CN110610371A - Latent user analysis method, system, and computer-readable storage medium - Google Patents
Latent user analysis method, system, and computer-readable storage medium Download PDFInfo
- Publication number
- CN110610371A CN110610371A CN201810611399.5A CN201810611399A CN110610371A CN 110610371 A CN110610371 A CN 110610371A CN 201810611399 A CN201810611399 A CN 201810611399A CN 110610371 A CN110610371 A CN 110610371A
- Authority
- CN
- China
- Prior art keywords
- user
- template
- script
- feature extraction
- user analysis
- 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
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
- 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
- G06Q30/0202—Market predictions or forecasting for commercial activities
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The disclosure provides a potential user analysis method, a potential user analysis system and a computer readable storage medium, and relates to the technical field of big data. The potential user analysis method comprises the following steps: acquiring a template calling request, wherein the template calling request comprises template identifications and configuration information of each template; acquiring a feature extraction template according to the template identifier, and filling the configuration information to the corresponding position of the feature extraction template to generate a feature extraction script; running a feature extraction script to obtain user features; and determining the consumption possibility of the user based on a predetermined user analysis method according to the user characteristics. By the method, the script can be generated according to the provided template identification and configuration information, the user characteristics are extracted by the script and the user analysis is carried out, the potential user analysis can be carried out on different service lines based on the configuration information without independent modeling, and the realization efficiency of the potential user analysis is improved.
Description
Technical Field
The present disclosure relates to the field of big data technology, and in particular, to a method, system, and computer-readable storage medium for potential user analysis.
Background
With the development of internet technology and logistics industry, online shopping is more popular with users, and besides some physical transactions, some virtual service lines, such as air tickets, hotels, train tickets, movie tickets, mobile phone recharging, game cards and the like, also rely on online transactions. Accurate marketing to users of various service lines can reduce disturbance to users and achieve expected activity effect, so the accurate marketing becomes an important development direction of virtual service lines.
Disclosure of Invention
The inventor finds that modeling in a specific field is often needed in related technologies, different behaviors and attributes of a user in the field are used as characteristic values, and accurate prediction of the user is achieved by utilizing an algorithm model. Although such a method realizes user prediction in a separate field, if each service line is built with such a model, the cost and time are increased, and rapid application deployment is not facilitated.
One object of the present disclosure is to improve the efficiency of implementation of potential user analytics.
According to some aspects of the present disclosure, a potential user analysis method is provided, including: acquiring template calling requests, wherein the template calling requests comprise template identifications and configuration information of each template; acquiring a feature extraction template according to the template identifier, and filling the configuration information to the corresponding position of the feature extraction template to generate a feature extraction script; running a feature extraction script to obtain user features; and determining the consumption possibility of the user based on a predetermined user analysis method according to the user characteristics.
Optionally, generating the feature extraction script comprises: acquiring a feature extraction template according to the template identifier, filling configuration information associated with the template identifier to a preset position, and generating a feature extraction sub-script; and summarizing the feature extraction sub-scripts according to all the called template identifications to generate a feature summarization script, wherein the feature extraction script comprises a feature summarization script and a plurality of feature extraction sub-scripts.
Optionally, the potential user analysis method further includes: after the feature extraction script is generated, uploading the feature extraction script to a script scheduling platform; the script scheduling platform creates a feature processing task and a feature summarizing task and issues a user analysis model so as to obtain user features by calling the user analysis model.
Optionally, the running the feature extraction script to obtain the user feature includes: calling a user analysis model and allocating a unique user analysis model identifier, wherein the user analysis model identifiers allocated each time the user analysis model is called are different; running scripts in the user analysis models with the same identification to obtain user characteristics, and associating the user characteristics with the user analysis model identification; determining the consumption likelihood of the user based on the predetermined user analysis method includes: and loading and running a preset user analysis method, traversing the user characteristics associated with the user analysis model identification, and determining the consumption possibility of the user.
Optionally, the feature extraction template comprises one or more of a browsing record extraction template, a search record extraction template, a coupon record extraction template, a user attribute information extraction template, or an associated category information extraction template.
Optionally, the configuration information includes one or more of target service line identification information, search keyword information, coupon identification information, data extraction time period limitation information, or data extraction threshold limitation information.
Optionally, the potential user analysis method further includes: determining potential users of the predetermined line of business by comparing the user's consumption likelihood to a predetermined threshold; and pushing the service associated with the predetermined service line to the determined potential user.
By the method, the script can be generated according to the provided template identification and configuration information, the user characteristics are extracted by the script and the user analysis is carried out, the potential user analysis can be carried out on different service lines based on the configuration information without independent modeling, and the realization efficiency of the potential user analysis is improved.
According to other aspects of the present disclosure, a potential user analysis system is provided, including: the request acquisition module is configured to acquire template calling requests, and each template calling request comprises a template identifier and configuration information of each template; the script generation module is configured to acquire a feature extraction template according to the template identification template, fill configuration information to corresponding positions of the feature extraction template and generate a feature extraction script; the characteristic acquisition module is configured to run a characteristic extraction script to acquire user characteristics; and the user analysis module is configured to determine the consumption possibility of the user based on a predetermined user analysis method according to the user characteristics.
Optionally, the script generating module includes: the sub-script generating unit is configured to acquire a feature extraction template according to the template identifier, fill configuration information associated with the template identifier to a preset position and generate a feature extraction sub-script; and the summarizing unit is configured to summarize the feature extraction sub-scripts according to all the called template identifications and generate a feature summarizing script, and the feature extraction script comprises a feature summarizing script and a plurality of feature extraction sub-scripts.
Optionally, the potential user analysis system further comprises: the script uploading module is configured to upload the feature extraction script generated by the script generating module to the script scheduling platform; and the analysis model publishing module is positioned on the script scheduling platform and is configured to create a feature processing task and a feature aggregation task and publish the user analysis model so as to obtain the user features by calling the user analysis model.
Optionally, the feature acquisition module is configured to: calling a user analysis model and allocating a unique user analysis model identifier, wherein the user analysis model identifiers allocated when the user analysis model is called each time are different; running scripts in the user analysis models with the same identification to obtain user characteristics, and associating the user characteristics with the user analysis model identification; the user analysis module is configured to load and run a predetermined user analysis method to traverse the user features associated with the user analysis model identification to determine the consumption likelihood of the user.
Optionally, the feature extraction template comprises one or more of a browsing record extraction template, a search record extraction template, a coupon record extraction template, a user attribute information extraction template, or an associated category information extraction template.
Optionally, the configuration information includes one or more of target service line identification information, search keyword information, coupon identification information, data extraction time period limitation information, or data extraction threshold limitation information.
Optionally, the potential user analysis system further comprises: a potential user determination module configured to determine potential users of the predetermined service line by comparing the consumption likelihood of the user with a predetermined threshold; and the business pushing module is configured to push the business associated with the predetermined business line to the determined potential user.
According to still further aspects of the present disclosure, a potential user analysis system is provided, including: a memory; and a processor coupled to the memory, the processor configured to perform any of the potential user analysis methods mentioned above based on instructions stored in the memory.
The user analysis system can generate the script according to the provided template identification and the configuration information, extract the user characteristics by using the script and analyze the user, can analyze the potential users of different service lines based on the configuration information without independent modeling, and improves the realization efficiency of the potential user analysis.
According to still further aspects of the present disclosure, a computer-readable storage medium is proposed, on which computer program instructions are stored, which instructions, when executed by a processor, implement the steps of any one of the potential user analysis methods mentioned above.
By executing the instructions on the computer-readable storage medium, the script can be generated according to the provided template identification and configuration information, the user characteristics are extracted by using the script and the user analysis is carried out, the potential user analysis can be carried out on different service lines based on the configuration information without independent modeling, and the implementation efficiency of the potential user analysis is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this disclosure, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure without limiting the disclosure in any way. In the drawings:
fig. 1 is a flow diagram of some embodiments of a potential user analysis method of the present disclosure.
FIG. 2 is a flow diagram of further embodiments of a potential user analytics method of the present disclosure.
Fig. 3 is a schematic diagram of some embodiments of a potential user analytics system of the present disclosure.
FIG. 4 is a schematic diagram of some embodiments of a script generation module in a potential user analytics system of the present disclosure.
Fig. 5 is a schematic architecture diagram of a potential user analysis system according to the present disclosure.
Fig. 6 is a schematic diagram of further embodiments of a potential user analytics system of the present disclosure.
Fig. 7 is a schematic diagram of further embodiments of potential user analytics systems of the present disclosure.
Detailed Description
The technical solution of the present disclosure is further described in detail below with reference to the accompanying drawings and examples.
A flow diagram of some embodiments of a potential user analytics method of the present disclosure is shown in fig. 1.
In step 101, a template invocation request is obtained, where the template invocation request includes template identifications and configuration information for each template. In some embodiments, the user may specify a plurality of feature extraction templates, each providing configuration information for a respective template. The characteristic extraction template comprises one or more of a browsing record extraction template, a search record extraction template, a coupon record extraction template, a user attribute information extraction template or an associated category information extraction template, and the configuration information comprises one or more of target service line identification information, search keyword information, coupon identification information, data extraction time period limitation information or data extraction threshold value limitation information.
In step 102, a feature extraction template is obtained according to the template identifier, and the configuration information is filled into the corresponding position of the feature extraction template to generate a feature extraction script. In some embodiments, the user specifies the position of the template corresponding to the configuration information when providing the configuration information, such as filling a service line of the feature extraction by specifying a uniform resource locator, filling a time interval of the feature extraction by specifying time, and the like; in other embodiments, the location where the configuration information is filled may be determined by means of character type judgment, for example, the location where the configuration information such as numbers or dates is filled to the time interval location, the location where the information such as address strings is filled to the service line location, and the like.
In step 103, a feature extraction script is run to obtain user features. In some embodiments, a script formed by each feature extraction template may be run separately to obtain a plurality of user features. In some embodiments, the task publishing operation may be performed after the script is formed, and the user feature extraction may be performed automatically or triggered by the user.
In step 104, a consumption possibility of the user is determined based on a predetermined user analysis method according to the user characteristics. In some embodiments, the user feature analysis may be performed by using a related user feature analysis tool, for example, invoking a corresponding function of the xgboost (e.g., tail) to transfer the feature data as a parameter into the xgboost method, i.e., obtaining a score of the estimated potential user likelihood. In some embodiments, the higher the score of the potential user likelihood, the greater the likelihood of consumption.
By the method, the script can be generated according to the provided template identification and configuration information, the user characteristics are extracted by the script and the user analysis is carried out, the potential user analysis can be carried out on different service lines based on the configuration information without independent modeling, and the realization efficiency of the potential user analysis is improved.
In some embodiments, since a plurality of feature extraction templates may be called, in order to ensure that the user features can be systematically analyzed and avoid feature omission, the feature extraction script may be divided into two types, namely, a feature summarizing script and a plurality of feature extraction sub-scripts, and the feature extraction scripts are summarized on the basis of each generated script. For example, a feature extraction template is obtained according to the template identifier, configuration information associated with the template identifier is filled into a preset position to generate a feature extraction sub-script, and then the feature extraction sub-script is summarized according to all the called template identifiers to generate a feature summarizing script. By the method, the features extracted by each script can be spliced to generate centralized and systematic feature information, so that the preset user analysis method can be conveniently called.
In some embodiments, after the feature extraction script is generated, the feature extraction script may be uploaded to a script scheduling platform, and the script scheduling platform creates a feature processing task and a feature summarization task and issues a user analysis model, so as to obtain user features by invoking the user analysis model. By the method, a user can conveniently call the formed script to perform user analysis, and development cost is saved; the workload of developing an independent scheduling user analysis platform is saved, and the deployment efficiency is further improved; the environment of generating and scheduling execution of the script file is separated, and the information safety is improved.
In some embodiments, different users are enabled to invoke the user analysis model at the same time after the user analysis model release is completed. When a user schedules and operates the user analysis model, a user analysis model identification is distributed, and the user analysis model identifications distributed when the user analysis model is called each time are different. And the scheduling platform runs the scripts in the user analysis models with the same identification to obtain the user characteristics, associates the user characteristics with the user analysis model identification, further loads and runs a preset user analysis method, traverses the user characteristics associated with the user analysis model identification and determines the consumption possibility of the user.
By the method, the same or different user analysis models can be simultaneously dispatched by a plurality of users without mutual interference, the possibility of system confusion is reduced, the user analysis efficiency is improved, and the user friendliness is also improved.
A flow diagram of further embodiments of the potential user analytics method of the present disclosure is shown in fig. 2.
In step 201, a template invocation request is obtained, where the template invocation request includes template identifications and configuration information for each template.
In step 202, a feature extraction template is obtained according to the template identifier, and the configuration information is filled into the corresponding position of the feature extraction template to generate a feature extraction script. Relevant data of the required service line can be extracted by means of specifying a data source or limiting a query condition, so that a potential user analysis result of the required service line can be obtained.
For example, when the called feature extraction template is a browsing record extraction template, the preset templates may be:
insert into [ target table name ]
select count (case where [ date ]) refers to sysdate (-3) the n 1else null end) browsed within 3 days, count (case where [ date ] - [ sysdate (-7) the n 1else null end) browsed within 7 days, … … from [ T ] where [ query condition ]
【】 With some being replaceable variables. Let T be a browsing table name of a service line "air ticket", the browsing table includes browsing record information, and can be used as a data source for extracting browsing records, dt be a browsing date, url be a browsing address, and T1 be a target table for loading the extracted user features. If the addresses corresponding to the service line are respectively 'jipiao.jd.com' and 'ijpiao.jd.com', the generated script statement is:
Insert into T1
select count (case where dt > -sysdate (-3) the 1else null end) is browsed within 3 days, count (case where dt > -sysdate (-7) the 1else null end) is browsed within 7 days, … … from T where R < u > r < u'
In some embodiments, it may be determined whether the script was successfully generated. If the success is achieved, the following steps are continuously executed; if the execution is not successful, an error prompt is popped up.
In some embodiments, after the feature extraction sub-scripts are generated, a feature summarization script may be created, and features output by all scripts can be summarized into the same feature table by using the feature summarization script, for example, connected into a wide table by join. Assuming that the browsing record features are recorded in the T1 table, the searching record features are recorded in the T2 table, and other features are recorded in the T3 table, and the target table is T _ ALL, the feature summary script may be:
insert inter T _ alloselect T1, T2, T3 from T1full outer join T2on T1. user id T2. user id full outer join T3on T1. user id T3
In some embodiments, it may be determined whether the feature aggregation script was generated successfully. If the success is achieved, the following steps are continuously executed; if the execution is not successful, an error prompt is popped up.
In step 203, the feature extraction script is uploaded to a script scheduling platform. In some embodiments, it may be determined whether the script was uploaded successfully. If the success is achieved, the following steps are continuously executed; if the execution is not successful, an error prompt is popped up.
In step 204, the script scheduling platform creates a feature processing task and a feature summarization task, issues a user analysis model, and waits for invocation.
In some embodiments, a scheduling platform task creation interface may be called, the feature extraction sub-script that is successfully uploaded is configured as a scheduling task, and information such as task running time and task dependency relationship is created through the interface. In some embodiments, whether the task is created successfully is judged, and if the task is created successfully, the task is waited for being called by a user; and if the creation is not successful, popping up an error prompt.
In some embodiments, a feature processing summary task may also be created, a scheduling platform task creation interface is called, the feature summary script which is successfully uploaded is configured as a scheduling task, and information such as task running time and task dependency relationship is created through the interface, so that the release of the user analysis model is completed.
In step 205, the user analysis model is invoked and assigned a unique user analysis model identification. In some embodiments, the user analysis task may be automatically triggered to execute at a predetermined frequency. In other embodiments, the user analysis task may also be actively performed by user triggering.
In step 206, a script in the user analysis model with the same identification is run, and the user feature is obtained and associated with the user analysis model identification.
In some embodiments, each script task may be executed in a timed manner, and the timing time for executing each feature extraction sub-script may be the same, that is, each task is started at the same time, or may be different, that is, each task has its own trigger time.
After the task execution of each feature extraction submodule is completed, the task of the feature summarization script may begin to be executed. In some embodiments, it may be detected whether each feature extraction sub-script has run to completion. If the operation is finished, the characteristic summarizing script is operated; if not, waiting until the execution is completed.
In step 207, a predetermined user analysis method is loaded and run, traversing the user features associated with the user analysis model identification, and determining the consumption likelihood of the user. In one embodiment, when the extraction of the user features is performed by the script scheduling platform, the script scheduling platform may continue to perform the user analysis operation, or after the script scheduling platform obtains the user features, receive user feature data from the script scheduling platform, and invoke a predetermined user analysis method to analyze the consumption probability of the user in another server or computer.
In step 208, potential subscribers of the predetermined service line are determined by comparing the consumption probability of the subscriber with a predetermined threshold, or the potential subscribers of the predetermined service line are determined according to a predetermined subscriber proportion.
In step 209, a service push associated with the predetermined service line is performed according to the determined potential user.
By the method, the execution error can be timely detected and reported in each link, the user can be helped to know the execution situation and timely find and solve problems, the consumption possibility of the user can be analyzed according to the service line, the potential user aiming at the service line can be determined, the disturbance to the user with low consumption possibility can be avoided, and targeted marketing can be realized.
A schematic diagram of some embodiments of a potential user analytics system of the present disclosure is shown in fig. 3. The request obtaining module 31 can obtain a template calling request including template identification and configuration information for each template. In some embodiments, the user may specify a plurality of feature extraction templates, each providing configuration information for a respective template. The characteristic extraction template comprises one or more of a browsing record extraction template, a search record extraction template, a coupon record extraction template, a user attribute information extraction template or an associated category information extraction template, and the configuration information comprises one or more of target service line identification information, search keyword information, coupon identification information, data extraction time period limitation information or data extraction threshold value limitation information.
The script generating module 32 can obtain the feature extraction template according to the template identifier, and fill the configuration information into the corresponding position of the feature extraction template to generate the feature extraction script. In some embodiments, the user specifies the position of the template corresponding to the configuration information when providing the configuration information, such as filling a service line of the feature extraction by specifying a uniform resource locator, filling a time interval of the feature extraction by specifying time, and the like; in other embodiments, the location of the configuration information padding may be determined by means of character type determination, such as padding the configuration information in the category of numbers or dates to the location of the time interval, padding the information in the category of address strings to the location of the service line, and the like.
The feature obtaining module 33 can run the feature extraction script to obtain the user features. In some embodiments, scripts formed by the feature extraction templates may be run separately to obtain a plurality of user features. In some embodiments, the task publishing operation may be performed after the script is formed, and the user feature extraction may be performed automatically or triggered by the user.
The user analysis module 34 can determine the consumption possibility of the user based on a predetermined user analysis method according to the user characteristics. In some embodiments, the user feature analysis may be performed by using a related user feature analysis tool, for example, invoking a corresponding function of xgboost (e.g., xgboost. train) to transfer the feature data as a parameter into the xgboost method, so as to obtain a score of the estimated potential user likelihood. In some embodiments, the higher the score of the potential user, the greater the likelihood of consumption.
The potential user analysis system can generate a script according to the provided template identification and configuration information, extract user characteristics by using the script and perform user analysis, can perform potential user analysis on different service lines based on the configuration information without independent modeling, and improves the implementation efficiency of the potential user analysis.
In some embodiments, as shown in FIG. 4, script generation module 32 includes a child script generation unit 421 and a summarization unit 422. The script generating module 421 can obtain a feature extraction template according to the template identifier, and fill configuration information associated with the template identifier into a predetermined position to generate a feature extraction sub-script; the summarization unit 422 summarizes the feature extraction sub-scripts according to all the called template identifications to generate feature summarization scripts. The system can splice the features extracted by each script to generate centralized and systematic feature information, and is convenient for a preset user analysis method to call.
In some embodiments, a potential user analytics system includes a script generation apparatus and a script scheduling platform. And calling a feature extraction template at the script scheduling device end by a user to generate a feature extraction script. The script scheduling device may include a script uploading module 35, which is capable of uploading the feature extraction script to the script scheduling platform; the script scheduling platform includes an analysis model publishing module 36 that is capable of creating a feature processing task and a feature summarization task and publishing a user analysis model to obtain user features by invoking the user analysis model.
The system can facilitate a user to call the formed script for user analysis, and saves development cost; the workload of developing an independent scheduling user analysis platform is saved, and the deployment efficiency is further improved; the environment of generating and scheduling execution of the script file is separated, and the information safety is improved.
In some embodiments, the feature obtaining module 33 is capable of assigning a user analysis model identification when the user analysis model is invoked, the user analysis model identification assigned to the user analysis model being different each time the user analysis model is invoked. And the scheduling platform runs the scripts in the user analysis models with the same identification to acquire the user characteristics, associates the user characteristics with the user analysis model identification, further loads and runs a preset user analysis method, traverses the user characteristics associated with the user analysis model identification and determines the consumption possibility of the user.
The system can realize that a plurality of users simultaneously schedule the same or different user analysis models without mutual interference, reduces the possibility of system confusion, improves the efficiency of user analysis and also improves the friendliness to the users.
In some embodiments, as shown in fig. 3, the potential user analysis system may further include a potential user determination module 37 and a traffic push module 38. The subscriber determining module 37 is capable of determining potential subscribers of the predetermined service line by comparing the consumption probability of the subscriber with a predetermined threshold, i.e. may be determined as potential subscribers of the predetermined service line if the consumption probability of the subscriber is greater than the predetermined threshold, or determining potential subscribers of the predetermined service line according to a predetermined subscriber proportion, i.e. may be determined as potential subscribers of the predetermined service line if the consumption probability of the subscriber is within the predetermined subscriber proportion among all subscribers. The service push module 38 can perform service push associated with a predetermined service line according to the determined potential user.
The system can analyze the consumption possibility of the user aiming at the service line, determine potential users aiming at the service line, avoid the disturbance to the user with low consumption possibility and realize targeted marketing.
An architectural schematic diagram of a potential user analytics system of the present disclosure is shown in fig. 5. The model configuration interface 51 provides for user selection of a feature extraction model and provision of configuration information. The model configuration interface 51 may include two functional modules of model management and feature management, and the model management module can perform functions of adding, deleting, modifying, deactivating, activating and the like of the feature extraction template; the feature management module can manage user-provided configuration information, such as URL configuration, search keyword configuration, and the like.
The service layer 52 can perform background processing based on the template identification and configuration information provided by the user. The service layer 52 includes a model and feature management service module and a model engine. The model and feature management service module is capable of performing creation and management of a background model, feature configuration maintenance, and the like according to configuration and call from the model configuration interface 51, corresponding to model management and feature management of the interface layer. The model engine comprises two core functions of model issuing and model operation, and the model issuing function comprises the following steps: splicing the characteristic values of the model into a characteristic extraction script according to a preset rule, uploading the characteristic extraction script to a scheduling system, and automatically creating a scheduling task; the model operation functions include: and running the generated script according to a timing rule, storing the processed feature data into a corresponding feature table, calling a preset user analysis method, inputting the processed feature data for operation, and finally outputting a prediction result for storage. The storage layer 53 can store the generated user analysis model and the feature information extracted after running the script. The configuration data in the storage layer 53 includes various information of the user analysis model, and is stored in various tables, such as a model table, a user attribute table, a browsing configuration table, a searching configuration table, and configuration tables of other preset features, and the system can extract a required template from the configuration data, and in some embodiments, can update the stored template according to the configuration and modification of the user.
In some embodiments, the various entries in the configuration data may be as shown in tables 1-3 below:
TABLE 1 model table
TABLE 2 browse configuration Table
Table 3 search configuration table
The model table records the name of the model, its associated service line and the id of the model, and may also include which configuration tables the model relates to, such as a browsing configuration table, a searching configuration table, a user attribute table, and the like. Because a plurality of models in the system can comprise browsing configuration tables, searching configuration tables and the like, the model to which each table belongs can be determined through the model id, so that the characteristic information is prevented from being polluted due to confusion of system execution, missing extraction of characteristics or addition of wrong characteristic information.
The feature data may include intermediate results and final results generated during analysis and operation of the potential user, for example, the extracted user features may be stored in corresponding feature tables (e.g., the user browsing features may be stored in a browsing feature table, the user search features may be stored in a search feature table, etc.), the features generated by running each script may be summarized, stored in a feature summary table, and the prediction results may be stored in a prediction result table.
In some embodiments, the entries in the feature data may be as shown in tables 4-7 below:
TABLE 4 browse feature List
Table 5 search feature table
TABLE 6 user Attribute Table
TABLE 7 characteristics summary sheet
The generated feature table and the feature summary table can be continuously stored so as to be extracted and called by a user, can be deleted after a preset time, or can be manually triggered by the user and a maintainer to delete, or can be covered and updated when the same model is operated again, so that the occupation of a storage space is reduced, and the stability of the system is improved.
The system can be deployed in three levels of storage, calling and interface, is convenient to use and maintain, and improves the efficiency of operation and maintenance.
A schematic structural diagram of some embodiments of the potential user analytics system of the present disclosure is shown in fig. 6. The potential user analysis system includes a memory 601 and a processor 602. Wherein: the memory 601 may be a magnetic disk, flash memory, or any other non-volatile storage medium. The memory is for storing instructions in the corresponding embodiments of the potential user analysis method above. The processor 602 is coupled to the memory 601 and may be implemented as one or more integrated circuits, such as a microprocessor or microcontroller. The processor 602 is configured to execute instructions stored in the memory, and can perform potential user analysis on different service lines based on the configuration information without separate modeling, thereby improving the efficiency of implementing the potential user analysis.
In some embodiments, as also shown in fig. 7, a potential user analysis system 700 includes a memory 701 and a processor 702. The processor 702 is coupled to the memory 701 by a BUS 703. The potential user analysis system 700 may also be coupled to an external storage device 705 via a storage interface 704 for invoking external data, and may also be coupled to a network or another computer system (not shown) via a network interface 706. And will not be described in detail herein.
In the embodiment, the data instructions are stored in the memory and processed by the processor, so that the potential user analysis can be performed on different service lines based on the configuration information without separately modeling, and the implementation efficiency of the potential user analysis is improved.
In other embodiments, a computer-readable storage medium has stored thereon computer program instructions which, when executed by a processor, implement the steps of the method in the corresponding embodiment of the potential user analysis method. As will be appreciated by one skilled in the art, embodiments of the present disclosure may be provided as a method, apparatus, or computer program product. Accordingly, the present disclosure may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present disclosure may take the form of a computer program product embodied on one or more computer-usable non-transitory storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present disclosure is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Thus far, the present disclosure has been described in detail. Some details that are well known in the art have not been described in order to avoid obscuring the concepts of the present disclosure. It will be fully apparent to those skilled in the art from the foregoing description how to practice the presently disclosed embodiments.
The methods and apparatus of the present disclosure may be implemented in a number of ways. For example, the methods and apparatus of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, firmware. The above-described order for the steps of the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above unless specifically stated otherwise. Further, in some embodiments, the present disclosure may also be embodied as programs recorded in a recording medium, the programs including machine-readable instructions for implementing the methods according to the present disclosure. Thus, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.
Finally, it should be noted that: the above examples are intended only to illustrate the technical solutions of the present disclosure and not to limit them; although the present disclosure has been described in detail with reference to preferred embodiments, those of ordinary skill in the art will understand that: modifications to the specific embodiments of the disclosure or equivalent substitutions for parts of the technical features may still be made; all such modifications are intended to be included within the scope of the claims of this disclosure without departing from the spirit thereof.
Claims (14)
1. A potential user analytics method, comprising:
acquiring a template calling request, wherein the template calling request comprises template identifications and configuration information of each template;
acquiring a feature extraction template according to the template identification, and filling the configuration information to the corresponding position of the feature extraction template to generate a feature extraction script;
running the feature extraction script to obtain the user features;
and determining the consumption possibility of the user based on a predetermined user analysis method according to the user characteristics.
2. The method of claim 1, wherein the generating a feature extraction script comprises:
acquiring the feature extraction template according to the template identification, filling configuration information associated with the template identification to a preset position, and generating a feature extraction sub-script;
and summarizing the feature extraction sub-scripts according to all called template identifications to generate the feature summarizing script, wherein the feature extraction script comprises the feature summarizing script and a plurality of feature extraction sub-scripts.
3. The method of claim 1, further comprising:
after the feature extraction script is generated, uploading the feature extraction script to a script scheduling platform;
the script scheduling platform creates a feature processing task and a feature summarizing task and issues a user analysis model so as to obtain user features by calling the user analysis model.
4. The method of claim 3, wherein,
the step of running the feature extraction script to acquire the user features comprises the following steps:
calling the user analysis model and allocating a unique user analysis model identifier, wherein the user analysis model identifiers allocated to the user analysis model are different each time the user analysis model is called;
running scripts in the user analysis model with the same identification to obtain user characteristics, and associating the user characteristics with the user analysis model identification;
the determining the consumption possibility of the user based on the predetermined user analysis method includes: and loading and running the predetermined user analysis method, traversing the user characteristics associated with the user analysis model identification, and determining the consumption possibility of the user.
5. The method according to any one of claims 1 to 4, wherein the feature extraction template comprises one or more of a browsing record extraction template, a search record extraction template, a coupon record extraction template, a user attribute information extraction template, or an associated category information extraction template;
and/or the presence of a gas in the gas,
the configuration information includes one or more of target service line identification information, search keyword information, coupon identification information, data extraction time period limitation information, or data extraction threshold limitation information.
6. The method of any of claims 1-4, further comprising:
determining potential users of a predetermined line of business by comparing the user's consumption likelihood to a predetermined threshold;
pushing the service associated with the predetermined service line to the determined potential user.
7. A potential user analytics system, comprising:
the request acquisition module is configured to acquire a template calling request, and the template calling request comprises template identifications and configuration information of each template;
the script generation module is configured to acquire a feature extraction template according to the template identification template, fill the configuration information to the corresponding position of the feature extraction template, and generate a feature extraction script;
the characteristic acquisition module is configured to run the characteristic extraction script to acquire user characteristics;
a user analysis module configured to determine a consumption possibility of the user based on a predetermined user analysis method according to the user characteristics.
8. The system of claim 7, wherein the script generation module comprises:
the sub-script generating unit is configured to acquire the feature extraction template according to the template identifier, fill the configuration information associated with the template identifier to a preset position and generate a feature extraction sub-script;
and the summarizing unit is configured to summarize the feature extraction sub-scripts according to all the called template identifications and generate the feature summarizing script, and the feature extraction script comprises the feature summarizing script and a plurality of feature extraction sub-scripts.
9. The system of claim 7, further comprising:
the script uploading module is configured to upload the feature extraction script generated by the script generating module to a script scheduling platform;
and the analysis model issuing module is positioned on the script scheduling platform and is configured to create a feature processing task and a feature summarizing task and issue a user analysis model so as to obtain user features by calling the user analysis model.
10. The system of claim 9, wherein,
the feature acquisition module is configured to:
calling the user analysis model and allocating a unique user analysis model identifier, wherein the user analysis model identifiers allocated to the user analysis model are different each time the user analysis model is called;
running scripts in the user analysis model with the same identification to obtain user characteristics, and associating the user characteristics with the user analysis model identification;
the user analysis module is configured to load and run the predetermined user analysis method to traverse the user features associated with the user analysis model identification to determine a consumption likelihood of the user.
11. The system of any one of claims 7 to 10, wherein the feature extraction template comprises one or more of a browsing record extraction template, a search record extraction template, a coupon record extraction template, a user attribute information extraction template, or an associated category information extraction template;
and/or the presence of a gas in the gas,
the configuration information includes one or more of target service line identification information, search keyword information, coupon identification information, data extraction time period limitation information, or data extraction threshold limitation information.
12. The system of any of claims 7 to 10, further comprising:
a potential user determination module configured to determine potential users of a predetermined line of service by comparing the user's consumption likelihood to a predetermined threshold;
a service pushing module configured to push the service associated with the predetermined service line to the determined potential user.
13. A potential user analytics system, comprising:
a memory; and
a processor coupled to the memory, the processor configured to perform the method of any of claims 1-6 based on instructions stored in the memory.
14. A computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the steps of the method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810611399.5A CN110610371A (en) | 2018-06-14 | 2018-06-14 | Latent user analysis method, system, and computer-readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810611399.5A CN110610371A (en) | 2018-06-14 | 2018-06-14 | Latent user analysis method, system, and computer-readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110610371A true CN110610371A (en) | 2019-12-24 |
Family
ID=68887688
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810611399.5A Pending CN110610371A (en) | 2018-06-14 | 2018-06-14 | Latent user analysis method, system, and computer-readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110610371A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111737405A (en) * | 2020-07-03 | 2020-10-02 | 广州云博互联网技术有限公司 | Image-text material archiving management method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104077332A (en) * | 2013-03-29 | 2014-10-01 | 上海城际互通通信有限公司 | User behavior analysis method based on charging information |
| CN105007171A (en) * | 2015-05-25 | 2015-10-28 | 上海欣方软件有限公司 | User data analysis system and method based on big data in communication field |
| CN106022800A (en) * | 2016-05-16 | 2016-10-12 | 北京百分点信息科技有限公司 | User feature data processing method and device |
| CN107908631A (en) * | 2017-07-25 | 2018-04-13 | 平安科技（深圳）有限公司 | Data processing method, device, storage medium and computer equipment |
-
2018
- 2018-06-14 CN CN201810611399.5A patent/CN110610371A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104077332A (en) * | 2013-03-29 | 2014-10-01 | 上海城际互通通信有限公司 | User behavior analysis method based on charging information |
| CN105007171A (en) * | 2015-05-25 | 2015-10-28 | 上海欣方软件有限公司 | User data analysis system and method based on big data in communication field |
| CN106022800A (en) * | 2016-05-16 | 2016-10-12 | 北京百分点信息科技有限公司 | User feature data processing method and device |
| CN107908631A (en) * | 2017-07-25 | 2018-04-13 | 平安科技（深圳）有限公司 | Data processing method, device, storage medium and computer equipment |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111737405A (en) * | 2020-07-03 | 2020-10-02 | 广州云博互联网技术有限公司 | Image-text material archiving management method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107239335B (en) | Job scheduling system and method for distributed system | |
| CN107925696B (en) | System and method for identifying, indexing and navigating to a depth state of a mobile application | |
| CN109814884A (en) | A kind of method and system carrying out resource management according to game resource type | |
| CN111068328B (en) | Game advertisement configuration form generation method, terminal equipment and medium | |
| CN105488223A (en) | Analyzing and subscribing method and system of business data events | |
| CN104423982B (en) | The processing method and processing equipment of request | |
| CN106873952B (en) | Data processing system and method for mobile terminal webpage development and application device | |
| CN113220657B (en) | Data processing method and device and computer equipment | |
| CN111124480B (en) | Method and device for generating application program package, electronic equipment and storage medium | |
| US20160019266A1 (en) | Query generating method and query generating device | |
| WO2022121971A1 (en) | Front-end project processing method and apparatus, device, management system, and storage medium | |
| CN117271460B (en) | Scientific research digital networking service method and system based on scientific research digital object language relation | |
| WO2019223172A1 (en) | Kettle task scheduling method and system, computer device, and storage medium | |
| CN111859076A (en) | Data crawling method and device, computer equipment and computer readable storage medium | |
| US9595014B1 (en) | System and method for executing workflow instance and modifying same during execution | |
| CN111813803A (en) | Statement block execution plan generation method, device, equipment and storage medium | |
| CN112905158B (en) | Marketing center platform system based on hierarchical series technology | |
| CN112860412B (en) | Service data processing method and device, electronic equipment and storage medium | |
| US20120197938A1 (en) | Search request control apparatus and search request control method | |
| CN110610371A (en) | Latent user analysis method, system, and computer-readable storage medium | |
| CN108897569A (en) | The method for cleaning and computer readable storage medium of iOS engineering discarded record | |
| CN107239399A (en) | For the index generation method of test, device, system and readable storage medium storing program for executing | |
| CN108021688B (en) | A kind of data processing method and device | |
| CN113792026B (en) | Method and device for deploying database script and computer-readable storage medium | |
| CN115857918A (en) | Data processing method and device, electronic equipment and storage medium |
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 |