CN113159213A - Service distribution method, device and equipment - Google Patents

Service distribution method, device and equipment Download PDF

Info

Publication number
CN113159213A
CN113159213A CN202110486442.1A CN202110486442A CN113159213A CN 113159213 A CN113159213 A CN 113159213A CN 202110486442 A CN202110486442 A CN 202110486442A CN 113159213 A CN113159213 A CN 113159213A
Authority
CN
China
Prior art keywords
service
service data
clustering
data
category
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
Application number
CN202110486442.1A
Other languages
Chinese (zh)
Inventor
徐玙萌
罗伟
江子扬
杨寰宇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Industrial and Commercial Bank of China Ltd ICBC
Original Assignee
Industrial and Commercial Bank of China Ltd ICBC
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Industrial and Commercial Bank of China Ltd ICBC filed Critical Industrial and Commercial Bank of China Ltd ICBC
Priority to CN202110486442.1A priority Critical patent/CN113159213A/en
Publication of CN113159213A publication Critical patent/CN113159213A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • 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
    • G06N3/04—Architecture, e.g. interconnection topology
    • 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
    • G06N3/088—Non-supervised learning, e.g. competitive learning
    • 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
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0631—Item recommendations
    • 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
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/02—Banking, e.g. interest calculation or account maintenance

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Marketing (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • General Business, Economics & Management (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Development Economics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Technology Law (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The embodiment of the specification provides a service distribution method, a service distribution device and service distribution equipment, which can be applied to the technical field of artificial intelligence. The method comprises the following steps: acquiring service data of a target user corresponding to at least two service scenes; the service data is used for representing the service processing content of the target user in the service scene; dividing the service data into at least two groups of clustering service data; the clustering service data comprises service data of the same type; the clustering service data corresponds to a preset clustering category; determining the user category of the target user based on the clustering service data and the corresponding preset clustering category; and distributing the recommended service corresponding to the user category to the target user. The method effectively ensures the processing of the multi-view data through data clustering and user category determination. In addition, multi-view data corresponding to the user is fully utilized, the accuracy of service distribution is improved, and the experience of the user in processing the corresponding service is improved.

Description

Service distribution method, device and equipment
Technical Field
The embodiment of the specification relates to the technical field of artificial intelligence, in particular to a service distribution method, a device and equipment.
Background
With the development of various industries, the subdivision degree of the service types is also continuously improved. These services may be services providing corresponding services for users, or services requiring users to process in time, and accordingly, services required to be acquired by different types of users are different. Therefore, services which are possibly acquired by the user are judged in advance according to the relevant information of the user, so that data and resources corresponding to the corresponding services are prepared in advance, the service processing efficiency in the subsequent process can be effectively improved, and the user experience is improved.
At present, when predicting services required by users, users are generally classified by historical service data corresponding to the users, so that the corresponding services are recommended to the users according to different classes. However, as the complexity of the service increases and the number of different fields and scenes increases, the types and application scenes corresponding to the historical service data become more complex, and it is difficult to effectively determine the user category by directly using the multi-view data, so that a suitable service cannot be recommended to the user. Therefore, a method for recommending a proper service to a user based on multi-view data is needed.
Disclosure of Invention
An object of the embodiments of the present specification is to provide a service allocation method, device and apparatus, so as to solve a problem how to recommend a service to a user based on multi-view data to improve user experience.
To solve the foregoing technical problem, an embodiment of the present specification provides a service allocation method, including: acquiring service data of a target user corresponding to at least two service scenes; the service data is used for representing the service processing content of the target user in the service scene; dividing the service data into at least two groups of clustering service data; the clustering service data comprises service data of the same type; the clustering service data corresponds to a preset clustering category; determining the user category of the target user based on the clustering service data and the corresponding preset clustering category; and distributing the recommended service corresponding to the user category to the target user.
An embodiment of this specification further provides a service allocation apparatus, including: the service data acquisition module is used for acquiring service data of a target user corresponding to at least two service scenes; the service data is used for representing the service processing content of the target user in the service scene; the clustering service data dividing module is used for dividing the service data into at least two groups of clustering service data; the clustering service data comprises service data of the same type; the clustering service data corresponds to a preset clustering category; the user category determining module is used for determining the user category of the target user based on the clustering service data and the corresponding preset clustering category; and the recommendation service distribution module is used for distributing the recommendation service corresponding to the user category to the target user.
The embodiment of the present specification further provides a service distribution device, including a memory and a processor; the memory to store computer program instructions; the processor to execute the computer program instructions to implement the steps of: acquiring service data of a target user corresponding to at least two service scenes; the service data is used for representing the service processing content of the target user in the service scene; dividing the service data into at least two groups of clustering service data; the clustering service data comprises service data of the same type; the clustering service data corresponds to a preset clustering category; determining the user category of the target user based on the clustering service data and the corresponding preset clustering category; and distributing the recommended service corresponding to the user category to the target user.
As can be seen from the technical solutions provided by the embodiments of the present specification, after obtaining multi-view service data from different service scenes, the embodiments of the present specification first perform clustering processing on the service data to obtain clustered service data, and determine the category of a user according to the clustered service data, so that a corresponding recommended service can be allocated to the user according to the category of the user. The method effectively ensures the processing of the multi-view data through the data clustering and the user category determining mode. In addition, the multi-view data corresponding to the user is fully utilized, and under the condition that the multi-view data has more information compared with single-view data, the accuracy of service distribution is improved, and the experience of the user in processing the corresponding service is further improved.
Drawings
In order to more clearly illustrate the embodiments of the present specification or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the specification, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a flowchart of a service allocation method according to an embodiment of the present disclosure;
FIG. 2 is a diagram illustrating a multi-user data set according to an embodiment of the present disclosure;
fig. 3 is a schematic diagram of a bank card recommendation scenario according to an embodiment of the present disclosure;
fig. 4 is a schematic flowchart of a recommendation scenario of a bank card according to an embodiment of the present disclosure;
FIG. 5 is a schematic structural diagram of a recommendation system according to an embodiment of the present disclosure;
fig. 6 is a block diagram of a service distribution apparatus according to an embodiment of the present disclosure;
fig. 7 is a block diagram of a service distribution apparatus according to an embodiment of the present disclosure.
Detailed Description
The technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, and not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present specification without any creative effort shall fall within the protection scope of the present specification.
In order to solve the above technical problem, a service allocation method according to an embodiment of the present disclosure is first introduced. The execution subject of the service allocation method may be a service allocation device. The service distribution equipment includes but is not limited to a server, an industrial personal computer, a PC machine and the like. As shown in fig. 1, the service allocation method may include the following implementation steps.
S110: acquiring service data of a target user corresponding to at least two service scenes; the service data is used for representing the service processing content of the target user in the service scene.
The target user may be a user who needs to perform service recommendation, and needs to allocate a corresponding service according to the type of the user, so that the target user can adapt to the allocated service, and the user experience is improved. Preferably, the target is used for the historical record of the processing service, so that the historical service data generated by the processing service can be effectively used. For example, when the target scenario is a bank card business scenario, the target user may be a bank card that has been handled and has performed corresponding business based on the bank card, so that historical business data of the target user on the bank card can be collected.
The traffic scenario may be different scenarios determined based on the currently targeted traffic, thereby serving to distinguish different traffic data. Correspondingly, the formats of the service data corresponding to different service scenarios may be different. For example, in the case that the service is a bank card service, the service scenario may include a bank card transaction scenario, a bank card inquiry service scenario, and the like, and formats of service data obtained respectively are also different. It should be noted that, service data in different formats may also be acquired in the same service scenario, which is not limited to this.
The service data may be data generated in the course of processing a corresponding service. Different service data of the target user can be used for reflecting different types of the target user, so that corresponding services can be recommended for the target user based on operation habits embodied in the service data, and the service processing effect of the user is improved. Specifically, the service data may be basic information of the user, such as data of a learned calendar, a sex, an age, and the like, or history information of the user processing the service, that is, a specific process of processing the service.
In this embodiment, the service data may be multi-view data. The multi-view data may refer to a plurality of data information acquired in different ways in different scenes, features of the data are often different, and included semantic information generally does not have comparability, but the data information also includes more information compared with data of a single view, so that a tag corresponding to a target user can be better determined.
In some embodiments, the business data includes at least one of customer account distribution information, customer transaction amount distribution information, customer transaction counter-party distribution information, customer transaction frequency distribution information, customer consumption merchant distribution information, and customer transaction channel distribution information.
The customer account distribution information may be used to indicate details of the account corresponding to the target user, such as information relating to different bank card accounts of the user in a bank card transaction. The customer transaction amount distribution information may indicate specific amounts consumed by the target user in different services and specific services corresponding to the different consumption amounts. The customer counterparty distribution information may represent detailed information about the target user's transaction object. The customer transaction frequency distribution information may represent transaction times corresponding to transaction records generated by the target user, as well as transaction frequency information generated based on the transaction records. The distribution information of the customer consumption merchants can be obtained by extracting consumption objects in the consumption records of the users, obtaining distribution positions corresponding to the consumption objects and comprehensively obtaining the distribution information of the customer consumption merchants. The distribution information of the customer transaction channel can be the transaction form or payment mode adopted in the transaction process.
It should be noted that the service data is service data in different formats corresponding to the service types of the bank card, and the application range of the service in actual application may be configured with service data in other formats, which is not limited to the above example and is not described herein again.
The specific method for acquiring the service data may be directly extracting from the corresponding log, or may be to store the corresponding service data in advance by using a corresponding data table. In practical application, the service data may also be obtained in other manners, which are not limited to the above examples and are not described herein again.
S120: dividing the service data into at least two groups of clustering service data; the clustering service data comprises service data of the same type; the clustering service data corresponds to a preset clustering category.
Since the service data is multi-view data, information contained in the service data is from different channels and further belongs to different types, and the service data can not be directly utilized to be effectively analyzed, so that the final prediction result lacks certain accuracy. Therefore, after the service data is obtained, the service data can be divided into at least two groups of clustered service data.
The clustering service data may be data obtained by clustering service data, and the same group of clustering service data may include service data of the same type. The clustering operation is an operation of dividing data of the same type into the same category, and in the embodiment of the present specification, after classifying the multi-view service data into a plurality of groups of clustered service data, behaviors of target users can be more effectively analyzed based on the clustered data.
Specifically, the group number of the clustering service data may be predetermined, so that a corresponding clustering process may be set according to the predetermined group number, so as to achieve a final clustering effect.
Using a specific example for illustration, for a data set containing N target users, assuming that the data set contains presentation data corresponding to V views, the data set can be represented as
Figure BDA0003050529460000041
Wherein
Figure BDA0003050529460000042
Representing the data expressed by the ith sample in the vth view, and so on, as shown in FIG. 2, the sample set can be represented as
Figure BDA0003050529460000051
In the form of (1), wherein XvRepresenting the expression data of the sample data in the view v, and
Figure BDA0003050529460000052
(dvis the dimensionality from the v-view data) is the ith data from the v-view. For the data in the data set, the N sample data can be divided into M clusters, so that the data can be effectively processed. In addition, the multi-view data of the same sample can form mutual limitation during clustering, so that noise is suppressed, and the purpose of automatic error correction is achieved.
In some embodiments, in order to more effectively obtain the clustered service data, service data features may be extracted from the service data, and then the service data may be divided into at least two groups of clustered service data based on the service data features. The service data features can be data obtained after feature extraction is performed on the clustered service data, and the feature extraction process is used for generalizing the clustered service data, for example, dimension reduction processing is performed on the clustered service data, so that the data volume can be reduced, influence of excessive redundant data on the processing process is avoided, the subsequent processing efficiency is improved, and the processing result is optimized. The specific feature extraction method may be set according to the actual application, and is not described herein again.
After the service data features are obtained, the service data may be divided into at least two groups of clustered service data based on the service data features. The specific dividing manner may refer to a manner of directly dividing the service data, and is not described herein again.
In some embodiments, different service scenarios may have different weight values, and accordingly, service data corresponding to different service scenarios may also have different weight values.
Specifically, a weight ω may be added to data of different views vvSo that it satisfies
Figure BDA0003050529460000053
With addition of an exponential parameter p for adjusting omegavSparsity of distribution; therefore, under different service scenes, the data use tendency can be better realized, such as: more credit card issuing popularization cooperating with a small merchant platform needs to pay more attention to 'consumption behavior data of a client', and credit card issuing popularization with a lower interest rate needs to pay more attention to 'transaction amount data condition of the client'. Through weighting and weighted sparsity control, the same set of data can be realized, and different clustering (label) results in different service scenes can be realized.
In some embodiments, the service data may be clustered by using a k-means clustering algorithm to obtain at least two groups of clustered service data. The k-means clustering algorithm is an unsupervised clustering algorithm, according to given sample data, based on the distance between the sample data, the sample data are divided into k clusters, so that the distance between the sample data in the clusters is as small as possible, and the distance between the clusters is as large as possible, thereby realizing the clustering of the sample data.
Accordingly, the clustering service data may correspond to a preset clustering category. The preset cluster category may be a category that is preset according to different types of data. Because the clustering service data is data obtained after clustering, the data in the same group generally has higher similarity, and the preset clustering category can be distributed based on the properties of the clustering service data in different groups.
In the present embodiment, the data set χ ═ x to be targeted to N target users1,x2,…,xNThe clustering is M classes, and the corresponding objective function can be
Figure BDA0003050529460000061
Wherein the content of the first and second substances,
Figure BDA0003050529460000062
Figure BDA0003050529460000063
for the representation of the centre point of the kth cluster (label) in the v (data class) view, δikIs a cluster distribution index: when x isiδ when (i-th data) belongs to k-th cluster (label)ik1, otherwise δikEach data can only belong to one cluster, and each target user finally obtains an attribution label, namely, the attribution label is obtained
Figure BDA0003050529460000064
And determining the preset clustering category of the clustering service data based on the label to which the clustering service data belongs. Adding a weight to each feature of the data of a single view
Figure BDA0003050529460000065
So that it satisfies
Figure BDA0003050529460000066
Wherein d isvFeature dimension (number of features contained) of data for the v-th view in order to weight features
Figure BDA0003050529460000067
The distribution of (a) is limited to a certain extent, so that the distribution of (b) is not too sparse, the result is more reliable, and an adjustment parameter beta is introduced to act on l2The validity of the norm limitation is adjusted. Aiming at the objective function, the obtained objective function is minimized, and the obtained objective function is obtainedAnd obtaining the final result which is the final clustering result.
For the specific process of obtaining the objective function, a Weighted Multi-view Clustering algorithm (WMCFS) based on Feature Selection may be used for calculation. View weighting (ω) may be fixed for the above equationv) Feature weighting
Figure BDA0003050529460000068
And cluster distribution (delta)ik) And obtaining a third parameter value by two of the three parameters, and circularly calculating until convergence.
The specific calculation process can be to use the business data of the target user
Figure BDA0003050529460000069
p (weighted sparsity of each type of data), M (number of labels), beta (reliability adjustment parameter), tmax(calculate maximum number of cycles) as an input parameter, weight view (ω)v) Feature weighting
Figure BDA00030505294600000610
And cluster distribution (delta)ik) As output parameters. Initializing the target function to make
Figure BDA00030505294600000611
And
Figure BDA00030505294600000612
t is 0. Thereafter, the utilization of the latest ω is repeatedly performedvAnd
Figure BDA00030505294600000613
calculate deltaikAnd covers the original deltaikA value of (d); using the latest deltaikAnd
Figure BDA00030505294600000614
calculate omegavAnd covers the original omegavA value; using the latest omegavAnd deltaikCalculate out
Figure BDA00030505294600000615
And cover the original
Figure BDA00030505294600000616
A value of (d); adding 1 to the value of t until t equals tmaxOr until the calculation results converge. t is tmaxThe maximum number of cycles may be preset. When the above-described loop process is finished, the final value of the final output parameter value may be determined, and clustering may be completed based on these calculated values.
The above specific calculation process is only an example for explaining the clustering process, and in practical application, the final result may be obtained in other manners, and is not limited to the above example.
S130: and determining the user category of the target user based on the clustering service data and the corresponding preset clustering category.
After the clustering service data is obtained, the user category of the target user can be determined according to the clustering service data. The user category can be used for distinguishing the types of the users, and for the service data of the users, the modes, habits and the like of the users for processing different services can be determined, and then the users are classified according to the differences.
In some embodiments, based on the calculation manner in step S120, when data clustering is performed based on the formula in which the objective function is obtained, corresponding user tags, that is, the preset clustering categories, may also be obtained, and corresponding user categories may also be determined for the user tags. Specifically, the corresponding clustering results can be determined according to the values of the user tags, and the clustering results are used as the final user category. Under the condition that the clustering type is set in advance, the user type determination of the target user can be effectively realized based on the clustering result of the service data.
In some embodiments, the user category may be determined by inputting the clustered service data into a pre-constructed category classification model, and outputting a corresponding user category through the category classification model. The class classification model may be a model obtained by training in advance using corresponding clustering sample data. The classification model may be a classification model, so that classification of the user is achieved by using the clustered service data.
In practical applications, the class classification model may be a mathematical model. The classification model may specifically include a regression model (e.g., a linear regression model, a non-linear regression model, etc.), a neural network model, a support vector machine model, a bayesian model, and the like.
By the class division model, clustering service data can be quickly and effectively utilized, so that a corresponding classification result is output, and service recommendation in subsequent steps is facilitated.
S140: and distributing the recommended service corresponding to the user category to the target user.
After the user category of the target user is determined, the recommended service corresponding to the user category can be allocated to the target user. Specifically, the corresponding recommended service may be set in advance for different user categories. After the user category is determined, the recommendation can be directly carried out according to the set recommended service, so that the service distribution is conveniently and effectively realized, and the service processing experience of the user is guaranteed.
To illustrate by using a specific example, for example, the preset user categories include "low service processing frequency", "medium service processing frequency" and "high service processing frequency", a service with a high consumption amount may be recommended to a user with a high service processing frequency, and a service with a low consumption amount may be recommended to a user with a "medium service processing frequency". When allocating a service to a plurality of target users, the service may be preferentially allocated to "medium service processing frequency" and "high service processing frequency", and no service may be recommended to a user having "low service processing frequency".
In some embodiments, after processing the recommended service, the target user may feed back a corresponding service evaluation, where the service evaluation may be used to describe the experience of the user in processing the recommended service, and the service evaluation may be an evaluation issued by the user, such as "good service recommendation" or "bad service recommendation", or may be a processing manner for the allocated service by the user, such as directly canceling or closing the allocated service, or directly entering into processing the service. In practical application, other service processing modes can be set according to specific situations.
Under the condition that the service evaluation can be used for describing the experience of a user for processing the service, the service evaluation can be utilized for optimizing the category division model so as to improve the identification accuracy. The specific optimization process can evaluate the accuracy of the classification model based on business evaluation, so that deep optimization or retraining of the classification model is performed based on the classification model. The specific optimization process may be set based on the actual application situation, and is not described herein again.
To illustrate with a specific scenario example, first, an example of a service distribution system 31 is described based on fig. 3, which includes a bank server and a client. The bank server comprises a client data storage chip 32 for storing the service data of the client, a feature selection chip 33 for extracting corresponding features from the service data, a cluster calculation chip 34 for clustering according to the service data features and determining different classes corresponding to the service data, and a client label storage chip 35 for storing the classes of the client, namely the client labels, after determining the user classes according to the clustering results. The client mainly comprises a client login module 36 through which a client can log in a corresponding account, and a marketing recommendation module 37 for obtaining the classification category of the client so as to recommend a corresponding service to the client.
Fig. 4 is a general flow of a bank card marketing recommendation process, which includes several major steps of raw data acquisition, feature extraction, multi-view clustering and clustering results. In the original data acquisition stage, customer account distribution data, customer consumption merchant type distribution data, customer consumption amount distribution data and the like can be acquired, in the feature extraction stage, single-view data feature extraction can be sequentially carried out on data in different views, and based on extracted features, data clustering is realized by using a weighted multi-view clustering algorithm based on feature selection. And giving a corresponding customer label to the user according to the clustering result, thereby realizing the marketing recommendation of the bank card by utilizing the customer label.
Based on the system and the overall process, a specific example of the bank card marketing recommendation process is introduced. As shown in fig. 5, step 51 is first executed to derive various service scenarios and various types of data information of the customer through the bank background, for example: the system comprises client account distribution information, client transaction amount distribution information, client transaction opponent distribution information, client transaction frequency distribution information, client consumption merchant distribution information, client transaction channel distribution information and the like. Thereafter, step 52 may be performed to perform feature extraction on the client data of different scenes and views. And step 53 is executed again, the poem graph data extracted by the characteristics are input into the algorithm front, and data clustering calculation is prepared. Based on the step 54, initializing parameters such as p (weighted sparsity of each type of data), M (number of tags), reliability adjustment parameters, tmax (maximum cycle count calculation), and the like, executing the step 55 to perform algorithm operation, and judging whether convergence or operation is performed to the maximum cycle count in the step 56, if not, continuing executing the step 55, otherwise, executing the step 57, checking whether part of typical client tags meet expected results, and if not, returning to the step 54 to perform clustering calculation again. If the expected result is met, then step 58 is executed, the label category to which each type of customer belongs is obtained according to the clustering result, a marketing strategy is established for the corresponding category, then step 59 can be executed, accurate marketing suggestions are directionally pushed to the customer through a bank terminal, such as a mobile phone bank, an e-bank, a personal internet bank and the like, correspondingly, step 510 can be executed, the prediction result of the model is perfected according to the subsequent feedback result (for example, directly closing or entering and completing newly-made business and the like) of the customer on marketing, and therefore the corresponding business is more accurately recommended to the customer, and the use experience of the customer is improved.
It should be noted that the above scenario example is only an example implemented based on a bank card application scenario, and a corresponding effect is to accurately recommend a corresponding bank card service to a user, and in practical application, the application scenario and the implemented technical effect may be adjusted according to requirements on the basis of the technical solution of the embodiment of the present specification, and are not limited to the above scenario example, and are not described herein again.
Based on the above embodiments and the introduction of the scene examples, it can be seen that, after the multi-view service data from different service scenes is obtained, the method first performs clustering processing on the service data to obtain clustered service data, and determines the category of the user according to the clustered service data, so that the corresponding recommended service can be allocated to the user according to the category of the user. The method effectively ensures the processing of the multi-view data through the data clustering and the user category determining mode. In addition, the multi-view data corresponding to the user is fully utilized, and under the condition that the multi-view data has more information compared with single-view data, the accuracy of service distribution is improved, and the experience of the user in processing the corresponding service is further improved.
A service allocation apparatus according to an embodiment of the present description is introduced based on a service allocation method corresponding to fig. 1. As shown in fig. 6, the traffic distribution apparatus includes the following modules.
A service data acquiring module 610, configured to acquire service data corresponding to at least two service scenarios from a target user; the service data is used for representing the service processing content of the target user in the service scene.
A clustering service data dividing module 620, configured to divide the service data into at least two groups of clustering service data; the clustering service data comprises service data of the same type; the clustering service data corresponds to a preset clustering category.
A user category determining module 630, configured to determine a user category of the target user based on the clustering service data and a corresponding preset clustering category.
A recommendation service assignment module 640, configured to assign recommendation services corresponding to the user category to the target user.
Based on the service allocation method corresponding to fig. 1, an embodiment of the present specification provides a service allocation device. As shown in fig. 7, the traffic distribution device may include a memory and a processor.
In this embodiment, the memory may be implemented in any suitable manner. For example, the memory may be a read-only memory, a mechanical hard disk, a solid state disk, a U disk, or the like. The memory may be used to store computer program instructions.
In this embodiment, the processor may be implemented in any suitable manner. For example, the processor may take the form of, for example, a microprocessor or processor and a computer-readable medium that stores computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, an embedded microcontroller, and so forth. The processor may execute the computer program instructions to perform the steps of: acquiring service data of a target user corresponding to at least two service scenes; the service data is used for representing the service processing content of the target user in the service scene; dividing the service data into at least two groups of clustering service data; the clustering service data comprises service data of the same type; the clustering service data corresponds to a preset clustering category; determining the user category of the target user based on the clustering service data and the corresponding preset clustering category; and distributing the recommended service corresponding to the user category to the target user.
It should be noted that the service allocation method, the device and the apparatus may be applied to the technical field of artificial intelligence, and may also be applied to other technical fields, which is not limited to this.
In the 90 s of the 20 th century, improvements in a technology could clearly distinguish between improvements in hardware (e.g., improvements in circuit structures such as diodes, transistors, switches, etc.) and improvements in software (improvements in process flow). However, as technology advances, many of today's process flow improvements have been seen as direct improvements in hardware circuit architecture. Designers almost always obtain the corresponding hardware circuit structure by programming an improved method flow into the hardware circuit. Thus, it cannot be said that an improvement in the process flow cannot be realized by hardware physical modules. For example, a Programmable Logic Device (PLD), such as a Field Programmable Gate Array (FPGA), is an integrated circuit whose Logic functions are determined by programming the Device by a user. A digital system is "integrated" on a PLD by the designer's own programming without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Furthermore, nowadays, instead of manually making an Integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to a software compiler used in program development and writing, but the original code before compiling is also written by a specific Programming Language, which is called Hardware Description Language (HDL), and HDL is not only one but many, such as abel (advanced Boolean Expression Language), ahdl (alternate Hardware Description Language), traffic, pl (core universal Programming Language), HDCal (jhdware Description Language), lang, Lola, HDL, laspam, hardward Description Language (vhr Description Language), vhal (Hardware Description Language), and vhigh-Language, which are currently used in most common. It will also be apparent to those skilled in the art that hardware circuitry that implements the logical method flows can be readily obtained by merely slightly programming the method flows into an integrated circuit using the hardware description languages described above.
The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
From the above description of the embodiments, it is clear to those skilled in the art that the present specification can be implemented by software plus the necessary first hardware platform. Based on such understanding, the technical solutions of the present specification may be essentially or partially implemented in the form of software products, which may be stored in a storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and include instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments of the present specification.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The description is operational with numerous first or special purpose computing system environments or configurations. For example: personal computers, server computers, hand-held or portable devices, tablet-type devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.
This description may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
While the specification has been described with examples, those skilled in the art will appreciate that there are numerous variations and permutations of the specification that do not depart from the spirit of the specification, and it is intended that the appended claims include such variations and modifications that do not depart from the spirit of the specification.

Claims (11)

1. A method for service allocation, comprising:
acquiring service data of a target user corresponding to at least two service scenes; the service data is used for representing the service processing content of the target user in the service scene;
dividing the service data into at least two groups of clustering service data; the clustering service data comprises service data of the same type; the clustering service data corresponds to a preset clustering category;
determining the user category of the target user based on the clustering service data and the corresponding preset clustering category;
and distributing the recommended service corresponding to the user category to the target user.
2. The method of claim 1, wherein the business data includes at least one of customer account distribution information, customer transaction amount distribution information, customer transaction counter-party distribution information, customer transaction frequency distribution information, customer consuming merchant distribution information, customer transaction channel distribution information.
3. The method of claim 1, wherein the dividing the traffic data into at least two groups of clustered traffic data comprises:
extracting service data characteristics from the service data;
and dividing the service data into at least two groups of clustering service data based on the service data characteristics.
4. The method of claim 1, wherein the dividing the traffic data into at least two groups of clustered traffic data comprises:
and clustering the service data by using a k-means clustering algorithm to obtain at least two groups of clustered service data.
5. The method of claim 1, wherein the at least two service scenarios respectively have different weight values; the dividing the service data into at least two groups of clustering service data comprises:
and dividing the service data into at least two groups of clustering service data based on the weight value corresponding to the service scene to which the service data belongs.
6. The method of claim 3, wherein the dividing the traffic data into at least two groups of clustered traffic data comprises:
and clustering the service data by using a weighted multi-view clustering algorithm based on feature selection to obtain at least two groups of clustered service data.
7. The method of claim 6, wherein the inputting the clustering service data into a pre-constructed category classification model to obtain the user category of the target user comprises:
repeatedly utilizing the weighted multi-view clustering algorithm based on feature selection to calculate clustering service data until the calculation result is converged or the maximum cycle number is reached;
and determining the user category of the target user according to the output calculation result.
8. The method of claim 1, wherein the determining the user category of the target user based on the clustered service data and a corresponding preset clustering category comprises:
inputting the clustering service data into a pre-constructed category division model to obtain the user category of the target user; the class division model is used for determining the class of the user according to the classification result of the service data.
9. The method of claim 8, wherein after assigning the recommended service corresponding to the user category to the target user, further comprising:
receiving service evaluation fed back by a target user; the service evaluation is used for describing the experience of the target user in processing the recommended service;
optimizing the category classification model based on the business evaluation.
10. A traffic distribution apparatus, comprising:
the service data acquisition module is used for acquiring service data of a target user corresponding to at least two service scenes; the service data is used for representing the service processing content of the target user in the service scene;
the clustering service data dividing module is used for dividing the service data into at least two groups of clustering service data; the clustering service data comprises service data of the same type; the clustering service data corresponds to a preset clustering category;
the user category determining module is used for determining the user category of the target user based on the clustering service data and the corresponding preset clustering category;
and the recommendation service distribution module is used for distributing the recommendation service corresponding to the user category to the target user.
11. A traffic distribution apparatus comprising a memory and a processor;
the memory to store computer program instructions;
the processor to execute the computer program instructions to implement the steps of: acquiring service data of a target user corresponding to at least two service scenes; the service data is used for representing the service processing content of the target user in the service scene; dividing the service data into at least two groups of clustering service data; the clustering service data comprises service data of the same type; the clustering service data corresponds to a preset clustering category; determining the user category of the target user based on the clustering service data and the corresponding preset clustering category; and distributing the recommended service corresponding to the user category to the target user.
CN202110486442.1A 2021-04-30 2021-04-30 Service distribution method, device and equipment Pending CN113159213A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110486442.1A CN113159213A (en) 2021-04-30 2021-04-30 Service distribution method, device and equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110486442.1A CN113159213A (en) 2021-04-30 2021-04-30 Service distribution method, device and equipment

Publications (1)

Publication Number Publication Date
CN113159213A true CN113159213A (en) 2021-07-23

Family

ID=76871439

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110486442.1A Pending CN113159213A (en) 2021-04-30 2021-04-30 Service distribution method, device and equipment

Country Status (1)

Country Link
CN (1) CN113159213A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113723945A (en) * 2021-09-15 2021-11-30 中国银行股份有限公司 Bank user data processing method and device
CN115545353A (en) * 2022-11-29 2022-12-30 支付宝(杭州)信息技术有限公司 Method and device for business wind control, storage medium and electronic equipment
CN113723945B (en) * 2021-09-15 2024-06-28 中国银行股份有限公司 Bank user data processing method and device

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113723945A (en) * 2021-09-15 2021-11-30 中国银行股份有限公司 Bank user data processing method and device
CN113723945B (en) * 2021-09-15 2024-06-28 中国银行股份有限公司 Bank user data processing method and device
CN115545353A (en) * 2022-11-29 2022-12-30 支付宝(杭州)信息技术有限公司 Method and device for business wind control, storage medium and electronic equipment
CN115545353B (en) * 2022-11-29 2023-04-18 支付宝(杭州)信息技术有限公司 Business wind control method, device, storage medium and electronic equipment

Similar Documents

Publication Publication Date Title
US20210287048A1 (en) System and method for efficient generation of machine-learning models
CN110069709B (en) Intention recognition method, device, computer readable medium and electronic equipment
CN111797320B (en) Data processing method, device, equipment and storage medium
CN110852881B (en) Risk account identification method and device, electronic equipment and medium
US10528889B2 (en) Stereoscopic learning for classification
CN112364937B (en) User category determination method and device, recommended content determination method and electronic equipment
CN110046806B (en) Method and device for customer service order and computing equipment
US20220180209A1 (en) Automatic machine learning system, method, and device
CN114638234A (en) Big data mining method and system applied to online business handling
CN113656699A (en) User feature vector determination method, related device and medium
CN116823410B (en) Data processing method, object processing method, recommending method and computing device
CN111667307B (en) Method and device for predicting financial product sales volume
CN113159213A (en) Service distribution method, device and equipment
CN113569955A (en) Model training method, user portrait generation method, device and equipment
CN112836742A (en) System resource adjusting method, device and equipment
CN113822390B (en) User portrait construction method and device, electronic equipment and storage medium
CN113034264A (en) Method and device for establishing customer loss early warning model, terminal equipment and medium
CN113837183B (en) Multi-stage certificate intelligent generation method, system and medium based on real-time mining
KR20200029647A (en) Generalization method for curated e-Commerce system by user personalization
CN116795357B (en) Processing method, device, equipment and storage medium of business rule
CN113159957B (en) Transaction processing method and device
CN117911116A (en) E-commerce information recommendation system based on artificial intelligence and control method thereof
CN115017341A (en) User tag processing method and device and electronic equipment
Babu et al. Resource-Based Prediction in Cloud Computing Using LSTM with Autoencoders
CN115545113A (en) KNN algorithm-based user classification 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