CN118277742A - User portrait construction method and device, storage medium and computer equipment - Google Patents
User portrait construction method and device, storage medium and computer equipment Download PDFInfo
- Publication number
- CN118277742A CN118277742A CN202410466393.9A CN202410466393A CN118277742A CN 118277742 A CN118277742 A CN 118277742A CN 202410466393 A CN202410466393 A CN 202410466393A CN 118277742 A CN118277742 A CN 118277742A
- Authority
- CN
- China
- Prior art keywords
- product
- user
- category
- record data
- transaction record
- 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
- 238000010276 construction Methods 0.000 title claims abstract description 24
- 230000002776 aggregation Effects 0.000 claims abstract description 75
- 238000004220 aggregation Methods 0.000 claims abstract description 75
- 238000012545 processing Methods 0.000 claims abstract description 55
- 238000013507 mapping Methods 0.000 claims abstract description 53
- 238000004458 analytical method Methods 0.000 claims description 39
- 238000000034 method Methods 0.000 claims description 29
- 238000011156 evaluation Methods 0.000 claims description 22
- 238000004891 communication Methods 0.000 claims description 17
- 239000011159 matrix material Substances 0.000 claims description 17
- 238000007621 cluster analysis Methods 0.000 claims description 6
- 238000004138 cluster model Methods 0.000 claims description 6
- 230000004931 aggregating effect Effects 0.000 claims description 2
- 238000007405 data analysis Methods 0.000 abstract description 2
- 230000007774 longterm Effects 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000010224 classification analysis Methods 0.000 description 3
- 238000003064 k means clustering Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000011022 operating instruction Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a user portrait construction method and device, a storage medium and computer equipment, which belong to the technical field of data analysis and are suitable for the financial field, and mainly solve the problems of large feature dimension and low feature saturation of a user portrait obtained by simply integrating data; determining a target product category corresponding to each transaction record data based on the mapping relation; performing category marking processing on the transaction record data by adopting the category of the target product to obtain marked transaction record data; carrying out aggregation statistics on the marked transaction record data based on the user numbers and the product categories to obtain a plurality of aggregation results of a plurality of product categories corresponding to each user number; and combining the aggregation results to obtain the user portrait corresponding to the user number.
Description
Technical Field
The invention relates to the technical field of data analysis, and is suitable for the financial field, in particular to a user portrait construction method and device, a storage medium and computer equipment.
Background
In the financial field, the user image can reflect the holding condition of the user on the financial product, and is not only a part of a user label system, but also an important component of the model entering feature of the product recommendation model. Constructing a user representation helps operators to better understand customers and also helps recommendation models to better identify potential customers. The financial product has the characteristics of relatively few product types and relatively slow product updating, so that the construction method of the user portrait in the financial field is relatively different from the construction method of the user portrait in other fields.
At present, in the financial field, source data of the holding condition of customer financial products in a database is adopted to construct user portraits, such as transaction lists or transaction detail data, and the transaction lists taking transaction flow as a main key are simply aggregated into user portraits taking customers as the main key. Although the types of financial products are relatively small, the quantity of the financial products can be tens of thousands or more, so that the feature dimension number of the user image obtained by simply integrating the existing data is large and the feature saturation is too low, and further the product holding condition of an analysis user is limited.
Disclosure of Invention
In view of this, the invention provides a method and a device for constructing a user portrait, a storage medium and a computer device, and mainly aims to solve the problems of larger feature dimension and low feature saturation of the user portrait obtained by simply integrating data in the prior art.
According to one aspect of the present invention, there is provided a user portrait construction method, including:
respectively acquiring corresponding transaction lists according to the product service range, and performing category analysis processing based on the transaction lists to obtain a plurality of product categories and mapping relations between the product categories and product names;
Determining a target product category corresponding to each transaction record data based on the mapping relation; performing category marking processing on the transaction record data by adopting the target product category to obtain marked transaction record data;
Carrying out aggregation statistics on the marked transaction record data based on the user numbers and the product categories to obtain a plurality of aggregation results of a plurality of product categories corresponding to each user number;
And combining the aggregation results to obtain the user portrait corresponding to the user number.
Further, the performing category analysis processing based on the transaction list to obtain a plurality of product categories, and the mapping relationship between the product categories and the product names includes:
Acquiring product attribute information in each transaction record data from the transaction list;
Performing duplication removal processing on the product attribute information to obtain a duplicate removed product attribute list;
And carrying out cluster analysis on the product attribute list by adopting a cluster model to obtain a plurality of product categories and mapping relations between the product categories and the product names.
Further, after the category analysis processing is performed based on the transaction list to obtain a plurality of product categories, the method further includes:
calculating covariance matrixes among categories and covariance matrixes of data in the categories based on a plurality of product categories;
evaluating a category analysis processing result based on the covariance matrix between the categories and the category internal data covariance matrix to obtain an evaluation value;
and if the evaluation value is smaller than the evaluation threshold value, adjusting the model parameters of the clustering model, and carrying out clustering analysis on the transaction record data based on the adjusted model parameters to obtain a plurality of adjusted product categories so as to construct the user image based on the adjusted product categories.
Further, the determining, based on the mapping relationship, the target product category corresponding to each transaction record data includes:
Obtaining a mapping relation between product categories and product names; obtaining a target product name from the transaction record data;
and determining the target product category corresponding to the target product name based on the mapping relation. Further, the step of performing category marking processing on the transaction record data by using the target product category to obtain marked transaction record data includes:
Setting a category label based on the product category;
And carrying out category marking processing on the transaction record data by adopting the category marking to obtain marked transaction record data.
Further, the aggregating statistics of the marked transaction record data based on the user numbers and the product categories, and obtaining a plurality of aggregate results of a plurality of product categories corresponding to each user number includes:
Acquiring all target mark transaction record data corresponding to the user number;
And respectively carrying out product holding quantity statistics, product holding amount statistics and product holding time interval statistics on the target mark transaction record data according to the category marks to obtain a plurality of aggregation results of a plurality of product categories corresponding to the user numbers.
Further, the merging the plurality of aggregation results to obtain a user portrait corresponding to the user number includes:
Combining a plurality of aggregation results corresponding to the user numbers to obtain a single-service user portrait corresponding to the product service range;
and combining a plurality of single-service user portraits corresponding to a plurality of product service ranges based on the user numbers to obtain the user portraits corresponding to the user numbers.
According to another aspect of the present invention, there is provided a user portrait construction apparatus, comprising:
the classification module is used for respectively acquiring corresponding transaction lists according to the product service range, and carrying out category analysis processing based on the transaction lists to obtain a plurality of product categories and mapping relations between the product categories and product names;
the marking module is used for determining a target product category corresponding to each transaction record data based on the mapping relation; performing category marking processing on the transaction record data by adopting the target product category to obtain marked transaction record data;
the aggregation statistics module is used for carrying out aggregation statistics on the marked transaction record data based on the user numbers and the product categories to obtain a plurality of aggregation results of a plurality of product categories corresponding to each user number;
And the merging module is used for merging the aggregation results to obtain the user portrait corresponding to the user number.
Further, the classification module is further configured to:
Acquiring product attribute information in each transaction record data from the transaction list;
Performing duplication removal processing on the product attribute information to obtain a duplicate removed product attribute list;
And carrying out cluster analysis on the product attribute list by adopting a cluster model to obtain a plurality of product categories and mapping relations between the product categories and the product names.
Further, the apparatus further comprises a classification evaluation module for:
calculating covariance matrixes among categories and covariance matrixes of data in the categories based on a plurality of product categories;
evaluating a category analysis processing result based on the covariance matrix between the categories and the category internal data covariance matrix to obtain an evaluation value;
and if the evaluation value is smaller than the evaluation threshold value, adjusting the model parameters of the clustering model, and carrying out clustering analysis on the transaction record data based on the adjusted model parameters to obtain a plurality of adjusted product categories so as to construct the user image based on the adjusted product categories.
Further, the marking module is further configured to:
Obtaining a mapping relation between product categories and product names; obtaining a target product name from the transaction record data;
And determining the target product category corresponding to the target product name based on the mapping relation.
Further, the marking module is further configured to:
Setting a category label based on the product category;
And carrying out category marking processing on the transaction record data by adopting the category marking to obtain marked transaction record data.
Further, the aggregation statistics module is further configured to:
Acquiring all target mark transaction record data corresponding to the user number;
And respectively carrying out product holding quantity statistics, product holding amount statistics and product holding time interval statistics on the target mark transaction record data according to the category marks to obtain a plurality of aggregation results of a plurality of product categories corresponding to the user numbers.
Further, the merging module is further configured to:
Combining a plurality of aggregation results corresponding to the user numbers to obtain a single-service user portrait corresponding to the product service range;
and combining a plurality of single-service user portraits corresponding to a plurality of product service ranges based on the user numbers to obtain the user portraits corresponding to the user numbers.
According to still another aspect of the present invention, there is provided a storage medium having stored therein at least one executable instruction for causing a processor to perform operations corresponding to the method for constructing a user representation as described above.
According to yet another aspect of the present invention, there is provided a computer device comprising a processor, a memory, a communication interface and a communication bus, said processor, said memory and said communication interface completing communication with each other via said communication bus;
The memory is used for storing at least one executable instruction, and the executable instruction enables the processor to execute the operation corresponding to the user portrait construction method.
By means of the technical scheme, the technical scheme provided by the embodiment of the invention has at least the following advantages:
Compared with the prior art, the method and the device for constructing the user portrait, provided by the invention, have the advantages that the product categories in different product service ranges and the mapping relation between the product categories and the product names are obtained by carrying out classification analysis processing on the transaction lists in different product service ranges; determining a target product category corresponding to each transaction record data based on the mapping relation; performing category marking processing on the transaction record data by adopting the category of the target product to obtain marked transaction record data; carrying out aggregation statistics on the marked transaction record data based on the user numbers and the product categories to obtain a plurality of aggregation results of a plurality of product categories corresponding to each user number; and combining the plurality of aggregation results to obtain user portraits corresponding to the user numbers, so that the construction of single service line product user portraits in different product service ranges is realized, the combination of user portraits in multiple product service ranges is also realized, and the comprehensive user portraits reflecting the product holding condition of the user in the financial field are formed. Because the method of the invention respectively carries out aggregation statistics according to the product category, the feature dimension number of the user image is greatly reduced, and the feature saturation is improved, thereby solving the problems of larger feature dimension number and too low feature saturation of the user image obtained by simply integrating the data in the prior art. In addition, when the service range of the product changes, the user portrait construction method can automatically and efficiently update the user portrait, thereby reducing the manpower and time required for constructing the user portrait.
The foregoing description is only an overview of the present invention, and is intended to be implemented in accordance with the teachings of the present invention in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present invention more readily apparent.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to designate like parts throughout the figures. In the drawings:
FIG. 1 shows a flow diagram of a user portrait construction method according to an embodiment of the present invention;
FIG. 2 is a flow chart illustrating another method for constructing a user portrait according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating a method for constructing a user portrait according to an embodiment of the present invention;
FIG. 4 is a flowchart illustrating a method for constructing a user portrait according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a user portrait constructing apparatus according to an embodiment of the present invention;
fig. 6 shows a schematic structural diagram of a computer device according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
The embodiment of the invention provides a user portrait construction method, as shown in figure 1, which comprises the following steps:
101. Respectively acquiring corresponding transaction lists according to the product service range, and performing category analysis processing based on the transaction lists to obtain a plurality of product categories and mapping relations between the product categories and product names;
In the embodiment of the invention, the current execution end respectively acquires the corresponding transaction list according to the product service range, wherein the product service range is used for representing the range for developing various financial services in the financial field, including a loan service range, a financial service range, an insurance service range and the like, and the embodiment of the invention is not particularly limited. The transaction list is used for representing a list for recording transaction data after the financial product is generated for transaction. The current execution end performs category analysis processing based on the transaction list to obtain a plurality of product categories and mapping relations between the product categories and the product names. Because the transaction list contains a large amount of transaction data, the category analysis processing can be realized by adopting a machine learning mode, such as a support vector machine model, a mean value clustering model and the like, and the embodiment of the invention is not particularly limited. The product categories are used for representing product categories which can be divided in a single product business range in the financial field and are used for representing single or multiple feature dimension information, for example, the product categories in the loan business range can be divided into long-term loans, short-term loans, mortgage-free loans, long-term mortgage-free loans, short-term mortgage-free loans and the like through machine learning, and the embodiment of the invention is not limited in detail.
102. Determining a target product category corresponding to each transaction record data based on the mapping relation; performing category marking processing on the transaction record data by adopting the target product category to obtain marked transaction record data;
In the embodiment of the invention, the target product category corresponding to each transaction record data is determined based on the mapping relation between the product category and the product name, if the product category corresponding to the transaction name A in the transaction record data is searched in the mapping relation to be the long-term mortgage type, the long-term mortgage type is determined as the target product category of the transaction record data, and the long-term mortgage type is adopted to carry out category marking processing on the transaction record data to obtain marked transaction record data of the transaction record data, and the embodiment of the invention is not particularly limited.
103. Carrying out aggregation statistics on the marked transaction record data based on the user numbers and the product categories to obtain a plurality of aggregation results of a plurality of product categories corresponding to each user number;
In the embodiment of the invention, the current execution terminal performs aggregation statistics on the marked transaction record data based on the user numbers and the product categories to obtain a plurality of aggregation results of a plurality of product categories corresponding to each user number. The user number is used for uniquely characterizing the identity information of the user, the user can be configured when the user registers, and all transaction record data generated by the user at a later stage can carry the user number. The user number may be one or more of a number, a letter, a special symbol, etc., which is not particularly limited in the embodiment of the present invention. In the specific aggregation statistics, all marked transaction record data of the same user number are found first, and then aggregation statistics is performed on each product category mark respectively, including product holding quantity statistics, product holding amount statistics, product holding time interval statistics and the like, and the embodiment of the invention is not limited in detail. The current execution end takes the aggregation statistical result of the marked transaction record data corresponding to the product category mark, such as the product holding quantity statistical value, the product holding amount statistical value, the product holding time interval statistical value and the like as an aggregation result. Because a single user number may participate in more than one product category transaction, a plurality of aggregation results corresponding to the same user number are obtained after aggregation statistics.
104. And combining the aggregation results to obtain the user portrait corresponding to the user number.
In the embodiment of the invention, the current execution end performs combination processing on a plurality of aggregation results to obtain the user portrait corresponding to the user number. And the merging processing is to combine the characteristics of the plurality of aggregation results in the characteristic dimension to obtain the user portrait containing all the characteristic dimensions of the aggregation results. For example, for user number 0001, the aggregate result a corresponding to the long-term loan product type, via steps 101-103, contains 4 feature dimensions; the aggregate result B corresponding to the short-term loan product type contains 4 feature dimensions; the aggregate result C corresponding to the mortgage product type contains 4 feature dimensions; the aggregate result D corresponding to the mortgage-free product type includes 4 feature dimensions, and the feature dimensions are combined during the combination processing to obtain a user portrait including 16 feature dimensions corresponding to the user number 0001.
Further, as a refinement and expansion of the foregoing embodiment, in order to further improve accuracy of product category analysis, another method for constructing a user portrait is provided, as shown in fig. 2, where the step of performing category analysis processing based on the transaction list to obtain a plurality of product categories, and a mapping relationship between the product categories and product names includes:
201. acquiring product attribute information in each transaction record data from the transaction list;
In the embodiment of the invention, the current execution end acquires the product attribute information in each transaction record data from the transaction list. The product attribute information includes a product name, a product amount, a product term, a product risk level, and the like, and the embodiment of the invention is not particularly limited.
202. Performing duplication removal processing on the product attribute information to obtain a duplicate removed product attribute list;
In the embodiment of the invention, the current execution end performs the duplication removal processing on the product attribute information to obtain a duplicate removed product attribute list. The duplicate removal process removes products with identical attribute information from the plurality of pieces of product attribute information, so that the obtained product attribute list is used for representing the list of the product attribute information corresponding to different financial products.
203. And carrying out cluster analysis on the product attribute list by adopting a cluster model to obtain a plurality of product categories and mapping relations between the product categories and the product names.
In the embodiment of the invention, the current execution end adopts a clustering model to perform clustering analysis on the product attribute list, so as to obtain a plurality of product categories and the mapping relation between the product categories and the product names. In the embodiment of the invention, the K-means clustering model is optimized to perform clustering analysis on the product attribute list, and the K-means clustering model can perform clustering analysis according to the set product types, namely products in the product attribute list in the loan service range can be classified into 8 types, 10 types, 12 types and the like according to the needs; dividing products in a product attribute list in a financial service range into 11 types, 13 types, 15 types and the like; products in the product attribute list within the insurance business range are classified into 9 types, 14 types, 16 types and the like, and the embodiment of the invention is not particularly limited. The K-means clustering model adopted in the embodiment of the invention can automatically generate the mapping relation between the product category and the product name after the clustering analysis, and the embodiment of the invention is not particularly limited.
Further, as a refinement and expansion of the foregoing embodiment, in order to evaluate the effect of the category analysis processing, thereby improving the accuracy of the category analysis processing, another method for constructing a user portrait is provided, as shown in fig. 3, where after the step of performing the category analysis processing based on the transaction list to obtain a plurality of product categories, the method further includes:
301. calculating covariance matrixes among categories and covariance matrixes of data in the categories based on a plurality of product categories;
302. Evaluating a category analysis processing result based on the covariance matrix between the categories and the category internal data covariance matrix to obtain an evaluation value;
In the embodiment of the invention, the current execution terminal calculates covariance matrixes among the categories based on a plurality of product categories and marks the covariance matrixes as matrix variables B k; calculating a covariance matrix of the data in the category, and marking the covariance matrix as a matrix variable W k; k represents the number of product categories.
It should be noted that, the current execution end evaluates the category analysis processing result based on the inter-category covariance matrix B k and the intra-category data covariance matrix W k, and a specific evaluation formula is as follows:
Wherein m is the number of product attribute information in the product attribute list, namely the number of products which are not repeated; tr is the trace of the matrix; s (k) is an evaluation value.
303. And if the evaluation value is smaller than the evaluation threshold value, adjusting the model parameters of the clustering model, and carrying out clustering analysis on the transaction record data based on the adjusted model parameters to obtain a plurality of adjusted product categories so as to construct the user image based on the adjusted product categories.
In the embodiment of the invention, the current execution end judges the evaluation value of the calculation and recall, if the evaluation value is smaller than the evaluation threshold value, the model parameters in the clustering model are adjusted, and then the clustering model after the parameters are adjusted is used for carrying out clustering analysis on the transaction record data to obtain a plurality of adjustment product categories, so that the user image is constructed based on the adjustment product categories. The specific process of performing cluster analysis on the transaction record data based on the cluster model after the parameter adjustment is the same as step 203, and the embodiment of the invention is not limited in detail.
Further, as a refinement and expansion of the foregoing embodiment, in order to facilitate division of transaction record data and perform modular analysis according to the divided data, another method for constructing a user portrait is provided, as shown in fig. 4, where the determining, based on the mapping relationship, a target product category corresponding to each transaction record data includes:
401. obtaining a mapping relation between product categories and product names; obtaining a target product name from the transaction record data;
402. And determining the target product category corresponding to the target product name based on the mapping relation.
In the embodiment of the invention, the current execution end acquires the mapping relation between the product category and the product name, acquires the target product name from the transaction record data, and searches the product category corresponding to the target product name in the mapping relation, namely the target product category. For example, a mapping table is used to represent the mapping relationship between the product category and the product name, as shown in the following table 1:
TABLE 1 mapping between product names and product categories
| Sequence number | Product name | Product category |
| 1 | AAAA001 | Class 2 |
| 2 | AAAA002 | Class 2 |
| 3 | BBBB001 | Class 1 |
| 4 | BBBB003 | Class 1 |
| 5 | CCCC001 | Class 4 |
| 6 | DDDD003 | Class 3 |
| ...... | ...... | ...... |
Based on the mapping table in table 1, when the target product name in the transaction record data is AAAA001, it may be determined that the target product type is type 2; when the target product name in the transaction record data is CCCC001, it may be determined that the target product type is type 4, which is not specifically limited in the embodiment of the present invention.
The step of carrying out category marking processing on the transaction record data by adopting the target product category, wherein the step of obtaining marked transaction record data comprises the following steps:
403. Setting a category label based on the product category;
404. And carrying out category marking processing on the transaction record data by adopting the category marking to obtain marked transaction record data.
In the embodiment of the present invention, the current execution end sets a category label based on the product category, where the category label may be one or more of letters, numbers and special symbols, and the embodiment of the present invention is not limited in detail. Product categories such as in the loan business scope may be labeled with DK01, DK02, DK03, etc. as category labels; the product category in the financial service range can be used as category labels by LC01, LC02, LC03 and the like; the product category in the insurance business range can be used as category labels by BX01, BX02, BX03 and the like, and the embodiment of the invention is not particularly limited. After the category label is set, the current execution end also needs to adopt the category label to carry out category label processing on the transaction record data to obtain the labeled transaction record data, so that the analysis of the data in later period is facilitated.
Further, as a refinement and expansion of the foregoing embodiment, in order to obtain the analysis of the transaction record data by modularization, another method for constructing a user portrait is provided, and the step of performing aggregation statistics on the marked transaction record data based on the user number and the product category, to obtain a plurality of aggregation results of a plurality of product categories corresponding to each user number includes:
Acquiring all target mark transaction record data corresponding to the user number;
And respectively carrying out product holding quantity statistics, product holding amount statistics and product holding time interval statistics on the target mark transaction record data according to the category marks to obtain a plurality of aggregation results of a plurality of product categories corresponding to the user numbers.
In the embodiment of the invention, the current execution end acquires all target mark transaction record data corresponding to the user number, for example, 20 target mark transaction record data corresponding to the user number 0001 in the loan service range, wherein 3 target mark transaction record data marked by category mark DK01, 3 target mark transaction record data marked by category mark DK02, 6 target mark transaction record data marked by category mark DK03 and 8 target mark transaction record data marked by category mark DK 04. The current execution end respectively carries out product holding quantity statistics, product holding amount statistics and product holding time interval statistics on the 3-item marked transaction record data marked by the category mark DK01 to obtain an aggregation result; carrying out product holding quantity statistics, product holding amount statistics and product holding time interval statistics on the 3-item label transaction record data marked by the category label DK02 to obtain another aggregation result; carrying out product holding quantity statistics, product holding amount statistics and product holding time interval statistics on 6 item mark transaction record data marked by category mark DK03, and obtaining an aggregation result; and counting the product holding quantity, the product holding amount and the product holding time interval according to the 8-item marked transaction record data marked by the category mark DK04, and obtaining an aggregation result again. In summary, 4 aggregation results of 4 product categories corresponding to the user number 0001 are obtained, and the embodiment of the present invention is not limited specifically.
Further, as a refinement and expansion of the specific implementation manner of the above embodiment, in order to obtain a complete portrait of a user based on a module analysis result, another method for constructing a user portrait is provided, where the step of merging a plurality of aggregation results to obtain a user portrait corresponding to the user number includes:
Combining a plurality of aggregation results corresponding to the user numbers to obtain a single-service user portrait corresponding to the product service range;
and combining a plurality of single-service user portraits corresponding to a plurality of product service ranges based on the user numbers to obtain the user portraits corresponding to the user numbers.
In the embodiment of the invention, the current execution end combines a plurality of aggregation results corresponding to the user numbers to obtain the single-service user portraits corresponding to the service range of the product. If the 4 aggregation results corresponding to the user number 0001 in the previous embodiment are combined to obtain a single service user portrait corresponding to the loan service range, the embodiment of the invention is not limited specifically. And combining the characteristics of the aggregation results in the characteristic dimension to obtain the user portrait containing all the characteristic dimensions of the aggregation results.
In addition, the current execution end also combines a plurality of single-service user portraits corresponding to a plurality of product service ranges based on the user numbers, and the user portraits corresponding to the user numbers are obtained. If the user number 0001 has corresponding single service user portraits in the loan service range, the financial service range and the insurance service range, the 3 single service user portraits are combined to obtain the user portraits corresponding to the user number 0001. At this time, user portrait construction is completed, and feature dimensions are greatly reduced, so that the saturation of the features is effectively improved.
Compared with the prior art, the method for constructing the user portrait obtains the product category in different product service ranges and the mapping relation between the product category and the product name by carrying out classification analysis processing on the transaction lists in different product service ranges; determining a target product category corresponding to each transaction record data based on the mapping relation; performing category marking processing on the transaction record data by adopting the category of the target product to obtain marked transaction record data; carrying out aggregation statistics on the marked transaction record data based on the user numbers and the product categories to obtain a plurality of aggregation results of a plurality of product categories corresponding to each user number; and combining the plurality of aggregation results to obtain user portraits corresponding to the user numbers, so that the construction of single service line product user portraits in different product service ranges is realized, the combination of user portraits in multiple product service ranges is also realized, and the comprehensive user portraits reflecting the product holding condition of the user in the financial field are formed. Because the method of the invention respectively carries out aggregation statistics according to the product category, the feature dimension number of the user image is greatly reduced, and the feature saturation is improved, thereby solving the problems of larger feature dimension number and too low feature saturation of the user image obtained by simply integrating the data in the prior art. In addition, when the service range of the product changes, the user portrait construction method can automatically and efficiently update the user portrait, thereby reducing the manpower and time required for constructing the user portrait.
As an implementation of the method shown in FIG. 1, an embodiment of the present invention provides a user portrait construction apparatus, as shown in FIG. 5, including:
The classification module 51 is configured to obtain corresponding transaction lists according to product service ranges, and perform category analysis processing based on the transaction lists to obtain a plurality of product categories and mapping relations between the product categories and product names;
a marking module 52, configured to determine a target product category corresponding to each transaction record data based on the mapping relationship; performing category marking processing on the transaction record data by adopting the target product category to obtain marked transaction record data;
an aggregation statistics module 53, configured to aggregate the marked transaction record data based on a user number and the product category, to obtain a plurality of aggregation results of a plurality of product categories corresponding to each user number;
And a merging module 54, configured to merge the plurality of aggregation results to obtain a user portrait corresponding to the user number.
Further, the classification module 51 is further configured to:
Acquiring product attribute information in each transaction record data from the transaction list;
Performing duplication removal processing on the product attribute information to obtain a duplicate removed product attribute list;
And carrying out cluster analysis on the product attribute list by adopting a cluster model to obtain a plurality of product categories and mapping relations between the product categories and the product names.
Further, the apparatus further comprises a classification evaluation module for:
calculating covariance matrixes among categories and covariance matrixes of data in the categories based on a plurality of product categories;
evaluating a category analysis processing result based on the covariance matrix between the categories and the category internal data covariance matrix to obtain an evaluation value;
and if the evaluation value is smaller than the evaluation threshold value, adjusting the model parameters of the clustering model, and carrying out clustering analysis on the transaction record data based on the adjusted model parameters to obtain a plurality of adjusted product categories so as to construct the user image based on the adjusted product categories.
Further, the marking module 52 is further configured to:
Obtaining a mapping relation between product categories and product names; obtaining a target product name from the transaction record data;
And determining the target product category corresponding to the target product name based on the mapping relation.
Further, the marking module 52 is further configured to:
Setting a category label based on the product category;
And carrying out category marking processing on the transaction record data by adopting the category marking to obtain marked transaction record data.
Further, the aggregation statistics module 53 is further configured to:
Acquiring all target mark transaction record data corresponding to the user number;
And respectively carrying out product holding quantity statistics, product holding amount statistics and product holding time interval statistics on the target mark transaction record data according to the category marks to obtain a plurality of aggregation results of a plurality of product categories corresponding to the user numbers.
Further, the merging module 54 is further configured to:
Combining a plurality of aggregation results corresponding to the user numbers to obtain a single-service user portrait corresponding to the product service range;
and combining a plurality of single-service user portraits corresponding to a plurality of product service ranges based on the user numbers to obtain the user portraits corresponding to the user numbers.
Compared with the prior art, the embodiment of the invention obtains the product category in different product service ranges and the mapping relation between the product category and the product name by carrying out classification analysis processing on the transaction list in different product service ranges; determining a target product category corresponding to each transaction record data based on the mapping relation; performing category marking processing on the transaction record data by adopting the category of the target product to obtain marked transaction record data; carrying out aggregation statistics on the marked transaction record data based on the user numbers and the product categories to obtain a plurality of aggregation results of a plurality of product categories corresponding to each user number; and combining the plurality of aggregation results to obtain user portraits corresponding to the user numbers, so that the construction of single service line product user portraits in different product service ranges is realized, the combination of user portraits in multiple product service ranges is also realized, and the comprehensive user portraits reflecting the product holding condition of the user in the financial field are formed. Because the method of the invention respectively carries out aggregation statistics according to the product category, the feature dimension number of the user image is greatly reduced, and the feature saturation is improved, thereby solving the problems of larger feature dimension number and too low feature saturation of the user image obtained by simply integrating the data in the prior art. In addition, when the service range of the product changes, the user portrait construction method can automatically and efficiently update the user portrait, thereby reducing the manpower and time required for constructing the user portrait.
According to one embodiment of the present invention, there is provided a storage medium storing at least one executable instruction for performing the user portrait construction method in any of the above method embodiments.
Fig. 6 is a schematic structural diagram of a computer device according to an embodiment of the present invention, and the specific embodiment of the present invention is not limited to the specific implementation of the computer device.
As shown in fig. 6, the computer device may include: a processor 602, a communication interface Communications Interface, a memory 606, and a communication bus 608.
Wherein: processor 602, communication interface 604, and memory 606 perform communication with each other via communication bus 608.
Communication interface 604 is used to communicate with network elements of other devices, such as clients or other servers.
Processor 602 is configured to execute program 610, and specifically to perform the steps associated with the user portrait construction method described above.
In particular, program 610 may include program code including computer-operating instructions.
The processor 602 may be a central processing unit CPU, or an Application-specific integrated Circuit ASIC (Application SPECIFIC INTEGRATED Circuit), or one or more integrated circuits configured to implement embodiments of the present invention. The one or more processors included in the computer device may be the same type of processor, such as one or more CPUs; but may also be different types of processors such as one or more CPUs and one or more ASICs.
A memory 606 for storing a program 610. The memory 606 may comprise high-speed RAM memory or may further comprise non-volatile memory (non-volatile memory), such as at least one disk memory.
The program 610 may be specifically operable to cause the processor 602 to:
respectively acquiring corresponding transaction lists according to the product service range, and performing category analysis processing based on the transaction lists to obtain a plurality of product categories and mapping relations between the product categories and product names;
Determining a target product category corresponding to each transaction record data based on the mapping relation; performing category marking processing on the transaction record data by adopting the target product category to obtain marked transaction record data;
Carrying out aggregation statistics on the marked transaction record data based on the user numbers and the product categories to obtain a plurality of aggregation results of a plurality of product categories corresponding to each user number;
And combining the aggregation results to obtain the user portrait corresponding to the user number.
It will be appreciated by those skilled in the art that the modules or steps of the invention described above may be implemented in a general purpose computing device, they may be concentrated on a single computing device, or distributed across a network of computing devices, they may alternatively be implemented in program code executable by computing devices, so that they may be stored in a memory device for execution by computing devices, and in some cases, the steps shown or described may be performed in a different order than that shown or described, or they may be separately fabricated into individual integrated circuit modules, or multiple modules or steps within them may be fabricated into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A user portrait construction method is characterized by comprising the following steps:
respectively acquiring corresponding transaction lists according to the product service range, and performing category analysis processing based on the transaction lists to obtain a plurality of product categories and mapping relations between the product categories and product names;
Determining a target product category corresponding to each transaction record data based on the mapping relation; performing category marking processing on the transaction record data by adopting the target product category to obtain marked transaction record data;
Carrying out aggregation statistics on the marked transaction record data based on the user numbers and the product categories to obtain a plurality of aggregation results of a plurality of product categories corresponding to each user number;
And combining the aggregation results to obtain the user portrait corresponding to the user number.
2. The method of claim 1, wherein the performing category analysis based on the transaction list to obtain a plurality of product categories, and wherein the mapping between the product categories and product names comprises:
Acquiring product attribute information in each transaction record data from the transaction list;
Performing duplication removal processing on the product attribute information to obtain a duplicate removed product attribute list;
And carrying out cluster analysis on the product attribute list by adopting a cluster model to obtain a plurality of product categories and mapping relations between the product categories and the product names.
3. The method of claim 2, wherein after the category analysis based on the transaction list, the method further comprises:
calculating covariance matrixes among categories and covariance matrixes of data in the categories based on a plurality of product categories;
evaluating a category analysis processing result based on the covariance matrix between the categories and the category internal data covariance matrix to obtain an evaluation value;
and if the evaluation value is smaller than the evaluation threshold value, adjusting the model parameters of the clustering model, and carrying out clustering analysis on the transaction record data based on the adjusted model parameters to obtain a plurality of adjusted product categories so as to construct the user image based on the adjusted product categories.
4. The method of claim 2, wherein the determining a target product category corresponding to each transaction record data based on the mapping relationship comprises:
Obtaining a mapping relation between product categories and product names; obtaining a target product name from the transaction record data;
And determining the target product category corresponding to the target product name based on the mapping relation.
5. The method of claim 1, wherein said performing category marking processing on said transaction record data using said target product category comprises:
Setting a category label based on the product category;
And carrying out category marking processing on the transaction record data by adopting the category marking to obtain marked transaction record data.
6. The method of claim 5, wherein aggregating the tagged transaction record data based on the user number and the product category to obtain a plurality of aggregate results for a plurality of the product categories corresponding to each of the user numbers comprises:
Acquiring all target mark transaction record data corresponding to the user number;
And respectively carrying out product holding quantity statistics, product holding amount statistics and product holding time interval statistics on the target mark transaction record data according to the category marks to obtain a plurality of aggregation results of a plurality of product categories corresponding to the user numbers.
7. The method according to any one of claims 1 to 6, wherein the merging the plurality of aggregation results to obtain the user representation corresponding to the user number includes:
Combining a plurality of aggregation results corresponding to the user numbers to obtain a single-service user portrait corresponding to the product service range;
and combining a plurality of single-service user portraits corresponding to a plurality of product service ranges based on the user numbers to obtain the user portraits corresponding to the user numbers.
8. A user portrait construction apparatus comprising:
the classification module is used for respectively acquiring corresponding transaction lists according to the product service range, and carrying out category analysis processing based on the transaction lists to obtain a plurality of product categories and mapping relations between the product categories and product names;
the marking module is used for determining a target product category corresponding to each transaction record data based on the mapping relation; performing category marking processing on the transaction record data by adopting the target product category to obtain marked transaction record data;
the aggregation statistics module is used for carrying out aggregation statistics on the marked transaction record data based on the user numbers and the product categories to obtain a plurality of aggregation results of a plurality of product categories corresponding to each user number;
And the merging module is used for merging the aggregation results to obtain the user portrait corresponding to the user number.
9. A storage medium having stored therein at least one executable instruction for performing operations corresponding to the method of constructing a user representation according to any one of claims 1-7.
10. A computer device comprising a processor, a memory, a communication interface and a communication bus, said processor, said memory and said communication interface completing communication with each other through said communication bus;
The memory is configured to store at least one executable instruction, where the executable instruction causes the processor to perform an operation corresponding to the user portrait construction method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410466393.9A CN118277742A (en) | 2024-04-17 | 2024-04-17 | User portrait construction method and device, storage medium and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410466393.9A CN118277742A (en) | 2024-04-17 | 2024-04-17 | User portrait construction method and device, storage medium and computer equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118277742A true CN118277742A (en) | 2024-07-02 |
Family
ID=91633568
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410466393.9A Pending CN118277742A (en) | 2024-04-17 | 2024-04-17 | User portrait construction method and device, storage medium and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118277742A (en) |
-
2024
- 2024-04-17 CN CN202410466393.9A patent/CN118277742A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106021362B (en) | Generation, image searching method and the device that the picture feature of query formulation represents | |
| WO2019214245A1 (en) | Information pushing method and apparatus, and terminal device and storage medium | |
| WO2021164382A1 (en) | Method and apparatus for performing feature processing for user classification model | |
| CN107633257B (en) | Data quality evaluation method and device, computer readable storage medium and terminal | |
| CN112035449B (en) | Data processing method and device, computer equipment and storage medium | |
| CN113918622A (en) | Information tracing method and system based on block chain | |
| CN113360768A (en) | Product recommendation method, device and equipment based on user portrait and storage medium | |
| CN112559524A (en) | Index database establishing method and device and storage medium | |
| CN113268665A (en) | Information recommendation method, device and equipment based on random forest and storage medium | |
| CN112434884A (en) | Method and device for establishing supplier classified portrait | |
| CN111898378B (en) | Industry classification method and device for government enterprise clients, electronic equipment and storage medium | |
| CN109144999B (en) | Data positioning method, device, storage medium and program product | |
| CN112818068A (en) | Big data and multidimensional feature-based data tracing method and system | |
| CN116719822A (en) | Method and system for storing massive structured data | |
| CN108830302B (en) | Image classification method, training method, classification prediction method and related device | |
| CN118277742A (en) | User portrait construction method and device, storage medium and computer equipment | |
| CN116155597A (en) | Access request processing method and device and computer equipment | |
| CN115293685A (en) | Logistics order state tracking method, device, equipment and storage medium | |
| CN111209397B (en) | Method for determining enterprise industry category | |
| CN109885710B (en) | User image depicting method based on differential evolution algorithm and server | |
| CN112560433A (en) | Information processing method and device | |
| CN113095064A (en) | Code field identification method and device, electronic equipment and storage medium | |
| CN111382343B (en) | Label system generation method and device | |
| CN114528452B (en) | Method and system for realizing data tag based on cigarette and wine sales | |
| CN117078112B (en) | Energy consumption detection method and data analysis system applied to enterprise abnormal electricity management |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication |