and analyze the clustered user group with the divided multiple user group types, and output an analysis value; When the analysis value is greater than a preset threshold, a target user group is generated; The task recommendation module is used to recommend at least one task corresponding to the target application scenario to the target user group.

8. A computer storage medium, characterized in that: The computer storage medium stores a plurality of instructions, which are suitable for being loaded by a processor and executing the method steps according to any one of claims 1 to 6.

9. A terminal, characterized in that: include: A processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the method steps as claimed in any one of claims 1 to 6.

Images