CN111062750A - User portrait label modeling and analyzing method, device, equipment and storage medium - Google Patents
User portrait label modeling and analyzing method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN111062750A CN111062750A CN201911279395.2A CN201911279395A CN111062750A CN 111062750 A CN111062750 A CN 111062750A CN 201911279395 A CN201911279395 A CN 201911279395A CN 111062750 A CN111062750 A CN 111062750A
- Authority
- CN
- China
- Prior art keywords
- target object
- data
- target
- preset
- label
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000004458 analytical method Methods 0.000 claims abstract description 63
- 238000012545 processing Methods 0.000 claims abstract description 44
- 238000004140 cleaning Methods 0.000 claims abstract description 20
- 238000007418 data mining Methods 0.000 claims abstract description 13
- 239000000523 sample Substances 0.000 claims description 97
- 238000000605 extraction Methods 0.000 claims description 51
- 230000002452 interceptive effect Effects 0.000 claims description 21
- 238000012549 training Methods 0.000 claims description 15
- 239000013074 reference sample Substances 0.000 claims description 10
- 238000003064 k means clustering Methods 0.000 claims description 7
- 230000003068 static effect Effects 0.000 claims description 7
- 238000013524 data verification Methods 0.000 claims description 5
- 238000003062 neural network model Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 3
- 238000004590 computer program Methods 0.000 claims 2
- 238000002372 labelling Methods 0.000 abstract description 7
- 239000000284 extract Substances 0.000 description 13
- 230000008569 process Effects 0.000 description 13
- 230000006399 behavior Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 6
- 238000005457 optimization Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000013500 data storage Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000013075 data extraction Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/27—Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The invention relates to the technical field of big data, and discloses a user portrait label modeling and analyzing method, device, equipment and storage medium. The user portrait label modeling and analyzing method comprises the following steps: extracting data from a preset data source in a preset mode to obtain original data; cleaning and combining original data to obtain sample data; extracting multi-dimensional target features from the sample data, and processing the multi-dimensional target features to obtain attribute tags of the target object, wherein the target object and the attribute tags of the target object form a tag system of the target object; constructing a target model according to a label system of a target object; and carrying out data mining analysis on the target service through the target model to obtain an analysis result. According to the invention, through constructing the label system of the user portrait, the efficiency of constructing the user portrait is improved, and meanwhile, the accuracy and efficiency of labeling the target user are improved.
Description
Technical Field
The invention relates to the field of user behavior portrayal, in particular to a user portrayal label modeling and analyzing method, a device, equipment and a storage medium.
Background
The user portrait is a popular concept at present, online and offline behaviors of the user are analyzed, various labels and characteristics are marked on the user, a plurality of typical user portraits are formed, and accurate marketing and recommendation are performed on a target user through the user portraits. The user portrait is a tagged user model abstracted according to information such as attributes, preferences, living habits, behaviors and the like of a user. Colloquially, a user is labeled, and the label is a highly refined characteristic mark obtained by analyzing user information. By tagging, a user may be described with some highly generalized, easily understandable features that may make it easier for a person to understand the user and may facilitate computer processing.
In the prior art, user portrayal mainly portrays users on the basis of internet behaviors, for example, behavior analysis and label processing are performed on users on the basis of browsing records, purchase records and the like of an online shopping mall. For some traditional industries, a related label system is lacking at present, so that the efficiency is low when a user portrait is constructed or modeling is carried out through business core data and user behavior data.
Disclosure of Invention
The invention mainly aims to solve the technical problems that a label system is lacked in preset services, and the user portrait construction efficiency is low.
To achieve the above object, a first aspect of the present invention provides a user portrait label modeling and analyzing method, including: extracting data from a preset data source through a preset mode to obtain original data, wherein the preset mode comprises a full quantity extraction mode and an increment extraction mode; cleaning and combining the original data to obtain sample data; extracting multidimensional target features from the sample data, and processing the multidimensional target features to obtain attribute tags of a target object, wherein the target object and the attribute tags of the target object form a tag system of the target object, and the target object comprises a vehicle, an interactive object, a user, accessory working hours, an answer and a road in a preset portrait system; constructing a target model according to a label system of the target object, wherein the label system is used for indicating that the corresponding target object is described according to the multi-dimensional target characteristics; and carrying out data mining analysis on the target service through the target model to obtain an analysis result, wherein the target service is determined in advance according to actual service requirements.
Optionally, in a first implementation manner of the first aspect of the present invention, the extracting multidimensional target features from the sample data, and processing the multidimensional target features to obtain attribute tags of a target object, where the target object and the attribute tags of the target object form a tag system of the target object, and the target object includes a vehicle, an interactive object, a user, accessory man-hours, an claim, and a road in a preset portrait system, and includes: reading a unique identifier of a target object from a preset data table, wherein the target object comprises a vehicle, an interactive object, a user, accessory working hours, an indemnity and a road in a preset portrait system; reading corresponding sample data from a second configuration unit hive database according to the unique identification of the target object and an object-oriented query language HQL grammar rule; extracting multi-dimensional target features from the read sample data according to a preset algorithm; carrying out clustering analysis on the multidimensional target characteristics according to a k-means clustering algorithm to obtain attribute labels of the target objects, wherein the attribute labels are named according to a preset label rule; setting the target object and the attribute tag of the target object as a tag system of the target object, and storing the tag system of the target object into the second hive database.
Optionally, in a second implementation manner of the first aspect of the present invention, the extracting, according to a preset algorithm, a multidimensional target feature from the read sample data includes: extracting a first feature from the read sample data according to a preset feature dimension; processing the read sample data through the trained model to obtain a second characteristic; and combining the first characteristic and the second characteristic to obtain a multi-dimensional target characteristic.
Optionally, in a third implementation manner of the first aspect of the present invention, the constructing a target model according to a label system of the target object, where the label system is configured to indicate that a target object corresponding to the multidimensional target feature description includes: extracting a label system of the target object from the second hive database according to preset target services to obtain an attribute label, wherein the label system is used for indicating the corresponding target object according to the multi-dimensional target feature description; dividing the attribute label into training sample data and reference sample data; and training a preset model according to the training sample data and the reference sample data to obtain a target model, wherein the preset model comprises a deep neural network model.
Optionally, in a fourth implementation manner of the first aspect of the present invention, the extracting data from the preset data source to obtain the original data, where the extracting data manner includes a full amount extraction manner and an incremental amount extraction manner, and includes: reading configuration information, wherein the configuration information is used for indicating that a preset data source is from a service database or a service file; performing full extraction on static metadata in the service database or data in the service file according to the configuration information, and performing incremental extraction on dynamic metadata in the service database according to the configuration information; performing data verification on the extracted metadata according to a preset data format; and performing redundancy processing on the checked data, and writing the data subjected to the redundancy processing into a first hive database to obtain original data.
Optionally, in a fifth implementation manner of the first aspect of the present invention, the cleaning and merging the original data to obtain sample data includes: when dirty data are detected to exist in the original data, cleaning the dirty data; merging the data with the same data structure after being cleaned according to a preset service theme to obtain sample data; and storing the sample data into a corresponding theme width table, wherein the theme width table is a data table of the second hive database.
Optionally, in a sixth implementation manner of the first aspect of the present invention, after the storing the tag generated by the target model into the second hive database according to the corresponding relationship of the target object, the user portrait tag modeling and analyzing method further includes: when detecting that a newly added object exists, acquiring sample data corresponding to the newly added object, extracting a label according to the sample data corresponding to the newly added object, and adding the newly added object to the target object; when detecting that a newly added field exists in the sample data of the target object and the newly added field value is label data to be extracted, performing clustering processing on the newly added field to obtain a newly added attribute label of the target object; and optimizing the attribute tag of the target object through a preset timing task, wherein the optimizing includes setting the attribute tag of the target object to be a null value or deleting the attribute tag of the target object.
A second aspect of the present invention provides a user portrait label modeling and analysis apparatus, comprising: the extraction unit is used for extracting data from a preset data source in a preset mode to obtain original data, wherein the preset mode comprises a full extraction mode and an increment extraction mode; the cleaning and merging unit is used for cleaning and merging the original data to obtain sample data; the processing unit is used for extracting multi-dimensional target features from the sample data and processing the multi-dimensional target features to obtain attribute tags of the target object, the target object and the attribute tags of the target object form a tag system of the target object, and the target object comprises a vehicle, an interactive object, a user, accessory working hours, a claim and a road in a preset portrait system; the construction unit is used for constructing a target model according to a label system of the target object, and the label system is used for indicating the target object corresponding to the multi-dimensional target feature description; and the analysis unit is used for carrying out data mining analysis on the target service through the target model to obtain an analysis result, wherein the target service is determined in advance according to the actual service requirement.
Optionally, in a first implementation manner of the second aspect of the present invention, the processing unit further includes: the system comprises a determining subunit, a display unit and a display unit, wherein the determining subunit is used for reading a unique identifier of a target object from a preset data table, and the target object comprises a vehicle, an interactive object, a user, accessory working hours, an indemnity and a road in a preset portrait system; the reading sub-unit is used for reading corresponding sample data from a second hive database of the configuration unit according to the unique identifier of the target object and an HQL grammar rule of an object-oriented query language; the first extraction subunit is used for extracting multi-dimensional target features from the read sample data according to a preset algorithm; the first clustering subunit is used for carrying out clustering analysis on the multidimensional target characteristics according to a k-means clustering algorithm to obtain attribute labels of the target objects, and the attribute labels are named according to a preset label rule; and the setting subunit is used for setting the target object and the attribute tag of the target object as a tag system of the target object, and storing the tag system of the target object into the second hive database.
Optionally, in a second implementation manner of the second aspect of the present invention, the first extraction subunit is specifically configured to: extracting a first feature from the read sample data according to a preset feature dimension; processing the read sample data through the trained model to obtain a second characteristic; and combining the first characteristic and the second characteristic to obtain a multi-dimensional target characteristic.
Optionally, in a third implementation manner of the second aspect of the present invention, the building unit is specifically configured to: extracting a label system of the target object from the second hive database according to preset target services to obtain an attribute label, wherein the label system is used for indicating the corresponding target object according to the multi-dimensional target feature description; dividing the attribute label into training sample data and reference sample data; and training a preset model according to the training sample data and the reference sample data to obtain a target model, wherein the preset model comprises a deep neural network model.
Optionally, in a fourth implementation manner of the second aspect of the present invention, the extracting unit is specifically configured to: reading configuration information, wherein the configuration information is used for indicating that a preset data source is from a service database or a service file; performing full extraction on static metadata in the service database or data in the service file according to the configuration information; performing incremental extraction on the dynamic metadata in the service database according to the configuration information; performing data verification on the extracted metadata according to a preset data format; and performing redundancy processing on the checked data, and writing the data subjected to the redundancy processing into a first hive database to obtain original data.
Optionally, in a fifth implementation manner of the second aspect of the present invention, the cleaning and merging unit is specifically configured to: when dirty data are detected to exist in the original data, cleaning the dirty data; merging the data with the same data structure after being cleaned according to a preset service theme to obtain sample data; and storing the sample data into a corresponding theme width table, wherein the theme width table is a data table of the second hive database.
Optionally, in a sixth implementation manner of the second aspect of the present invention, the processing unit further includes: the second extraction subunit is used for acquiring sample data corresponding to the newly added object when detecting that the newly added object exists, extracting a label according to the sample data corresponding to the newly added object, and adding the newly added object to the target object; the second clustering subunit is used for clustering the newly added fields to obtain newly added attribute tags of the target object when detecting that the newly added fields exist in the sample data of the target object and the newly added field values are tag data to be extracted; and the optimizing subunit is used for optimizing the attribute tag of the target object through a preset timing task, wherein the optimizing includes setting the attribute tag of the target object to be a null value or deleting the attribute tag of the target object.
A third aspect of the invention provides a user portrait label modeling and analysis device comprising: a memory having instructions stored therein and at least one processor, the memory and the at least one processor interconnected by a line; the at least one processor invokes the instructions in the memory to cause the user representation tag modeling and analysis device to perform the method of the first aspect described above.
A fourth aspect of the present invention provides a computer-readable storage medium having stored therein instructions which, when run on a computer, cause the computer to perform the method of the first aspect described above.
According to the technical scheme, the invention has the following advantages:
in the technical scheme provided by the invention, data are extracted from a preset data source through a preset mode to obtain original data, wherein the preset mode comprises a full quantity extraction mode and an increment extraction mode; cleaning and combining the original data to obtain sample data; extracting multidimensional target features from the sample data, and processing the multidimensional target features to obtain attribute tags of a target object, wherein the target object and the attribute tags of the target object form a tag system of the target object, and the target object comprises a vehicle, an interactive object, a user, accessory working hours, an answer and a road in a preset portrait system; constructing a target model according to a label system of the target object, wherein the label system is used for indicating that the corresponding target object is described according to the multi-dimensional target characteristics; and carrying out data mining analysis on the target service through the target model to obtain an analysis result, wherein the target service is determined in advance according to actual service requirements. In the embodiment of the invention, the efficiency of constructing the user portrait is improved and the accuracy and efficiency of labeling the target user are improved by constructing the label system and the target model of the user portrait.
Drawings
FIG. 1 is a schematic diagram of one embodiment of a user representation tag modeling and analysis method in an embodiment of the invention;
FIG. 2 is a schematic diagram of another embodiment of a user representation tag modeling and analysis method in an embodiment of the invention;
FIG. 3 is a schematic diagram of an embodiment of a user representation tag modeling and analysis apparatus in an embodiment of the invention;
FIG. 4 is a schematic diagram of another embodiment of a user representation tag modeling and analysis apparatus in an embodiment of the invention;
FIG. 5 is a schematic diagram of an embodiment of a user portrait label modeling and analysis device in an embodiment of the present invention.
Detailed Description
The embodiment of the invention provides a user portrait label modeling and analyzing method, a device, equipment and a storage medium, which are used for improving the efficiency of constructing a user portrait and the accuracy and efficiency of labeling a target user by constructing a label system and a target model of the user portrait.
In order to make the technical field of the invention better understand the scheme of the invention, the embodiment of the invention will be described in conjunction with the attached drawings in the embodiment of the invention.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It will be appreciated that the data so used may be interchanged under appropriate circumstances such that the embodiments described herein may be practiced otherwise than as specifically illustrated or described herein. Furthermore, the terms "comprises," "comprising," or "having," and any variations thereof, are intended to cover non-exclusive inclusions, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
For ease of understanding, a detailed flow of an embodiment of the present invention is described below, with reference to fig. 1, where an embodiment of a user portrait tag modeling and analysis method in an embodiment of the present invention includes:
101. extracting data from a preset data source through a preset mode to obtain original data, wherein the preset mode comprises a full quantity extraction mode and an increment extraction mode;
the server extracts data from a preset data source through a preset mode to obtain original data, wherein the preset mode comprises a full quantity extraction mode and an increment extraction mode, and the preset data source comprises a service database and a service file. Specifically, the server extracts metadata from the service database and the service file through a preset data collection task, and stores the extracted metadata into a first hive database in a unified manner, wherein the first hive database is deployed in a Hadoop cluster in advance, and the data stored in the first hive database is the metadata extracted by the server according to the corresponding service storage.
It should be noted that, the Hadoop provides a distributed storage system for storing mass data, and meanwhile, the Hadoop is widely applied in big data processing application and benefits from the natural advantages of the Hadoop in the aspects of data extraction, deformation and loading, and the MapReduce function of the Hadoop realizes that a single task is broken, the broken task is sent to a plurality of nodes, and then the broken task is loaded into a data warehouse in the form of a single data set.
102. Cleaning and combining original data to obtain sample data;
and the server cleans and merges the original data to obtain sample data. Specifically, the server determines a preset service theme; the server cleans and combines the extracted original data according to a preset service theme to obtain sample data; the server warehouses the sample data to a second hive database. For example, the preset business theme comprises an underwriting theme and a claim settlement theme, wherein the underwriting theme is divided into an insurance policy theme, a insurance application theme and a quotation theme, the claim settlement theme is divided into an investigation theme, a loss settlement theme and a claim settlement theme, the server extracts corresponding original data according to the contracting theme and the claim settlement theme, cleans and merges the original data, and stores sample data according to the preset business theme.
It should be noted that hive is a data warehouse tool based on Hadoop, and can map a structured data file into a database table and provide an object-oriented query language HQL query function.
103. Extracting multidimensional target features from the sample data, and processing the multidimensional target features to obtain attribute tags of a target object, wherein the target object and the attribute tags of the target object form a tag system of the target object, and the target object comprises a vehicle, an interactive object, a user, accessory working hours, an claim and a road in a preset portrait system;
the server extracts multi-dimensional target features from the sample data and processes the multi-dimensional target features to obtain attribute tags of the target object, the target object and the attribute tags of the target object form a tag system of the target object, and the target object comprises vehicles, interactive objects, users, accessory working hours, claims and roads in a preset portrait system. Specifically, the server determines a target object and preset dimensional characteristics; the server extracts multi-dimensional target features from the sample data according to the target object and the preset dimension features, and carries out clustering processing on the multi-dimensional target features to obtain attribute tags of the target object, wherein the attribute tags are named according to a preset tag rule; the server sets the target object and the attribute tag of the target object as a tag system of the target object.
It should be noted that basic attributes and usage of the vehicle may affect risk and value evaluation of the vehicle; the basic attribute and the behavior attribute of the client are helpful for recommending service for the user; the fitting man-hour is to take pictures from the two aspects of the whole replacement of the fitting or the maintenance man-hour; the claim picture comprises accident information, risks, cost and the like; the attributes of the road, the road conditions and the like have important significance for the risk assessment of vehicle driving.
104. Constructing a target model according to a label system of a target object, wherein the label system is used for indicating the target object corresponding to the multi-dimensional target feature description;
and the server constructs a target model according to a label system of the target object, wherein the label system is used for indicating the target object corresponding to the multi-dimensional target feature description. Specifically, the server manages and expands the attribute tags of the target object according to the four-level classification principle of the target object, the main classification principle, the secondary classification principle and the secondary classification principle, selects the attribute tags of the target object from the corresponding classifications according to a specific requirement scene when generating the target model, for example, when pricing modeling is performed according to two objects of vehicle and accessory working hours, the tags are read from tag tables of the two objects of the vehicle and the accessory working hours, and then the pricing model is created.
It can be understood that the primary classification, the secondary classification and the secondary classification are label classifications preset according to actual business requirements, and are not limited herein, and the preset label rules are named according to the attribute labels of the target object, the primary classification, the secondary classification and the target object.
105. And carrying out data mining analysis on the target service through the target model to obtain an analysis result, wherein the target service is determined in advance according to actual service requirements.
The method comprises the steps that a server carries out data mining analysis on target services through a target model to obtain an analysis result, the target services are determined in advance according to actual service requirements, for example, for vehicles and clients which have expired for renewal, the server extracts renewal labels from a vehicle label library and a client label library and constructs a renewal model, the server calculates the renewal difficulty degree of users to be renewed according to the renewal model to obtain a renewal score, divides renewal types for the users to be renewed according to the renewal score and outputs the renewal labels, wherein the renewal types comprise extremely easy renewal type, easier renewal type, general renewal type, difficult renewal type and extremely difficult renewal type, and the renewal labels are used for indicating and describing factors top5 influencing the renewal rate of the clients; and the server recommends related services to the target user according to the renewal type and the renewal label, wherein the renewal type and the renewal label are analysis results.
In the embodiment of the invention, the efficiency of constructing the user portrait is improved and the accuracy and efficiency of labeling the target user are improved by constructing the label system and the target model of the user portrait.
Referring to FIG. 2, another embodiment of a method for modeling and analyzing a user portrait label according to an embodiment of the present invention includes:
201. extracting data from a preset data source to a first hive database through a preset mode to obtain original data, wherein the preset mode comprises a full quantity extraction mode and an increment extraction mode;
the server extracts data from a preset data source to a first hive database through a preset mode to obtain original data, wherein the preset mode comprises a full quantity extraction mode and an increment extraction mode. Specifically, the server reads configuration information, and the server configuration information is used for indicating that the preset data source is from a service database or a service file; the server extracts the static metadata in the service database or the data in the service file in full according to the configuration information; the server performs incremental extraction on the dynamic metadata in the service database according to the configuration information, further, the server sets an incremental extraction period according to the updating frequency of the metadata in the service database, for example, for a policy service with transactions every day, the server performs daily extraction on the dynamic metadata through a timing task, and for an exchange rate service maintained monthly, the server performs monthly extraction on the dynamic metadata through the timing task, and the specific details are not limited herein; the server performs data verification on the extracted metadata according to a preset data format, for example, when the metadata is imported through a file for full extraction, the server verifies whether the format of the imported file meets a preset import rule or not, or the server detects whether actual data corresponding to a date format field of the metadata is a non-date character string or not, and the specific details are not limited herein; the server conducts redundancy processing on the checked data and writes the data after the redundancy processing into the first hive database to obtain original data, wherein the redundancy processing mainly refers to removing repeated data, particularly when the metadata increment is updated, the extracted metadata is stored in the first hive database as historical data, and the server deletes the corresponding historical data first and writes the updated metadata.
It should be noted that the full extraction refers to extracting static metadata of a table or a view in a preset data source without moving, and converting the static metadata into data in a preset format; incremental extraction refers to extracting only new or modified dynamic metadata from a table to be extracted.
202. Cleaning and combining the original data to obtain sample data, and warehousing the sample data into a second hive database;
and the server cleans and merges the original data to obtain sample data, and the sample data is put into a second hive database. Specifically, when dirty data is detected to exist in original data, the server cleans the dirty data, for example, for data imported by a file, the server determines whether a target date field value is greater than a preset threshold value; if the target date field value is larger than a preset threshold value, the server determines that dirty data exists in the extracted metadata; the server cleans the dirty data; the server merges the cleaned data with the same data structure according to the preset service theme to obtain sample data, and further, the server merges and processes the data of different data tables in the first hive database according to the preset service theme to obtain the sample data, wherein the data cleaning and merging process can be carried out simultaneously with the data extraction, and the field extraction is carried out on the data with various sources and disorder according to the preset service theme to obtain the sample data; the server stores sample data into a corresponding theme wide table, wherein the theme wide table is a data table of a second hive database, that is, the server organizes data of a first hive database to form a data mart layer corresponding to a preset service theme, and for the same theme wide table, the data may come from different sources, for example, for vehicle information, the data includes information recorded in a policy after successful application and information recorded in a price enquiry list after successful non-application, so that the server merges data with the same structure but different sources, writes the merged data into the vehicle information theme wide table, and provides the merged data as the same data source for downstream services.
It should be noted that the first hive database and the second hive database are pre-deployed in a Hadoop cluster, the Hadoop provides a distributed storage system for storing mass data, the MapReduce of the Hadoop is used for calculating the mass data and loading the mass data into a data warehouse, and the hive is a data warehouse tool based on the Hadoop, can map a structured data file into a database table, and provides an object-oriented query language HQL query function.
203. Reading corresponding sample data from a second hive database according to a target object of a preset portrait system, wherein the target object comprises a vehicle, an interactive object, a user, accessory working hours, a claim and a road;
the server reads corresponding sample data from the second hive database according to a target object of the preset image system, wherein the target object comprises a vehicle, an interactive object, a user, an accessory man-hour, a claim and a road, and specifically, the server reads a unique identifier of the target object from a preset data table, and the target object comprises a vehicle, an interactive object, a user, an accessory man-hour, a claim and a road in the preset image system, wherein the preset data table stores the unique identifier of the target object and a name of the target object, for example, the server reads the unique identifiers and names corresponding to the vehicle, the interactive object, the user, the accessory man-hour, the claim and the road in the preset image system from the preset data table, and the unique identifiers and names are (ID _1, vehicle), (ID _2, inter _ object), (ID _3, user), (ID _4, task _ time), (ID _5, close _ setting) and (ID _6, the unique identification is a unique mark for identifying the target object, the server can be set according to a universal unique identification code or a globally unique incremental identification, and is not limited herein, and the server presets the unique identification and the name of the target object and stores the unique identification and the name in a preset data table; and the server reads corresponding sample data from the second hive database according to the unique identifier of the target object and the object-oriented query language HQL grammar rule.
It can be understood that the target object and the theme width table have a mapping relationship, the server associates with the corresponding theme width table according to the unique identifier of the target object, and the server queries and obtains the theme width table of the sample data storage corresponding to each target object according to the unique identifier of the target object; the server reads corresponding sample data from the theme width table of the corresponding sample data storage, for example, for the unique identifier of the interactive object being ID _2, the server determines the name of the theme width table as inter _ object from the mapping relation table, and the server reads the sample data of the interactive object from the inter _ object table.
204. Extracting multi-dimensional target features from the read sample data according to a preset algorithm;
and the server extracts multi-dimensional target features from the read sample data according to a preset algorithm. The server extracts a first feature from the read sample data according to a preset feature dimension; the server processes the read sample data through the trained model to obtain a second characteristic; and the server combines the first characteristic and the second characteristic to obtain a multi-dimensional target characteristic. On one hand, the preset feature dimension integrates features from the application angle, for example, the features influencing the renewal rate are extracted according to the business requirements and business experiences; on the other hand, features are sorted from scene contacts, for example, by analyzing data generated by customers in the pricing process and extracting features.
It should be noted that the multi-dimensional target features include basic features and behavior features, and the basic features are natural attribute descriptions of the target object, such as gender and age of the user; behavioral characteristics are characteristics generated by the behavior of the target object, for example, characteristics generated by the application or claim process.
205. Performing clustering analysis on the multidimensional target characteristics according to a k-means clustering algorithm to obtain attribute labels of target objects, wherein the attribute labels are named according to a preset label rule;
the server carries out clustering analysis on the multidimensional target features according to a k-means clustering algorithm to obtain attribute labels of the target objects, and the attribute labels are named according to preset label rules, wherein the preset label rules refer to that the server names the target objects according to the attribute labels of the target objects, main classification, secondary classification and secondary classification, the main classification, the secondary classification and the secondary classification are label classifications preset according to actual service requirements, and the specific details are not limited here.
It should be noted that the K-means clustering algorithm is a clustering analysis algorithm for iterative solution, and the steps are that K objects are randomly selected as initial clustering centers, then the distance between each object and each seed clustering center is calculated, and each object is assigned to the closest clustering center; the cluster centers and the objects assigned to them represent a cluster; each sample is allocated, and the clustering center of the cluster is recalculated according to the existing object in the cluster; this process will be repeated until some termination condition is met. Wherein the termination condition may be that no (or minimum number) objects are reassigned to different clusters, no (or minimum number) cluster centers are changed again, and the sum of squared errors is locally minimal.
206. Setting the target object and the attribute tag of the target object as a tag system of the target object, and storing the tag system of the target object into a second hive database;
the server sets the target object and the attribute label of the target object as a label system of the target object, stores the label system of the target object into a second hive database, and specifically manages and expands the attribute label of the target object according to the principles of the target object, a main classification, a secondary classification and a secondary classification; the server logically divides the target object and the attribute tags of the target object into separate libraries, such as a vehicle tag library, a customer tag library, and the like. When storing the attribute tag of the target object, the server can adopt a first storage mode and a second storage mode, wherein the first storage mode refers to that the server stores the target object, the primary classification and the secondary classification in a dividing way through an independent data table of a second hive database, and sets the attribute tags of the secondary classification and the target object as specific tag fields; the second storage mode is that the server sets independent data tables in a second hive database to store the target object, the main classification, the secondary classification and the secondary classification, and sets the attribute tag of the target object as a field to store. It should be noted that the specific storage principle requires that the server performs selection after judgment according to the actual scene, the server judges whether the attribute tag data under the classification of the target object is greater than a preset value, and if the tag data under the classification of the target object is greater than the preset value, a second storage mode is adopted; and if the classified label data is not greater than the preset value, adopting a first storage mode. For example, if the classification of "vehicle-use information-use vehicle" exceeds 300 tags, the server uses a second storage mode, the server divides the data table according to the classification of fine granularity, and the comparison relationship between the attribute tag classification and the final field can be managed through the system.
For example, for the premium scale business of vehicles, the label data storage hierarchy relationship is divided into vehicle, usage information, vehicle usage, underwriting and premium scale according to the target object, the primary classification, the secondary classification and the attribute label of the target object. When the server stores the vehicle, the use information and the vehicle in a first storage mode, the vehicle, the use information and the vehicle are named as a vechiel _ usetag _ use _ attr data table, and the insurance acceptance and premium scale is set as a final field named as ply _ prev _ tag; when the server adopts the second storage mode for storage, the data table of the vehicle, the use information, the vehicle utilization and the insurance acceptance is named as vecile _ useljply _ attr, and the insurance premium size is set as a final field named as prem _ tag.
Optionally, when detecting that a newly added object exists, the server determines sample data corresponding to the newly added object, extracts a tag according to the sample data corresponding to the newly added object, and adds the newly added object to the target object; when detecting that a newly added field exists in sample data of a target object and the newly added field value is label data to be extracted, the server carries out clustering processing on the newly added field to obtain a newly added attribute label of the target object; and the server carries out optimization processing on the attribute tag of the target object through a preset timing task, wherein the optimization comprises the step of setting the attribute tag of the target object to be a null value or deleting the attribute tag of the target object.
207. Constructing a target model according to a label system of a target object, wherein the label system is used for indicating the target object corresponding to the multi-dimensional target feature description;
and the server constructs a target model according to a label system of the target object, wherein the label system is used for indicating the target object corresponding to the multi-dimensional target feature description. Specifically, the server reads a tag system for extracting a target object from the second hive database according to preset target services to obtain an attribute tag, wherein the tag system is used for indicating the target object corresponding to the multi-dimensional target feature description; dividing the attribute label into training sample data and reference sample data; and training the preset model according to the training sample data and the reference sample data to obtain a target model, wherein the preset model comprises a deep neural network model.
It can be understood that the server manages and expands the attribute tags of the target object according to the four-level classification principle of the target object, the main classification principle, the secondary classification principle and the secondary classification principle, selects the attribute tags of the target object from the corresponding classification according to a specific requirement scene when generating the target model, for example, when pricing modeling is performed according to two objects of vehicle and accessory working hours, the tags are read from the tag tables of the two objects of the vehicle and accessory working hours, and then the pricing model is created.
208. And carrying out data mining analysis on the target service through the target model to obtain an analysis result, wherein the target service is determined in advance according to actual service requirements.
The server carries out data mining analysis on target services through a target model to obtain an analysis result, the target services are determined in advance according to actual service requirements, the data mining analysis comprises prediction, the prediction is that historical data is used for finding out a change rule, a model is built, the model is used for predicting the type and the characteristics of future data, the prediction focuses on precision and uncertainty, and measurement is usually carried out by means of prediction variance. For example, for vehicles and clients which are due for renewal, the server extracts renewal labels from the vehicle label library and the client label library and constructs a renewal model, the server calculates the renewal difficulty of the users to be renewed according to the renewal model to obtain a renewal score, divides renewal types for the users to be renewed according to the renewal score, and outputs the renewal labels, wherein the renewal types comprise extremely easy renewal type, relatively easy renewal type, general renewal type, difficult renewal type and extremely difficult renewal type, and the renewal labels are used for indicating and describing a factor top5 influencing the renewal rate of the clients; and the server recommends related services to the target user according to the renewal type and the renewal label, wherein the renewal type and the renewal label are analysis results.
In the embodiment of the invention, the efficiency of constructing the user portrait is improved and the accuracy and efficiency of labeling the target user are improved by constructing the label system and the target model of the user portrait.
With reference to fig. 3, a user portrait label modeling and analyzing apparatus according to an embodiment of the present invention is described above, and an embodiment of a user portrait label modeling and analyzing apparatus according to an embodiment of the present invention includes:
an extraction unit 301, configured to extract data from a preset data source through a preset manner to obtain original data, where the preset manner includes a full extraction manner and an incremental extraction manner;
a cleaning and merging unit 302, configured to clean and merge original data to obtain sample data;
the processing unit 303 is configured to extract multidimensional target features from the sample data, and process the multidimensional target features to obtain attribute tags of a target object, where the target object and the attribute tags of the target object form a tag system of the target object, and the target object includes a vehicle, an interactive object, a user, accessory man-hours, an answer and a road in a preset portrait system;
a building unit 304, configured to build a target model according to a tag system of a target object, where the tag system is used to indicate that a corresponding target object is described according to a multi-dimensional target feature;
and the analysis unit 305 is configured to perform data mining analysis on the target service through the target model to obtain an analysis result, where the target service is determined in advance according to actual service requirements.
In the embodiment of the invention, the efficiency of constructing the user portrait is improved and the accuracy and efficiency of labeling the target user are improved by constructing the label system and the target model of the user portrait.
Referring to FIG. 4, another embodiment of a user portrait tag modeling and analysis apparatus according to an embodiment of the present invention includes:
an extraction unit 301, configured to extract data from a preset data source through a preset manner to obtain original data, where the preset manner includes a full extraction manner and an incremental extraction manner;
a cleaning and merging unit 302, configured to clean and merge original data to obtain sample data;
the processing unit 303 is configured to extract multidimensional target features from the sample data, and process the multidimensional target features to obtain attribute tags of a target object, where the target object and the attribute tags of the target object form a tag system of the target object, and the target object includes a vehicle, an interactive object, a user, accessory man-hours, an answer and a road in a preset portrait system;
a building unit 304, configured to build a target model according to a tag system of a target object, where the tag system is used to indicate that a corresponding target object is described according to a multi-dimensional target feature;
and the analysis unit 305 is configured to perform data mining analysis on the target service through the target model to obtain an analysis result, where the target service is determined in advance according to actual service requirements.
Optionally, the processing unit 303 may further include:
a determination subunit 3031, configured to read a unique identifier of a target object from a preset data table, where the target object includes a vehicle, an interactive object, a user, an accessory hour, an indemnity, and a road in a preset portrait system;
the reading sub-unit 3032 is configured to read corresponding sample data from the hive database of the second configuration unit according to the unique identifier of the target object and the object-oriented query language HQL grammar rule;
a first extraction subunit 3033, configured to extract a multidimensional target feature from the read sample data according to a preset algorithm;
the first clustering subunit 3034 is configured to perform clustering analysis on the multidimensional target features according to a k-means clustering algorithm to obtain attribute tags of the target objects, where the attribute tags are named according to a preset tag rule;
the setting subunit 3035 is configured to set the target object and the attribute tag of the target object as a tag system of the target object, and store the tag system of the target object in the second hive database.
Optionally, the first extraction subunit 3033 may be further specifically configured to:
extracting a first feature from the read sample data according to a preset feature dimension;
processing the read sample data through the trained model to obtain a second characteristic;
and combining the first characteristic and the second characteristic to obtain the multi-dimensional target characteristic.
Optionally, the constructing unit 304 may be further specifically configured to:
extracting a label system of the target object from the second hive database according to preset target services to obtain an attribute label, wherein the label system is used for indicating the corresponding target object according to multi-dimensional target feature description;
dividing the attribute label into training sample data and reference sample data;
and training the preset model according to the training sample data and the reference sample data to obtain a target model, wherein the preset model comprises a deep neural network model.
Optionally, the extracting unit 301 may be further specifically configured to:
reading configuration information, wherein the configuration information is used for indicating that a preset data source is from a service database or a service file;
performing full extraction on static metadata in a service database or data in a service file according to the configuration information;
performing incremental extraction on the dynamic metadata in the service database according to the configuration information;
performing data verification on the extracted metadata according to a preset data format;
and performing redundancy processing on the checked data, and writing the data subjected to the redundancy processing into a first hive database to obtain original data.
Optionally, the cleaning and merging unit 302 may be further specifically configured to:
when dirty data exist in the original data, cleaning the dirty data;
merging the data with the same data structure after being cleaned according to a preset service theme to obtain sample data;
and storing the sample data into a corresponding theme width table, wherein the theme width table is a data table of the second hive database.
Optionally, the processing unit 303 may further include:
a second extracting subunit 3036, configured to, when detecting that a new object exists, obtain sample data corresponding to the new object, extract a tag according to the sample data corresponding to the new object, and add the new object to the target object;
a second clustering subunit 3037, configured to, when detecting that a newly added field exists in sample data of the target object and the newly added field value is tag data to be extracted, perform clustering on the newly added field to obtain a newly added attribute tag of the target object;
an optimizing subunit 3038, configured to perform optimization processing on the attribute tag of the target object through a preset timing task, where the optimization processing includes setting the attribute tag of the target object to be null or deleting the attribute tag of the target object.
In the embodiment of the invention, the efficiency of constructing the user portrait is improved and the accuracy and efficiency of labeling the target user are improved by constructing the label system and the target model of the user portrait.
Fig. 3 and 4 describe the user portrait label modeling and analyzing apparatus in the embodiment of the present invention in detail from the perspective of the modular functional entity, and the user portrait label modeling and analyzing apparatus in the embodiment of the present invention is described in detail from the perspective of hardware processing.
FIG. 5 is a schematic diagram of a user representation tag modeling and analyzing apparatus 500 according to an embodiment of the present invention, which may include one or more processors (CPUs) 501 (e.g., one or more processors) and a memory 509, one or more storage media 508 (e.g., one or more mass storage devices) for storing applications 507 or data 506, and may generate relatively large differences according to configuration or performance. Memory 509 and storage medium 508 may be, among other things, transient storage or persistent storage. The program stored on storage medium 508 may include one or more modules (not shown), each of which may include a series of instruction operations in a device for modeling and analyzing user representation tags. Still further, the processor 501 may be configured to communicate with a storage medium 508 to execute a series of instruction operations in the storage medium 508 on the user representation tag modeling and analysis device 500.
User representation tag modeling and analysis apparatus 500 may also include one or more power supplies 502, one or more wired or wireless network interfaces 503, one or more input-output interfaces 504, and/or one or more operating systems 505, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, and the like. Those skilled in the art will appreciate that the user representation tag modeling and analysis device configuration shown in FIG. 5 does not constitute a limitation of user representation tag modeling and analysis devices, and may include more or fewer components than those shown, or some components in combination, or a different arrangement of components.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the embodiments provided in the present invention, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A user portrait label modeling and analysis method, comprising:
extracting data from a preset data source through a preset mode to obtain original data, wherein the preset mode comprises a full quantity extraction mode and an increment extraction mode;
cleaning and combining the original data to obtain sample data;
extracting multidimensional target features from the sample data, and processing the multidimensional target features to obtain attribute tags of a target object, wherein the target object and the attribute tags of the target object form a tag system of the target object, and the target object comprises a vehicle, an interactive object, a user, accessory working hours, an answer and a road in a preset portrait system;
constructing a target model according to a label system of the target object, wherein the label system is used for indicating that the corresponding target object is described according to the multi-dimensional target characteristics;
and carrying out data mining analysis on the target service through the target model to obtain an analysis result, wherein the target service is determined in advance according to actual service requirements.
2. The user portrait label modeling and analysis method of claim 1, wherein the extracting multi-dimensional target features from the sample data and processing the multi-dimensional target features to obtain attribute labels of the target object, the target object and the attribute labels of the target object form a label system of the target object, and the target object includes a vehicle, an interactive object, a user, an accessory man-hour, a claim and a road in a preset portrait system, and the method includes:
reading a unique identifier of a target object from a preset data table, wherein the target object comprises a vehicle, an interactive object, a user, accessory working hours, an indemnity and a road in a preset portrait system;
reading corresponding sample data from a second configuration unit hive database according to the unique identification of the target object and an object-oriented query language HQL grammar rule;
extracting multi-dimensional target features from the read sample data according to a preset algorithm;
carrying out clustering analysis on the multidimensional target characteristics according to a k-means clustering algorithm to obtain attribute labels of the target objects, wherein the attribute labels are named according to a preset label rule;
setting the target object and the attribute tag of the target object as a tag system of the target object, and storing the tag system of the target object into the second hive database.
3. The user portrait tag modeling and analysis method of claim 2, wherein the extracting multi-dimensional target features from the read sample data according to a preset algorithm comprises:
extracting a first feature from the read sample data according to a preset feature dimension;
processing the read sample data through the trained model to obtain a second characteristic;
and combining the first characteristic and the second characteristic to obtain a multi-dimensional target characteristic.
4. The user representation label modeling and analysis method of claim 2, wherein said building a target model according to a label hierarchy of the target objects, the label hierarchy for indicating that a corresponding target object is described according to the multi-dimensional target features comprises:
extracting a label system of the target object from the second hive database according to preset target services to obtain an attribute label, wherein the label system is used for indicating the corresponding target object according to the multi-dimensional target feature description;
dividing the attribute label into training sample data and reference sample data;
and training a preset model according to the training sample data and the reference sample data to obtain a target model, wherein the preset model comprises a deep neural network model.
5. The method of modeling and analyzing a user portrait label of claim 1, wherein the extracting data from a preset data source by a preset manner to obtain raw data, the preset manner comprising a full extraction manner and an incremental extraction manner comprises:
reading configuration information, wherein the configuration information is used for indicating that a preset data source is from a service database or a service file;
performing full extraction on static metadata in the service database or data in the service file according to the configuration information;
performing incremental extraction on the dynamic metadata in the service database according to the configuration information;
performing data verification on the extracted metadata according to a preset data format;
and performing redundancy processing on the checked data, and writing the data subjected to the redundancy processing into a first hive database to obtain original data.
6. The user representation tag modeling and analysis method of any of claims 1 to 5, wherein the cleansing and merging the raw data to obtain sample data comprises:
when dirty data are detected to exist in the original data, cleaning the dirty data;
merging the data with the same data structure after being cleaned according to a preset service theme to obtain sample data;
and storing the sample data into a corresponding theme width table, wherein the theme width table is a data table of the second hive database.
7. The user representation tag modeling and analysis method of claim 2, wherein after setting the target object and the attribute tag of the target object as the tag hierarchy of the target object and storing the tag hierarchy of the target object in the second hive database, the user representation tag modeling and analysis method further comprises:
when detecting that a newly added object exists, acquiring sample data corresponding to the newly added object, extracting a label according to the sample data corresponding to the newly added object, and adding the newly added object to the target object;
when detecting that a newly added field exists in the sample data of the target object and the newly added field value is label data to be extracted, performing clustering processing on the newly added field to obtain a newly added attribute label of the target object;
and optimizing the attribute tag of the target object through a preset timing task, wherein the optimizing includes setting the attribute tag of the target object to be a null value or deleting the attribute tag of the target object.
8. A user portrait label modeling and analysis apparatus, comprising:
the extraction unit is used for extracting data from a preset data source in a preset mode to obtain original data, wherein the preset mode comprises a full extraction mode and an increment extraction mode;
the cleaning and merging unit is used for cleaning and merging the original data to obtain sample data;
the processing unit is used for extracting multi-dimensional target features from the sample data and processing the multi-dimensional target features to obtain attribute tags of the target object, the target object and the attribute tags of the target object form a tag system of the target object, and the target object comprises a vehicle, an interactive object, a user, accessory working hours, a claim and a road in a preset portrait system;
the construction unit is used for constructing a target model according to a label system of the target object, and the label system is used for indicating the target object corresponding to the multi-dimensional target feature description;
and the analysis unit is used for carrying out data mining analysis on the target service through the target model to obtain an analysis result, wherein the target service is determined in advance according to the actual service requirement.
9. A user portrait label modeling and analysis apparatus, comprising: a memory having instructions stored therein and at least one processor, the memory and the at least one processor interconnected by a line;
the at least one processor invoking the instructions in the memory to cause the user representation tag modeling and analysis device to perform the method of any of claims 1-7.
10. A computer-readable storage medium having stored thereon a computer program, characterized in that: the computer program realizing the steps of the method according to any one of claims 1-7 when executed by a processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911279395.2A CN111062750A (en) | 2019-12-13 | 2019-12-13 | User portrait label modeling and analyzing method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911279395.2A CN111062750A (en) | 2019-12-13 | 2019-12-13 | User portrait label modeling and analyzing method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111062750A true CN111062750A (en) | 2020-04-24 |
Family
ID=70300785
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911279395.2A Pending CN111062750A (en) | 2019-12-13 | 2019-12-13 | User portrait label modeling and analyzing method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111062750A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111611483A (en) * | 2020-05-11 | 2020-09-01 | 腾讯科技（深圳）有限公司 | Object portrait construction method, device, equipment and storage medium |
| CN111861697A (en) * | 2020-07-02 | 2020-10-30 | 北京睿知图远科技有限公司 | Loan multi-head data-based user portrait generation method and system |
| CN111984898A (en) * | 2020-06-29 | 2020-11-24 | 平安国际智慧城市科技股份有限公司 | Label pushing method and device based on big data, electronic equipment and storage medium |
| CN112381563A (en) * | 2020-11-02 | 2021-02-19 | 上海淇玥信息技术有限公司 | Automatic label generation method and device and electronic equipment |
| CN112860808A (en) * | 2020-12-30 | 2021-05-28 | 深圳市华傲数据技术有限公司 | User portrait analysis method, device, medium and equipment based on data tag |
| CN113051914A (en) * | 2021-04-09 | 2021-06-29 | 淮阴工学院 | Enterprise hidden label extraction method and device based on multi-feature dynamic portrait |
| CN113204714A (en) * | 2021-03-23 | 2021-08-03 | 北京中交兴路信息科技有限公司 | User portrait based task recommendation method and device, storage medium and terminal |
| CN113313344A (en) * | 2021-04-13 | 2021-08-27 | 武汉烽火众智数字技术有限责任公司 | Label system construction method and system fusing multiple modes |
| CN113722305A (en) * | 2021-08-26 | 2021-11-30 | 广东电网有限责任公司 | Analysis application system and method |
| CN114841570A (en) * | 2022-05-07 | 2022-08-02 | 金腾科技信息(深圳)有限公司 | Data processing method, device, equipment and medium for customer relationship management system |
| CN115600840A (en) * | 2022-12-07 | 2023-01-13 | 南方电网数字电网研究院有限公司（Cn） | User group carbon emission dynamic portrait model construction method, device and equipment |
| CN117726359A (en) * | 2024-02-08 | 2024-03-19 | 成都纳宝科技有限公司 | Interactive marketing method, system and equipment |
| CN117726359B (en) * | 2024-02-08 | 2024-04-26 | 成都纳宝科技有限公司 | Interactive marketing method, system and equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108053322A (en) * | 2017-12-15 | 2018-05-18 | 东峡大通（北京）管理咨询有限公司 | The customer investment return evaluation method and system of vehicle |
| CN108764984A (en) * | 2018-05-17 | 2018-11-06 | 国网冀北电力有限公司电力科学研究院 | A kind of power consumer portrait construction method and system based on big data |
| CN109461016A (en) * | 2018-09-10 | 2019-03-12 | 平安科技（深圳）有限公司 | Data methods of marking, device, computer equipment and storage medium |
| CN109858953A (en) * | 2019-01-02 | 2019-06-07 | 深圳壹账通智能科技有限公司 | User's portrait method, apparatus, computer equipment and storage medium |
| CN110428231A (en) * | 2019-08-06 | 2019-11-08 | 重庆仙桃前沿消费行为大数据有限公司 | Administrative information recommended method, device, equipment and readable storage medium storing program for executing |
-
2019
- 2019-12-13 CN CN201911279395.2A patent/CN111062750A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108053322A (en) * | 2017-12-15 | 2018-05-18 | 东峡大通（北京）管理咨询有限公司 | The customer investment return evaluation method and system of vehicle |
| CN108764984A (en) * | 2018-05-17 | 2018-11-06 | 国网冀北电力有限公司电力科学研究院 | A kind of power consumer portrait construction method and system based on big data |
| CN109461016A (en) * | 2018-09-10 | 2019-03-12 | 平安科技（深圳）有限公司 | Data methods of marking, device, computer equipment and storage medium |
| CN109858953A (en) * | 2019-01-02 | 2019-06-07 | 深圳壹账通智能科技有限公司 | User's portrait method, apparatus, computer equipment and storage medium |
| CN110428231A (en) * | 2019-08-06 | 2019-11-08 | 重庆仙桃前沿消费行为大数据有限公司 | Administrative information recommended method, device, equipment and readable storage medium storing program for executing |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111611483B (en) * | 2020-05-11 | 2022-07-22 | 腾讯科技（深圳）有限公司 | Object portrait construction method, device and equipment and storage medium |
| CN111611483A (en) * | 2020-05-11 | 2020-09-01 | 腾讯科技（深圳）有限公司 | Object portrait construction method, device, equipment and storage medium |
| CN111984898A (en) * | 2020-06-29 | 2020-11-24 | 平安国际智慧城市科技股份有限公司 | Label pushing method and device based on big data, electronic equipment and storage medium |
| CN111861697A (en) * | 2020-07-02 | 2020-10-30 | 北京睿知图远科技有限公司 | Loan multi-head data-based user portrait generation method and system |
| CN111861697B (en) * | 2020-07-02 | 2021-05-18 | 北京睿知图远科技有限公司 | Loan multi-head data-based user portrait generation method and system |
| CN112381563A (en) * | 2020-11-02 | 2021-02-19 | 上海淇玥信息技术有限公司 | Automatic label generation method and device and electronic equipment |
| CN112860808A (en) * | 2020-12-30 | 2021-05-28 | 深圳市华傲数据技术有限公司 | User portrait analysis method, device, medium and equipment based on data tag |
| CN113204714A (en) * | 2021-03-23 | 2021-08-03 | 北京中交兴路信息科技有限公司 | User portrait based task recommendation method and device, storage medium and terminal |
| CN113051914A (en) * | 2021-04-09 | 2021-06-29 | 淮阴工学院 | Enterprise hidden label extraction method and device based on multi-feature dynamic portrait |
| CN113313344A (en) * | 2021-04-13 | 2021-08-27 | 武汉烽火众智数字技术有限责任公司 | Label system construction method and system fusing multiple modes |
| CN113722305A (en) * | 2021-08-26 | 2021-11-30 | 广东电网有限责任公司 | Analysis application system and method |
| CN114841570A (en) * | 2022-05-07 | 2022-08-02 | 金腾科技信息(深圳)有限公司 | Data processing method, device, equipment and medium for customer relationship management system |
| CN115600840A (en) * | 2022-12-07 | 2023-01-13 | 南方电网数字电网研究院有限公司（Cn） | User group carbon emission dynamic portrait model construction method, device and equipment |
| CN117726359A (en) * | 2024-02-08 | 2024-03-19 | 成都纳宝科技有限公司 | Interactive marketing method, system and equipment |
| CN117726359B (en) * | 2024-02-08 | 2024-04-26 | 成都纳宝科技有限公司 | Interactive marketing method, system and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111062750A (en) | User portrait label modeling and analyzing method, device, equipment and storage medium | |
| Reinschmidt et al. | Business intelligence certification guide | |
| CN101408885B (en) | Modeling topics using statistical distributions | |
| JP4634685B2 (en) | Data linking system and method using tokens | |
| JP5368665B2 (en) | Expert database forwarded back to link weighted association rules | |
| CN108805598B (en) | Similarity information determination method, server and computer-readable storage medium | |
| US6493723B1 (en) | Method and system for integrating spatial analysis and data mining analysis to ascertain warranty issues associated with transportation products | |
| CN102521233A (en) | Adaptive image retrieval database | |
| CN103733194A (en) | Dynamically organizing cloud computing resources to facilitate discovery | |
| CN103164804A (en) | Personalized method and personalized device of information push | |
| US10977250B1 (en) | Responding to similarity queries using vector dimensionality reduction | |
| US20180246958A1 (en) | Analysis server device, data analysis system, and data analysis method | |
| CN109543040A (en) | Similar account recognition methods and device | |
| CN113032668A (en) | Product recommendation method, device and equipment based on user portrait and storage medium | |
| CN112541111A (en) | Commodity retrieval and commodity recommendation system based on decentralized big data retrieval market | |
| CN109783726A (en) | A kind of real estate big data computation processing method and its system | |
| CN111310032A (en) | Resource recommendation method and device, computer equipment and readable storage medium | |
| CN111581479A (en) | One-stop data processing method and device, storage medium and electronic equipment | |
| CN105359172A (en) | Calculating a probability of a business being delinquent | |
| KR101026544B1 (en) | Method and Apparatus for ranking analysis based on artificial intelligence, and Recording medium thereof | |
| Nordeen | Learn Data Warehousing in 24 Hours | |
| Yang et al. | Discovery of online shopping patterns across websites | |
| US20150142782A1 (en) | Method for associating metadata with images | |
| CN113706253A (en) | Real-time product recommendation method and device, electronic equipment and readable storage medium | |
| CN113298145A (en) | Label filling method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |