Personnel management method and system based on immigrant portrait technology

Abstract

The invention discloses a personnel management method and system based on an immigrant portrait technology, and belongs to the technical field of data management. The method comprises the following steps: collecting existing immigrant data, and preprocessing the immigrant data to obtain user data, business data, behavior data and theme content data; labeling the user data and the theme content data to form immigrant user portraits and theme content portraits, and storing the immigrant user portraits and the theme content portraits; and according to the stored portrait system and the user resources, intelligent matching and application services are carried out on the requirements of the platform user. A perfect three-layer immigrant tag classification system is designed, basic information of the immigrant is concerned, the immigrant is portrayed from multiple dimensions such as placement conditions, economic income conditions, satisfaction conditions and social demands, and a comprehensive immigrant portraying system is formed. The multi-dimensional portraying method is helpful to more comprehensively understand the living state and demand of immigrants.

Description

Technical Field

[0001] The present invention relates to the field of data management technology, and in particular to a personnel management method and system based on immigration profiling technology. Background Art

[0002] With the continuous development of hydropower projects, a large number of reservoir migrant groups have been formed. These migrant groups have generated a large amount of data in the process of resettlement, including basic personal information, family composition, social relations, economic activities, etc. Effective social management and services for these specific groups require the integration, analysis and application of these large amounts of scattered information. The existing immigration management system often cannot meet the needs of accurate identification and in-depth understanding of individual immigrants. There are generally problems such as incomplete information collection, inaccurate analysis, and low efficiency, and it is difficult to meet the needs of large-scale data analysis. With the development of big data and artificial intelligence technology, the present invention proposes a technical application method that can automatically identify, effectively integrate, quickly match and analyze, and generate portraits of individual and group immigrants, so as to provide immigrants with more efficient, accurate and personalized services, and further improve the management efficiency and service quality of immigration management agencies. Summary of the invention

[0003] In view of the above-mentioned problems, the present invention is proposed.

[0004] Therefore, the problem to be solved by the present invention is that the existing immigration management system is often unable to meet the needs of accurate identification and in-depth understanding of individual immigrants, and generally has problems such as incomplete information collection, inaccurate analysis, and low efficiency, and it is difficult to meet the needs of large-scale data analysis.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a personnel management method based on immigrant portrait technology, which includes collecting existing immigrant data, pre-processing the immigrant data to obtain user data, business data, behavior data and subject content data; labeling the user data and subject content data to form immigrant user portraits and subject content portraits and store them; based on the stored portrait system and user resources, intelligent matching and application services are performed for the needs of platform users.

[0006] As a preferred solution of the personnel management method based on immigrant portrait technology described in the present invention, wherein: the preprocessing includes data processing and feature extraction, the data processing includes data collection, data cleaning, data conversion and data integration; the feature extraction includes extracting basic characteristics of immigrants, immigrant resettlement characteristics, immigrant economic characteristics, immigrant satisfaction characteristics, user login characteristics, user behavior characteristics, immigrant statistical characteristics, login statistical characteristics and behavior statistical characteristics from the processed data.

[0007] As a preferred solution of the personnel management method based on immigrant portrait technology described in the present invention, the labeling processing of user data and subject content data includes: user labeling is to perform secondary modeling processing and calculation on user-related data in the data warehouse, analyze data distribution and correlation, combine the biometric information of individual immigrants and immigrant-related data to discover the relationship and pattern between data, extract representative features for immigrant portrait construction, and label the features to construct a complete immigrant user portrait; subject content labeling is to extract corresponding subject words and keywords from the article content, label them and form a subject content portrait.

[0008] As a preferred solution of the personnel management method based on immigrant portrait technology described in the present invention, the label includes three layers, namely label subject, label type and label value; the label subject is divided and calibrated according to the extracted features; the label type is the sub-feature information under the label subject, and at the same time, the update frequency and update period are set on each label type; the label value represents the value range of the label under its corresponding label type, which is used to specifically describe the immigrant portrait.

[0009] As a preferred solution of the personnel management method based on the immigrant portrait technology described in the present invention, the specific steps of constructing the portrait include user topic weight calculation, user tag score calculation and calculation result storage;

[0010] The user topic weight calculation is to calculate the total score of the weights of each behavior type of the user under a certain topic page resource on the platform in the most recent time. This score is used as the basic data for subsequent user tag score calculation. The formula for the weight score WV of a user on a certain topic page resource on the platform is expressed as: WV=W 浏览 ×N 浏览 ×TD 浏览 +W 收藏 ×N 收藏 ×TD 收藏 +W 分享 ×N 分享 ×TD 分享

[0011] +W 点击 ×N 点击 ×TD 点击

[0012] Where W represents the weight of the user in the current behavior type, N represents the number of times the user has the current behavior type, if the subscript is browsing behavior, N is the duration, and TD represents time decay; the calculation formula of time decay TD is expressed as,

[0013]

[0014] Among them, t represents the distance between the time when the event occurred and the current time; the user tag score is calculated by combining the weight score of the user's topic page resource with the corresponding tag score under the page resource to calculate the user tag score LV. The calculation formula is expressed as:

[0015] LV=T / (T 标签1 +T 标签2 +....T 标签n )×WV

[0016] Among them, T represents the score corresponding to the tag under the topic page. If T has no subscript, it represents the score corresponding to the current tag under the topic page. The current tag is any one of tags 1 to n. The score calculated by the user tag is stored in . The user portrait table constitutes and describes the user preference and habit information.

[0017] As a preferred solution of the personnel management method based on immigration portrait technology described in the present invention, wherein: the intelligent matching includes automatic matching and manual matching; the automatic matching is based on the principle of matching a maximum of M pieces of information each time, combined with the corresponding scores of users on each tag in the user portrait table, and the proportion of the total score is calculated. The number of page resources corresponding to each tag is obtained through the proportion and the total number of recommended resources and stored in the user recommendation table. The matching algorithm is expressed as:

[0018] S=V / (V 标签1 +V 标签2 +....V 标签n )×M

[0019] Among them, S represents the number of page resources to be recommended corresponding to the current tag, V represents the score corresponding to the user tag, if V has no subscript, it represents the score corresponding to the current user tag, and the current user tag is any one of tags 1 to n; the manual matching includes, for a specific group of people, manually setting the tag rules corresponding to the user grouping, analyzing and calculating the user groups that meet the conditions, combining the recommended and to-be-recommended tables of users within the range, and obtaining a fixed number of page resources from the corresponding tag scores in the subject portrait table from high to low and storing them in the user to-be-recommended table.

[0020] As a preferred solution of the personnel management method based on immigrant portrait technology described in the present invention, the application service includes personalized recommendation service, immigrant portrait platform and data analysis platform; the personalized recommendation service targets individual immigrants, and the immigrant portrait platform and data analysis platform are used by business personnel; the personalized recommendation service includes obtaining the corresponding page resources of the current user from the table to be recommended for push service, and transferring the exposed page resource information to the recommended table for users to view the browsing history. If there is no available resource record in the table to be recommended, the intelligent matching program is automatically executed; the immigrant portrait platform includes four functions: tag management, user analysis, user query and user grouping;

[0021] The data analysis platform includes visually displaying various basic data in multiple dimensions through visualization tools, which are used to analyze the statistics of key immigration indicators and data.

[0022] Another object of the present invention is to provide a personnel management system based on immigrant portrait technology, which can automatically identify, effectively integrate, quickly match and analyze, and generate individual and group portraits of immigrants.

[0023] In order to solve the above technical problems, the present invention provides the following technical solutions: a system of personnel management method based on immigrant portrait technology, comprising: a basic data layer, a data processing layer and an application service layer; the basic data layer collects existing immigrant data, pre-processes the immigrant data to obtain user data, business data, behavior data and subject content data; the data processing layer labels the user data and subject content data to form immigrant user portraits and subject content portraits and store them; the application service layer performs intelligent matching and application services for the needs of platform users based on the stored portrait system and user resources

[0024] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the personnel management method based on immigration profiling technology as described above are implemented.

[0025] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a personnel management method based on immigration profiling technology as described above.

[0026] The beneficial effects of the present invention are as follows: 1. Automated data collection: using automated methods such as API interface calls and local file queries to collect massive amounts of data related to immigrants in real time, including immigrants' personal information, resettlement information, log information, etc. This automated data collection method reduces manpower input, reduces costs, and greatly improves the efficiency and accuracy of data collection.

[0027] 2. Multi-source and multi-type data integration: The group's existing immigration system data, platform logs and embedded data, external data and other multi-source data are efficiently processed and uniformly stored using advanced technologies such as Hadoop HDFS and NoSQL database to build a comprehensive immigration portrait dataset that integrates multiple data formats such as structured data (such as population information), semi-structured data (such as text descriptions) and unstructured data (such as pictures, videos).

[0028] 3. Comprehensive immigrant portrait construction: A well-designed three-tier immigrant label classification system not only focuses on the basic information of immigrants, but also portraits immigrants from multiple dimensions such as resettlement, economic income, satisfaction, and social needs, forming a comprehensive immigrant portrait system. This multi-dimensional portrait method helps to more comprehensively understand the living conditions and needs of immigrants.

[0029] 4. Intelligent matching and precise recommendation: We have designed a precise intelligent matching calculation model that can quickly and automatically search and match corresponding service resources based on the immigrants’ portraits. At the same time, by building a recommendation algorithm, combined with the immigrants’ personal information, needs and preferences, we can intelligently match and recommend the most appropriate service solutions, providing immigrants with more precise and personalized services.

[0030] 5. Data visualization: The platform interacts with users through the immigration profiling platform and data analysis platform, and intuitively displays analysis results to users in the form of charts, maps, reports and dashboards, making complex data information intuitive and easy to understand, so that non-professionals can also use the technology easily, thereby improving the popularity and ease of use of the technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0032] Figure 1 A flowchart for constructing an immigrant portrait based on immigrant portrait technology in Example 1.

[0033] Figure 2 This is a data collection, analysis and storage design diagram for a personnel management method based on immigrant profiling technology in Example 1.

[0034] Figure 3 FIG1 is an immigration label classification system 1 of a personnel management method based on immigration profiling technology in Example 1.

[0035] Figure 4FIG2 is an immigration label classification system of a personnel management method based on immigration profiling technology in Example 2.

[0036] Figure 5 FIG3 is an immigration label classification system of a personnel management method based on immigration profiling technology in Example 1.

[0037] Figure 6 FIG1 is a diagram of a database table structure design for a personnel management method based on immigration profiling technology in Example 1.

[0038] Figure 7 FIG2 is a database table structure design of a personnel management method based on immigration profiling technology in Example 1.

[0039] Figure 8 Figure 3 is a database table structure design of a personnel management method based on immigration profiling technology in Example 1.

[0040] Fig. 9 This is a schematic diagram of personalized immigration recommendations for a personnel management method based on immigration profiling technology in Example 1.

[0041] Fig.10 This is a structural diagram of a personnel management system based on immigration profiling technology in Example 2. DETAILED DESCRIPTION

[0042] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0043] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0044] Example 1, reference Figure 1-Figure 9 , which is the first embodiment of the present invention, and which provides a personnel management method based on immigrant portrait technology, including: Figure 1 As shown:

[0045] Step 1: Collect existing immigration data and pre-process the immigration data to obtain user data, business data, behavior data and subject content data.

[0046] The group collects and imports the existing immigration system data, cloud platform logs, embedded data, external data and other multi-source data into HDFS through API interface calls and local file queries. The collected data is processed and integrated in batches through Hive calling MapReduce tasks. The specific data flow diagram is as follows: Figure 2 shown.

[0047] Preprocessing includes data processing and feature extraction, and data processing includes data collection, data cleaning, data conversion and data integration.

[0048] (1) Data collection: The group collects data from multiple sources, including the group's existing immigration system data, cloud platform logs and tracking data, and external data, to form four types of original data sets, including user data, business data, behavior data, and subject content data. This data, as the most original data that has not been processed, is the source data of other upper-level data. In terms of structure, its table or field naming is basically consistent with the business system. At the same time, the time dimension is added as a partition field, and a combination of incremental and full data is used for data storage. Incremental storage is performed in units of days, with business dates as partitions. Each partition stores daily incremental business data, such as platform logs and tracking data; full storage is performed in units of days, with business dates as partitions. Each partition stores full business data up to the business date, such as basic immigration information, immigration resettlement information, and other data.

[0049] (2) Data cleaning: Use HiveQL string functions, date functions, and conditional expressions to clean data and ensure data quality. For example, string processing functions such as trim, lower, and regexp_replace can be used to remove spaces, convert character case, and replace strings with specific patterns. CASE statements and WHERE clauses can be used to filter data based on specific conditions to remove abnormal values ​​and data that meets specific conditions. Missing values ​​can be filtered out using the is null and notnull conditions. Group by and having clauses can be used to identify and process duplicate data.

[0050] (3) Data conversion: Perform type conversion, format unification, and calculation of derived attributes on the cleaned data. For example, use HiveQL functions and expressions to convert date strings into date types and convert all numeric data into a unified numeric type; use HiveQL arithmetic operators and functions to calculate derived attributes, such as calculating the difference in days between two dates and calculating composite indicators based on multiple fields.

[0051] (4) Data integration: Use Hive's table connection, view, and index functions to merge the cleaned and converted data according to the specified key columns and integrate them into a unified data analysis model to create the immigration cloud platform fact table, immigration system migration fact table, statistical aggregation table, etc., in preparation for data analysis and feature extraction.

[0052] Feature extraction includes in-depth analysis of the cleaned and organized data, and the use of classification, clustering and other technologies to extract representative and useful key features based on immigration business needs and data characteristics. Specifically, the following features are included:

[0053] Basic characteristics of immigrants: including age, gender, marital status, education level, major category, survey scope, occupation type, etc.

[0054] Characteristics of resettlement of immigrants: including definition status, relocation and resettlement methods, relocation and resettlement destination, relocation and resettlement location, production resettlement methods, etc.

[0055] Economic characteristics of immigrants: including total annual income, wage income, business income, other income, total annual expenditure, family business expenditure, expenditure on purchasing fixed assets, other expenditure, etc.

[0056] Characteristics of immigrant satisfaction: including satisfaction with resettlement planning and design, satisfaction with resource allocation, satisfaction with relocation adaptation, satisfaction with living environment, satisfaction with village infrastructure, satisfaction with publicity, and satisfaction with smooth complaint channels.

[0057] User login characteristics: including login method, login address, login time period, terminal type, device model, etc.

[0058] User behavior characteristics: including behavior time, behavior type, behavior object, etc.

[0059] Statistical characteristics of immigrants: family size, house area, per capita house area, land area, per capita land area, number of pieces of wood, amount of compensation subsidy, per capita annual income, proportion of wage income, proportion of business income, per capita annual expenditure, proportion of living consumption, etc.

[0060] Login statistical features: number of logins in the past 3 days, number of logins in the past 7 days, number of logins in the past 30 days, cumulative number of logins, average activity in the past 3 days, average activity in the past 7 days, average activity in the past 30 days, cumulative average activity, active time period, etc.

[0061] Behavioral statistical characteristics: number of views, number of clicks, number of favorites, number of follows, etc.

[0062] Database selection and design: The platform uses a combination of multiple databases to store data according to different application scenarios.

[0063] In order to facilitate the analysis of the massive amount of data collected (including the group's existing immigration system data, platform logs and embedded data, external data, etc.), Hive is used to build a data warehouse. Hive is a data warehouse tool based on Hadoop, relying on HDFS to store data, and the SQL language provided can easily query the data stored in HDFS.

[0064] In order to quickly read and use the immigrant user portraits, we chose to store the portrait data in HBase. HBase is a high-performance, column-based, scalable, real-time read-write distributed storage system. Unlike Hive, HBase can run in real time on the database and is suitable for real-time query of big data. Since HBase cannot solve the problem of efficient retrieval of data under multiple conditions, we use Elasticsearch to establish a secondary index for storage. The secondary index is used to quickly locate the primary index rowkey to meet the needs of querying data based on complex conditions.

[0065] For basic data such as tag definition, user group definition, dictionary library, monitoring information, etc. in the platform, conventional relational databases are used for data storage, such as the DAMENG database.

[0066] Step 2: Label the user data and subject content data to form immigrant user portraits and subject content portraits and store them.

[0067] The data processing layer mainly performs labeling on two parts of data: users and subject content, which is an important part of portrait modeling.

[0068] Thematic content labeling mainly involves extracting corresponding subject terms and keywords from article content, labeling them and thus forming a thematic content portrait.

[0069] User labeling mainly involves secondary modeling, processing and calculation of user-related data in the data warehouse, in-depth analysis to understand data distribution and correlation, and combining individual immigrant biometric information with various related data such as immigrant resettlement and economy to discover relationships and patterns between data, extract useful features that are representative of immigrant portrait construction, and use statistics, rules and other methods to formulate labels to construct a complete immigrant user portrait.

[0070] The generated immigrant user portrait and subject content portrait related results are stored to facilitate the subsequent updating and application of the portrait.

[0071] To facilitate management, tags need to be named in a standardized manner. According to actual business needs and scenarios, the entire tag system is divided into three layers, namely tag subject, tag type, and tag value.

[0072] (1) Labeling system

[0073] Tag topic (first level): used to indicate which category the tag belongs to, such as user basic attributes (BASIC), user behavior attributes (ACTION), immigrant resettlement attributes (RESETTLE), immigrant economic attributes (ECONOMY), immigrant satisfaction attributes (SATISFACTION), etc.

[0074] Tag type (secondary): Under each tag category, the tag type is further subdivided, such as age, gender, etc. under the basic attributes of the user. At the same time, the update frequency and update period are set on each tag type to facilitate the scheduling of the corresponding computing tasks according to different update rules.

[0075] Label value (Level 3): The label value indicates the value range of the label under its corresponding label type, which is used to describe the specific portrait of immigrants.

[0076] (2) Tag naming

[0077] Tags are named using the rule of tag subject_tag type_tag value. For example, the user's gender tag is named as follows:

[0078] [Male]: BASIC_02_001

[0079]

Female

[0080] The specific label classification system is as follows Figure 3-Figure 5 shown.

[0081] The specific steps of constructing the portrait are as follows: Integrate the extracted features and the created labels to construct a complete immigrant portrait. The steps are mainly divided into three parts: user topic weight calculation, user label score calculation and calculation result storage.

[0082] The user topic weight calculation is to calculate the total score of the weight of each behavior type of the user under a certain topic page resource on the platform in the most recent time. This score is used as the basic data for the subsequent user tag score calculation. The formula for the weight score WV of a user on a certain topic page resource on the platform is expressed as: WV = W 浏览 ×N 浏览 ×TD 浏览 +W 收藏 ×N 收藏 ×TD 收藏 +W 分享 ×N 分享 ×TD 分享

[0083] +W 点击 ×N 点击 ×TD 点击

[0084] Where W represents the weight of the user in the current behavior type, N represents the number of times the user has the current behavior type, if the subscript is browsing behavior, N is the duration, and TD represents time decay.

[0085] The weights assigned to each user behavior type are shown in the following table:

[0086] Behavior Behavior Weight Browsing time (<2000ms) 1 Browsing time (>=2000ms) 2 collect 2 share 3 Click 5

[0087] Since the user's preference decays with time, that is, the farther away from the current time, the greater the decay, the calculation formula of time decay TD is expressed as:

[0088]

[0089] Among them, t represents the distance between the time when the event occurs and the current time.

[0090] The user tag score is calculated by combining the weight score of the user's topic page resource with the corresponding tag score under the page resource to calculate the user tag score LV. The calculation formula is expressed as:

[0091] LV=T / (T 标签1 +T 标签2 +....T 标签n )×WV

[0092] Among them, T represents the score corresponding to the tag under the topic page. If T has no subscript, it represents the score corresponding to the current tag under the topic page. The current tag is any one of tag 1 to tag n.

[0093] The scores calculated by the above user tags are stored in the user portrait table, which constitutes and describes various information such as user preferences and habits. The user portrait platform can be directly used to display the personal portrait of immigrants and serve as the basis for the platform's subsequent intelligent matching and recommendations.

[0094] Database structure design: In combination with the characteristics of immigration business and data, the database structure mainly includes platform fact table, migration fact table, statistical aggregation table, portrait table, push table, and other basic tables. The specific table structure design is as follows: Figure 6-Figure 8 shown.

[0095] The platform fact table is a built-in table of the platform, which mainly includes user registration information, user login logs, user behavior logs and tracking logs.

[0096] The migration fact table is a data storage space that is migrated synchronously from other systems. It is mainly for the convenience of subsequent analysis operations. When the table is created, it is consistent with the original data structure. It mainly includes immigrant household information, immigrant population information, immigrant resettlement information, immigration indicators and funding information, immigrant income and expenditure information, and immigrant satisfaction information.

[0097] The statistical aggregation table is based on the fact table migrated from the platform and performs statistical summary, and stores the commonly used statistical results, so that the later data analysis platform can quickly and easily extract relevant data for users to use, thereby improving the statistical speed. It mainly includes the population basic information summary table, the immigrant resettlement situation summary table, the immigrant economic situation summary table, the immigrant satisfaction situation summary table, the user login situation summary table, and the user behavior situation summary table.

[0098] The portrait table is mainly used to store the relevant results calculated by the algorithm, including the user login attribute table, user behavior attribute table, user portrait weight calculation table, user portrait table, and theme portrait table. The user login attribute table and user behavior attribute table are used to store the user's login and behavior-related feature attribute value information on the platform; the portrait weight calculation table is used to store the weighted scores of each behavior type on the page; the user portrait table and theme portrait table are mainly used to store the scores of users and topics under various tags, so as to facilitate the push of pages based on the portrait table calculation in the later stage.

[0099] The push table is mainly used to store the user's to-be-pushed and pushed table topic page resource data, so that users can directly obtain the push page and view the history. It mainly includes the user recommended table and the user to-be-recommended table.

[0100] Other basic tables mainly use relational databases to store the addition, deletion, modification and query data of relevant functions in the platform interface. They mainly include tag definition tables and user group related tables. The tag definition table is mainly used to store basic information such as the name, code, enabled status and update frequency of the defined tag; the user group related table is mainly used to store basic information of specific groups in the platform and their tag rule settings, so as to facilitate subsequent analysis and calculation of user groups that meet the conditions and complete the push service of corresponding information.

[0101] Step 3: Based on the stored portrait system and user resources, intelligent matching and application services are provided for the needs of platform users.

[0102] Smart matching includes automatic matching and manual matching. Automatic matching follows the principle of matching a maximum of M pieces of information each time. Combined with the corresponding scores of users on each tag in the user portrait table, the proportion of the total score is calculated. The number of page resources corresponding to each tag is obtained through the proportion and the total number of recommended resources and stored in the user's waiting recommendation table. The matching algorithm is expressed as:

[0103] S=V / (V 标签1 +V 标签2 +....V 标签n )×M

[0104] Among them, S represents the number of page resources to be recommended corresponding to the current tag, V represents the score corresponding to the user tag, if V has no subscript, it represents the score corresponding to the current user tag, the current user tag is any one of tag 1 to tag n, and the initial fixed value of M in this embodiment is 10.

[0105] Manual matching includes, for a specific group of people, manually setting the label rules corresponding to user grouping, analyzing and calculating the user groups that meet the conditions, combining the recommended and to-be-recommended tables of users in the range, obtaining a fixed number (the platform is set to 3) of page resources with corresponding label scores from high to low in the subject portrait table and storing them in the user to-be-recommended table. For example, users with "high" in [Activity] in [User Behavior Attributes] and "Employment Section" in [Content Preference Attributes] are grouped and calculated as users with employment needs, and page resources containing employment-related tags are sent to them and stored in the to-be-recommended table.

[0106] The application services mainly include personalized recommendation services, immigration portrait platform and data analysis platform. The personalized recommendation service targets individual immigrants, while the immigration portrait platform and data analysis platform are used by business personnel.

[0107] The personalized recommendation service includes obtaining the page resources corresponding to the current user from the table to be recommended for push service, and transferring the exposed page resource information to the recommended table for users to view the browsing history. In addition, it is determined whether there is an available resource record in the table to be recommended. If not, the intelligent matching program is automatically executed to ensure that the recommended resource is stored in the table to be recommended. The specific diagram is as follows Fig. 9 shown.

[0108] The immigration portrait platform mainly includes four functions: label management, user analysis, user query and user grouping.

[0109] Tag management: As the core of the profiling platform, it is used to create, modify, delete and query various tags. These tags are generated based on various data of immigrants and are used to describe and classify immigrant groups.

[0110] User analysis: Use label data to conduct in-depth analysis of immigrant groups, including analysis of immigration trends, multi-dimensional assessment of immigration demand characteristics, etc., to provide a basis for formulating targeted and differentiated immigration policies and measures.

[0111] User query: By entering specific conditions (such as name, ID number, etc.), you can quickly locate and query the information and label data of relevant immigrants. The query results are displayed in the form of a list or chart.

[0112] User grouping: Crowd selection is performed based on rules such as intersection, union, and difference operations between tags, and immigrant groups are divided into different sub-groups to enable more accurate management and services and meet more diverse selection needs.

[0113] As a statistical analysis platform for key indicators and data of immigration, the data analysis platform intuitively displays various basic data in multiple dimensions including resettlement location, time period, etc., such as basic information of immigrants, resettlement status of immigrants, satisfaction of immigrants, etc., through tables, charts or other visualization tools, providing managers with comprehensive and accurate immigration effectiveness data.

[0114] Example 2, reference Fig.10 , which is the second embodiment of the present invention, and which is different from the first embodiment in that: a system of a personnel management method based on immigrant portrait technology, characterized in that: it includes a basic data layer, a data processing layer and an application service layer; the basic data layer collects existing immigrant data, and pre-processes the immigrant data to obtain user data, business data, behavior data and subject content data; the data processing layer labels the user data and subject content data to form immigrant user portraits and subject content portraits and store them; the application service layer performs intelligent matching and application services for the needs of platform users based on the stored portrait system and user resources.

[0115] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0116] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0117] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0118] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0119] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. A personnel management method based on immigration profiling technology, characterized by: include, Collect existing immigration data and pre-process the immigration data to obtain user data, business data, behavior data and subject content data; Label the user data and subject content data to form and store immigrant user portraits and subject content portraits; Based on the stored portrait system and user resources, intelligent matching and application services are provided to meet the needs of platform users.

2. A personnel management method based on immigrant profiling technology as claimed in claim 1, characterized in that: The preprocessing includes data processing and feature extraction, and the data processing includes data collection, data cleaning, data conversion and data integration; The feature extraction This includes extracting basic characteristics of immigrants, immigrant resettlement characteristics, immigrant economic characteristics, immigrant satisfaction characteristics, user login characteristics, user behavior characteristics, immigrant statistical characteristics, login statistical characteristics and behavior statistical characteristics from the processed data.

3. A personnel management method based on immigrant profiling technology as claimed in claim 2, characterized in that: The labeling of user data and subject content data includes: user labeling is to perform secondary modeling and calculation on the data related to the user in the data warehouse, analyze the data distribution and correlation, combine the biometric information of individual immigrants and immigration-related data to discover the relationship and pattern between the data, extract the representative features for constructing immigrant portraits, and label the features to construct a complete immigrant user portrait; Topic content labeling is to extract corresponding subject words and keywords from the article content, label them and thus form a topic content portrait.

4. A personnel management method based on immigrant profiling technology as claimed in claim 3, characterized in that: The tag includes three layers, namely, tag subject, tag type and tag value; The label topics are divided and calibrated according to the extracted features; The tag type is the sub-feature information under the tag theme. At the same time, the update frequency and update period are set on each tag type; The tag value represents the value range of the tag under its corresponding tag type, and is used to specifically describe the immigrant portrait.

5. A personnel management method based on immigrant profiling technology as claimed in claim 4, characterized in that: The specific steps of constructing the portrait include user topic weight calculation, user tag score calculation and calculation result storage; The user topic weight calculation is to calculate the total score of the weights of each behavior type of the user under a certain topic page resource on the platform in the most recent time. This score is used as the basic data for subsequent user tag score calculation. The formula for the weight score WV of a user on a certain topic page resource on the platform is expressed as: WV=W 浏览 ×N 浏览 ×TD 浏览 +W 收藏 ×N 收藏 ×TD 收藏 +W 分享 ×N 分享 ×TD 分享 +W 点击 ×N 点击 ×TD 点击 Where W represents the weight of the user in the current behavior type, N represents the number of times the user has the current behavior type, if the subscript is browsing behavior, N is the duration, and TD represents time decay; The calculation formula of time decay TD is expressed as: Among them, t represents the distance between the time when the event occurs and the current time; The user tag score is calculated by combining the weight score of the user's topic page resource with the corresponding tag score of the page resource to calculate the user tag score LV. The calculation formula is: LV=T / (T 标签1 +T 标签2 +.....T 标签n )×WV Where T represents the score corresponding to the tag on the topic page. If T has no subscript, it represents the score corresponding to the current tag on the topic page. The current tag is any one of tag 1 to tag n. The scores calculated by the user tags are stored in , and the user portrait table constitutes and describes the user preference and habit information.

6. A personnel management method based on immigrant profiling technology as claimed in claim 5, characterized in that: The intelligent matching includes automatic matching and manual matching; The automatic matching is based on the principle of matching a maximum of M pieces of information each time, and combines the corresponding scores of users on each tag in the user portrait table to calculate their proportion under the total score. The number of page resources corresponding to each tag is obtained through the proportion and the total number of recommended resources and stored in the user recommendation table. The matching algorithm is expressed as: S=V / (V 标签1 +V 标签2 +.....V 标签n )×M Where S represents the number of recommended page resources corresponding to the current tag, V represents the score corresponding to the user tag, and if V has no subscript, it represents the score corresponding to the current user tag, and the current user tag is any one from tag 1 to tag n; The manual matching includes, for a specific group of people, manually setting the label rules corresponding to the user grouping, analyzing and calculating the user groups that meet the conditions, combining the recommended and to-be-recommended tables of users within the range, and obtaining a fixed number of page resources from the corresponding label scores in the subject portrait table from high to low and storing them in the user to-be-recommended table.

7. A personnel management method based on immigrant profiling technology as claimed in claim 6, characterized in that: The application services include personalized recommendation services, immigration profiling platform and data analysis platform; The personalized recommendation service targets individual immigrants, and the immigrant portrait platform and data analysis platform are used by business personnel; The personalized recommendation service includes obtaining the page resources corresponding to the current user from the table to be recommended for push service, transferring the exposed page resource information to the recommended table for the user to view the browsing history, and automatically executing the intelligent matching program if there is no available resource record in the table to be recommended; The immigration portrait platform includes four functions: tag management, user analysis, user query and user grouping; The data analysis platform includes visually displaying various basic data in multiple dimensions through visualization tools, which are used to analyze the statistics of key immigration indicators and data.

8. A system using a personnel management method based on immigration profiling technology as claimed in any one of claims 1 to 7, characterized in that: Includes basic data layer, data processing layer and application service layer; The basic data layer collects existing immigration data and pre-processes the immigration data to obtain user data, business data, behavior data and subject content data; The data processing layer performs labeling on user data and subject content data to form and store immigrant user portraits and subject content portraits; The application service layer performs intelligent matching and application services for the needs of platform users based on the stored portrait system and user resources.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of a personnel management method based on immigration profiling technology described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a personnel management method based on immigration profiling technology described in any one of claims 1 to 7 are implemented.

Images