CN110400202A - Bank product recommended method and system - Google Patents
Bank product recommended method and system Download PDFInfo
- Publication number
- CN110400202A CN110400202A CN201910700946.1A CN201910700946A CN110400202A CN 110400202 A CN110400202 A CN 110400202A CN 201910700946 A CN201910700946 A CN 201910700946A CN 110400202 A CN110400202 A CN 110400202A
- Authority
- CN
- China
- Prior art keywords
- product
- user
- text
- information
- dictionary
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—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
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Technology Law (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides a kind of bank product recommended method and systems.The described method includes: obtaining user basic information and user preference information according to user characteristics description standard and its dictionary, files on each of customers text is determined using the user basic information and the user preference information;According to product description standard and its dictionary, product information is obtained, determines product description text using the product information;According to the files on each of customers text and the product description text, vector space is generated, in the way of similarity assessment, user to be recommended and corresponding recommended products are determined in the vector space.The present invention is based on files on each of customers texts and product description text that user's portrait and product information extract, with the association of natural language technology searching user and product more extensive in the vector space of language description, realize that personalized bank product is recommended.
Description
Technical field
The present invention relates to bank product recommended technology field, espespecially a kind of bank product recommended method and system.
Background technique
The analyses such as the characteristics of recommender system utilizes special Information Filtering Technology, feature or product according to client are therebetween
Relevance, by different Products Shows to possible interested client.
In recent years, recommender system plays a significant role in the product layout of internet scientific & technical corporation, in e-commerce
Recommendations, news content recommendation of websites article, music video recommendation of websites video etc., applying for these personalized recommendations is obtaining
New visitor, precision marketing are taken, enlivens and plays better effects on old visitor.Bank product marketing is reviewed, there are types to enrich for bank product,
Alternative feature high, purchase threshold is low, the financial product that bank product can purchase from a minority, becoming masses may participate in
Appreciation of fixed assets channel.The marketing activity that the marketing of bank product at present is based primarily upon parent body's task distributed pushes, this
The mobile terminal APP of external bank universal Generalization bounds are that rule-based engine or traditional algorithm single-point type product model are recommended.
With increasing for bank product type, product user range is more universal, and the marketing of existing bank product is recommended to support to be increasingly difficult to
With meet demand.Therefore, how bank product is directed to client and carries out personalized accurate recommendation into an important demand.
Summary of the invention
To solve the above-mentioned problems, the embodiment of the present invention provides a kind of bank product recommended method, which comprises
According to user characteristics description standard and its dictionary, user basic information and user preference information are obtained, using described
User basic information and the user preference information determine files on each of customers text;
According to product description standard and its dictionary, product information is obtained, determines product description text using the product information
This;
According to the files on each of customers text and the product description text, vector space is generated, similarity assessment side is utilized
Formula determines user to be recommended and corresponding recommended products in the vector space.
Optionally, in an embodiment of the present invention, described according to user characteristics description standard and its dictionary, obtain user's base
This information and user preference information include: the maintenance user characteristics description standard and its dictionary, are retouched according to the user characteristics
Standard and its dictionary are stated, the user basic information and the user preference information are obtained.
It is optionally, in an embodiment of the present invention, described that product information is obtained according to product description standard and its dictionary,
Determine that product description text includes: the maintenance product description standard and its dictionary using the product information, according to the production
Product description standard and its dictionary obtain the product information, determine product description text using the product information.
It is optionally, in an embodiment of the present invention, described according to the files on each of customers text and the product description text,
Vector space is generated, in the way of similarity assessment, determines that user to be recommended and corresponding recommendation produce in the vector space
It is text vector that product, which include: by the files on each of customers text and the product description text conversion, generates vector space;Determine institute
State the cosine value between text vector, using the cosine value carry out similarity assessment, in the vector space determine to
Recommended user and corresponding recommended products.
The embodiment of the present invention also provides a kind of Recommendation System for Financial Products, the system comprises: user portrait module, product
Characteristic extracting module and algorithm recommending module;
The user draws a portrait module according to user characteristics description standard and its dictionary, obtains user basic information and user is inclined
Good information determines files on each of customers text using the user basic information and the user preference information;
The product feature extraction module obtains product information according to product description standard and its dictionary, utilizes the production
Product information determines product description text;
The algorithm recommending module generates vector space according to the files on each of customers text and the product description text,
In the way of similarity assessment, user to be recommended and corresponding recommended products are determined in the vector space.
Optionally, in an embodiment of the present invention, user's portrait module includes: that user's specification describes unit, is used
In the maintenance user characteristics description standard and its dictionary;User information extraction unit, for being described according to the user characteristics
Standard and its dictionary obtain the user basic information;User preference extraction unit, for describing to mark according to the user characteristics
Quasi- and its dictionary, obtains the user preference information.
Optionally, in an embodiment of the present invention, the product feature extraction module includes: that the standardization of products describes mould
Block, for safeguarding the product description standard and its dictionary;Product information extraction unit, for according to the product description standard
And its dictionary, the product information is obtained, determines product description text using the product information.
Optionally, in an embodiment of the present invention, the algorithm recommending module includes: text vector unit, and being used for will
The files on each of customers text and the product description text conversion are text vector, generate vector space;Similarity assessment unit,
For determining the cosine value between the text vector, similarity assessment is carried out using the cosine value, with empty in the vector
Between middle determination user to be recommended and corresponding recommended products.
The embodiment of the present invention also provides a kind of computer equipment, including memory, processor and storage are on a memory simultaneously
The computer program that can be run on a processor, the processor perform the steps of when executing the computer program
According to user characteristics description standard and its dictionary, user basic information and user preference information are obtained, using described
User basic information and the user preference information determine files on each of customers text;
According to product description standard and its dictionary, product information is obtained, determines product description text using the product information
This;
According to the files on each of customers text and the product description text, vector space is generated, similarity assessment side is utilized
Formula determines user to be recommended and corresponding recommended products in the vector space.
The embodiment of the present invention also provides a kind of computer readable storage medium, is stored thereon with computer program, the meter
Calculation machine program performs the steps of when being executed by processor
According to user characteristics description standard and its dictionary, user basic information and user preference information are obtained, using described
User basic information and the user preference information determine files on each of customers text;
According to product description standard and its dictionary, product information is obtained, determines product description text using the product information
This;
According to the files on each of customers text and the product description text, vector space is generated, similarity assessment side is utilized
Formula determines user to be recommended and corresponding recommended products in the vector space.
The present invention is based on files on each of customers texts and product description text that user's portrait and product information extract, with nature
The association of language technology searching user and product more extensive in the vector space of language description realizes that personalized bank produces
Product are recommended.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, embodiment will be described below
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only of the invention some
Embodiment for those of ordinary skill in the art without any creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is a kind of flow chart of bank product recommended method of the embodiment of the present invention;
Fig. 2 is the flow chart that files on each of customers text is generated in the embodiment of the present invention;
Fig. 3 is the flow chart that product description text is generated in the embodiment of the present invention;
Fig. 4 is the flow chart that recommended user and corresponding recommended products are determined in the embodiment of the present invention;
Fig. 5 is a kind of structural schematic diagram of Recommendation System for Financial Products of the embodiment of the present invention;
Fig. 6 is the structural schematic diagram of user's portrait module in the embodiment of the present invention;
Fig. 7 is the structural schematic diagram of product feature extraction module in the embodiment of the present invention;
Fig. 8 is the structural schematic diagram of algorithm recommending module in the embodiment of the present invention.
Specific embodiment
The embodiment of the present invention provides a kind of bank product recommended method and system.
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
It is as shown in Figure 1 a kind of flow chart of bank product recommended method of the embodiment of the present invention, method as shown in the figure includes:
Step S1 obtains user basic information and user preference information according to user characteristics description standard and its dictionary, benefit
Files on each of customers text is determined with the user basic information and the user preference information;
Step S2 obtains product information according to product description standard and its dictionary, determines product using the product information
Text is described;
Step S3 generates vector space, utilizes similarity according to the files on each of customers text and the product description text
Assessment mode determines user to be recommended and corresponding recommended products in the vector space.
In the present embodiment, by user personality essential information and its preference profiles and product can Expressive Features be converted to
Files on each of customers text and product description text find user by text vector spatial simlanty algorithm from the angle of content of text
Think the product for being accustomed to purchase in associated groups of users and group, realize the accurate recommendation of consumer products, is user and silver
Row marketing personnel provide a reference.
As an embodiment of the present invention, according to user characteristics description standard and its dictionary, user basic information is obtained
And user preference information includes: the maintenance user characteristics description standard and its dictionary, according to the user characteristics description standard
And its dictionary, obtain the user basic information and the user preference information.
As an embodiment of the present invention, according to product description standard and its dictionary, product information is obtained, using described
Product information determines that product description text includes: the maintenance product description standard and its dictionary, according to the product description mark
Quasi- and its dictionary, obtains the product information, determines product description text using the product information.
As an embodiment of the present invention, according to the files on each of customers text and the product description text, generate to
Quantity space determines user to be recommended in the vector space and corresponding recommended products includes: in the way of similarity assessment
It is text vector by the files on each of customers text and the product description text conversion, generates vector space;Determine the text
Cosine value between vector carries out similarity assessment using the cosine value, to determine use to be recommended in the vector space
Family and corresponding recommended products.
In a specific embodiment of the invention, as shown in Fig. 2, generating files on each of customers text, specific step is as follows:
Step 101: the standard and its dictionary that maintenance user's portrait describes user characteristics.Specifically, integrating and safeguarding use
Family feature describes label, including labeling, label dictionary value, label value range etc., into next step.
Step S102: establishing criteria and dictionary extract user information from customer relationship information system.Specifically, by from visitor
Family relation information system extract user information, such as age segmentations, occupational group, whether senior executive user describe, into next step
Suddenly.
Step S103: establishing criteria and dictionary hold product historical user's preference information from product systems user.Specifically,
User preference information is extracted by holding product history from product systems user, as credit grade, silver demonstrate,prove frequency of transferring accounts, purchase produces
Product and products thereof description.
In a specific embodiment of the invention, as shown in figure 3, generating product description text, specific step is as follows:
Step S201: standard and its dictionary of the maintenance items feature to product description.Specifically, full dose can be increased newly and be safeguarded
Product feature to the standard and its dictionary, including product time limit, income rank, the mode that expires, name of product etc. of product description, into
Enter next step.
Step S202: establishing criteria and dictionary extract product information from each product systems.Specifically, by being produced from service class
The product systems such as product, deposit class product, financing class product, sinking funds bonds product extract product feature and obtain product description text.
In a specific embodiment of the invention, as shown in figure 4, determining the specific of recommended user and corresponding recommended products
Steps are as follows:
Step S301: load files on each of customers text and product description text generation vector space.Specifically, with
Doc2vec algorithm loads files on each of customers text and product description text, vector space is generated, into next step.
Step S302: the group and products thereof that user in vector space recommends is searched with similarity assessment.Specifically, In
It is searched in vector space and the corresponding group of the files on each of customers of recommended products user and group is needed to be accustomed to the product list held,
Product list is exported, service is finally externally exported with service form.
The present invention is based on files on each of customers texts and product description text that user's portrait and product information extract, with nature
The association of language technology searching user and product more extensive in the vector space of language description realizes that personalized bank produces
Product are recommended.On the one hand, a kind of new bank product Generalization bounds are provided, on the other hand, using the existing software and hardware resources of bank,
Further artificial intelligence is pushed to assist production decision.
It is illustrated in figure 5 a kind of structural schematic diagram of Recommendation System for Financial Products of the embodiment of the present invention, system as shown in the figure
It include: user's portrait module 1, product feature extraction module 2 and algorithm recommending module 3;
The user draws a portrait module 1 according to user characteristics description standard and its dictionary, obtains user basic information and user
Preference information determines files on each of customers text using the user basic information and the user preference information;
The product feature extraction module 2 obtains product information according to product description standard and its dictionary, utilizes the production
Product information determines product description text;
The algorithm recommending module 3 generates vector space according to the files on each of customers text and the product description text,
In the way of similarity assessment, user to be recommended and corresponding recommended products are determined in the vector space.
In the present embodiment, present system includes user's portrait module, product feature extraction module, algorithm recommendation mould
Block.User draws a portrait module for providing the files on each of customers text input after user's specificationization is extracted for recommender system, and product is special
Sign extraction module is used to provide the product description text after the standardization of products is extracted for system, and algorithm recommending module is for providing
The relevant basic algorithm tenability of natural language.The recommendation process of recommender system is with the files on each of customers of user's portrait module output
Text and the product description text of product feature extraction module output are to input, at the natural language in access algorithm recommending module
Relevant algorithm is managed, such as: text vector conversion, Text similarity computing.System can provide personalization for public users
Products Show, comprising service product, deposit product, finance product, sinking funds bonds etc., to realize the precision marketing of user.
In the present embodiment, Recommendation System for Financial Products isolated node is disposed, and user's portrait module 1, calculation product feature mention
Modulus block 2, algorithm recommending module 3 are interacted as the module of Recommendation System for Financial Products.Recommendation System for Financial Products is to take
Business form externally exports recommendation results, can provide recommendation service for Internetbank, Mobile banking, third party system, or battalion
Pin personnel, which provide, recommends marketing program.Wherein: user's portrait module 1 can increase and safeguard that full dose user characteristics describe label newly, wrap
Include labeling, label dictionary value, label value range etc..By extracting user information, such as age from customer relationship information system
Segmentation, occupational group, whether users' description such as senior executive.User preference letter is extracted by holding product history from product systems user
Breath, as credit grade, silver demonstrate,prove frequency of transferring accounts, purchase product and products thereof description.Product feature extraction module 2 can be increased newly and be tieed up
Full dose product feature is protected to the standard and its dictionary of product description, including product time limit, income rank, the mode that expires, ProductName
Claim etc..It is special by extracting product from product systems such as service class product, deposit class product, financing class product, sinking funds bonds products
Obtain product description text.Algorithm recommending module 3, the module as Recommendation System for Financial Products can be accessed directly, provide
The relevant basic algorithm of natural language processing is supported, files on each of customers text and product description text generation vector space, fortune are loaded
The group and products thereof that user in vector space recommends is searched with similarity assessment.
As an embodiment of the present invention, user's portrait module includes: that user's specification describes unit, for safeguarding
State user characteristics description standard and its dictionary;User information extraction unit, for according to the user characteristics description standard and its
Dictionary obtains the user basic information;User preference extraction unit, for according to the user characteristics description standard and its word
Allusion quotation obtains the user preference information.Wherein, files on each of customers text includes user basic information and user preference information.
As an embodiment of the present invention, product feature extraction module includes: standardization of products describing module, for tieing up
Protect the product description standard and its dictionary;Product information extraction unit is used for according to the product description standard and its dictionary,
The product information is obtained, determines product description text using the product information.
As an embodiment of the present invention, algorithm recommending module includes: text vector unit, is used for the user
Archives text and the product description text conversion are text vector, generate vector space;Similarity assessment unit, for determining
Cosine value between the text vector carries out similarity assessment using the cosine value, to determine in the vector space
User to be recommended and corresponding recommended products.
In a specific embodiment of the invention, as shown in fig. 6, user draws a portrait, module 1 is specifically included: user's specification is retouched
State unit 11, user information extraction unit 12, user preference extraction unit 13.Wherein:
User's specification describes unit 11, for safeguarding standard and its dictionary that user's portrait describe user characteristics, wraps
Include ability labeling, label dictionary value, label value range etc..
User information extraction unit 12 extracts user information from customer relationship information system for establishing criteria and dictionary.
It is inclined from product systems user to hold product historical user for establishing criteria and dictionary for user preference extraction unit 13
Good information.
In a specific embodiment of the invention, as shown in fig. 7, product feature extraction module 2 specifically includes: product standard
Change description unit 21, product information extraction unit 22.Wherein:
The standardization of products describes unit 21, is responsible for newly-increased and maintenance full dose product feature to the standard and its word of product description
Allusion quotation, including product time limit, income rank, the mode that expires, name of product etc.
Product information extraction unit 22, by being produced from service class product, deposit class product, financing class product, sinking funds bonds
The product systems such as product extract product feature and obtain product description text.
In a specific embodiment of the invention, as shown in figure 8, algorithm recommending module 3 specifically includes: text vector list
First 31, similarity assessment unit 32.Wherein:
Text vector unit 31 encodes text by text vector method, is computer text conversion
It will be appreciated that form.Text vector common method have onehot, bow, tf-idf, word2vec, doc2vec, Hash to
Quantization etc..Text vector unit 31 uses the files on each of customers text and production that doc2vec mode exports user's portrait module 1
The product description text conversion that product characteristic extracting module 2 exports is text vector.
Similarity assessment unit 32, for assessing the similarity degree between text vector.Usually calculate two sections of texts
Cosine value between text vector, numerical value is bigger, and to represent similarity bigger.The similarity based method of doc2vec can be used, equally
The similarity for being simply considered that text vector is exactly the cosine value between vector, according to the feature difference of data can be changed to LDA,
Docsim scheduling algorithm.
The present invention is based on files on each of customers texts and product description text that user's portrait and product information extract, with nature
The association of language technology searching user and product more extensive in the vector space of language description realizes that personalized bank produces
Product are recommended.On the one hand, a kind of new bank product Generalization bounds are provided, on the other hand, using the existing software and hardware resources of bank,
Further artificial intelligence is pushed to assist production decision.
The embodiment of the present invention also provides a kind of computer equipment, including memory, processor and storage are on a memory simultaneously
The computer program that can be run on a processor, the processor perform the steps of when executing the computer program
According to user characteristics description standard and its dictionary, user basic information and user preference information are obtained, using described
User basic information and the user preference information determine files on each of customers text;
According to product description standard and its dictionary, product information is obtained, determines product description text using the product information
This;
According to the files on each of customers text and the product description text, vector space is generated, similarity assessment side is utilized
Formula determines user to be recommended and corresponding recommended products in the vector space.
The embodiment of the present invention also provides a kind of computer readable storage medium, is stored thereon with computer program, the meter
Calculation machine program performs the steps of when being executed by processor
According to user characteristics description standard and its dictionary, user basic information and user preference information are obtained, using described
User basic information and the user preference information determine files on each of customers text;
According to product description standard and its dictionary, product information is obtained, determines product description text using the product information
This;
According to the files on each of customers text and the product description text, vector space is generated, similarity assessment side is utilized
Formula determines user to be recommended and corresponding recommended products in the vector space.
Conceived based on application identical with a kind of above-mentioned bank product recommended method, the present invention also provides a kind of above-mentioned meters
Calculate machine equipment and a kind of computer readable storage medium.Due to a kind of computer equipment and a kind of computer readable storage medium
The principle solved the problems, such as is similar to a kind of bank product recommended method, therefore a kind of computer equipment and a kind of computer-readable
The implementation of storage medium may refer to a kind of implementation of bank product recommended method, and overlaps will not be repeated.
The present invention is based on files on each of customers texts and product description text that user's portrait and product information extract, with nature
The association of language technology searching user and product more extensive in the vector space of language description realizes that personalized bank produces
Product are recommended.On the one hand, a kind of new bank product Generalization bounds are provided, on the other hand, using the existing software and hardware resources of bank,
Further artificial intelligence is pushed to assist production decision.
Those of ordinary skill in the art will appreciate that implementing the method for the above embodiments can lead to
Program is crossed to instruct relevant hardware and complete, which can be stored in a computer readable storage medium, such as
ROM/RAM, magnetic disk, CD etc..
Particular embodiments described above has carried out further in detail the purpose of the present invention, technical scheme and beneficial effects
Describe in detail it is bright, it should be understood that the above is only a specific embodiment of the present invention, the guarantor being not intended to limit the present invention
Range is protected, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should be included in this
Within the protection scope of invention.
Claims (10)
1. a kind of bank product recommended method, which is characterized in that the described method includes:
According to user characteristics description standard and its dictionary, user basic information and user preference information are obtained, utilizes the user
Essential information and the user preference information determine files on each of customers text;
According to product description standard and its dictionary, product information is obtained, determines product description text using the product information;
According to the files on each of customers text and the product description text, vector space, in the way of similarity assessment, In are generated
User to be recommended and corresponding recommended products are determined in the vector space.
2. obtaining the method according to claim 1, wherein described according to user characteristics description standard and its dictionary
It include: the maintenance user characteristics description standard and its dictionary to user basic information and user preference information, according to the use
Family feature description standard and its dictionary obtain the user basic information and the user preference information.
3. being produced the method according to claim 1, wherein described according to product description standard and its dictionary
Product information determines that product description text includes: to safeguard the product description standard and its dictionary using the product information, according to
The product description standard and its dictionary, obtain the product information, determine product description text using the product information.
4. the method according to claim 1, wherein described retouch according to the files on each of customers text and the product
It states text, generates vector space, in the way of similarity assessment, user to be recommended and corresponding is determined in the vector space
Recommended products includes:
It is text vector by the files on each of customers text and the product description text conversion, generates vector space;
It determines the cosine value between the text vector, carries out similarity assessment using the cosine value, with empty in the vector
Between middle determination user to be recommended and corresponding recommended products.
5. a kind of Recommendation System for Financial Products, which is characterized in that the system comprises: user's portrait module, product feature are extracted
Module and algorithm recommending module;
The user draws a portrait module according to user characteristics description standard and its dictionary, obtains user basic information and user preference letter
Breath, determines files on each of customers text using the user basic information and the user preference information;
The product feature extraction module obtains product information according to product description standard and its dictionary, is believed using the product
It ceases and determines product description text;
The algorithm recommending module generates vector space according to the files on each of customers text and the product description text, utilizes
Similarity assessment mode determines user to be recommended and corresponding recommended products in the vector space.
6. system according to claim 5, which is characterized in that user module of drawing a portrait includes:
User's specification describes unit, for safeguarding the user characteristics description standard and its dictionary;
User information extraction unit, for obtaining the user and believing substantially according to the user characteristics description standard and its dictionary
Breath;
User preference extraction unit, for obtaining the user preference letter according to the user characteristics description standard and its dictionary
Breath.
7. system according to claim 5, which is characterized in that the product feature extraction module includes:
Standardization of products describing module, for safeguarding the product description standard and its dictionary;
Product information extraction unit, for obtaining the product information, utilizing institute according to the product description standard and its dictionary
It states product information and determines product description text.
8. system according to claim 5, which is characterized in that the algorithm recommending module includes:
Text vector unit, it is raw for being text vector by the files on each of customers text and the product description text conversion
At vector space;
Similarity assessment unit carries out similarity using the cosine value for determining the cosine value between the text vector
Assessment, to determine user to be recommended and corresponding recommended products in the vector space.
9. a kind of computer equipment including memory, processor and stores the meter that can be run on a memory and on a processor
Calculation machine program, which is characterized in that the processor performs the steps of when executing the computer program
According to user characteristics description standard and its dictionary, user basic information and user preference information are obtained, utilizes the user
Essential information and the user preference information determine files on each of customers text;
According to product description standard and its dictionary, product information is obtained, determines product description text using the product information;
According to the files on each of customers text and the product description text, vector space, in the way of similarity assessment, In are generated
User to be recommended and corresponding recommended products are determined in the vector space.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
It is performed the steps of when being executed by processor
According to user characteristics description standard and its dictionary, user basic information and user preference information are obtained, utilizes the user
Essential information and the user preference information determine files on each of customers text;
According to product description standard and its dictionary, product information is obtained, determines product description text using the product information;
According to the files on each of customers text and the product description text, vector space, in the way of similarity assessment, In are generated
User to be recommended and corresponding recommended products are determined in the vector space.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910700946.1A CN110400202A (en) | 2019-07-31 | 2019-07-31 | Bank product recommended method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910700946.1A CN110400202A (en) | 2019-07-31 | 2019-07-31 | Bank product recommended method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110400202A true CN110400202A (en) | 2019-11-01 |
Family
ID=68326893
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910700946.1A Pending CN110400202A (en) | 2019-07-31 | 2019-07-31 | Bank product recommended method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110400202A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111383095A (en) * | 2020-03-11 | 2020-07-07 | 中国工商银行股份有限公司 | Method and device for matching public deposit products |
CN112613906A (en) * | 2020-12-18 | 2021-04-06 | 北京思特奇信息技术股份有限公司 | Marketing recommendation evaluation method and device based on mobile communication network and computer equipment |
CN113592605A (en) * | 2021-08-10 | 2021-11-02 | 平安银行股份有限公司 | Product recommendation method, device, equipment and storage medium based on similar products |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020130868A1 (en) * | 2000-11-28 | 2002-09-19 | Aston Guardian Limited | Method and apparatus for providing financial instrument interface |
CN106097044A (en) * | 2016-06-01 | 2016-11-09 | 腾讯科技(深圳)有限公司 | A kind of data recommendation processing method and device |
CN109636510A (en) * | 2018-11-28 | 2019-04-16 | 阿里巴巴集团控股有限公司 | A kind of determining consumer's risk preference, the recommended method of finance product and device |
CN109636538A (en) * | 2018-12-20 | 2019-04-16 | 成都知数科技有限公司 | Bank product recommended method, device and server |
CN109635184A (en) * | 2018-11-02 | 2019-04-16 | 平安科技(深圳)有限公司 | Financial product recommended method, device and computer equipment based on data analysis |
-
2019
- 2019-07-31 CN CN201910700946.1A patent/CN110400202A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020130868A1 (en) * | 2000-11-28 | 2002-09-19 | Aston Guardian Limited | Method and apparatus for providing financial instrument interface |
CN106097044A (en) * | 2016-06-01 | 2016-11-09 | 腾讯科技(深圳)有限公司 | A kind of data recommendation processing method and device |
CN109635184A (en) * | 2018-11-02 | 2019-04-16 | 平安科技(深圳)有限公司 | Financial product recommended method, device and computer equipment based on data analysis |
CN109636510A (en) * | 2018-11-28 | 2019-04-16 | 阿里巴巴集团控股有限公司 | A kind of determining consumer's risk preference, the recommended method of finance product and device |
CN109636538A (en) * | 2018-12-20 | 2019-04-16 | 成都知数科技有限公司 | Bank product recommended method, device and server |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111383095A (en) * | 2020-03-11 | 2020-07-07 | 中国工商银行股份有限公司 | Method and device for matching public deposit products |
CN112613906A (en) * | 2020-12-18 | 2021-04-06 | 北京思特奇信息技术股份有限公司 | Marketing recommendation evaluation method and device based on mobile communication network and computer equipment |
CN112613906B (en) * | 2020-12-18 | 2024-04-30 | 北京思特奇信息技术股份有限公司 | Marketing recommendation evaluation method and device based on mobile communication network and computer equipment |
CN113592605A (en) * | 2021-08-10 | 2021-11-02 | 平安银行股份有限公司 | Product recommendation method, device, equipment and storage medium based on similar products |
CN113592605B (en) * | 2021-08-10 | 2023-08-22 | 平安银行股份有限公司 | Product recommendation method, device, equipment and storage medium based on similar products |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
See-To et al. | Customer reviews for demand distribution and sales nowcasting: a big data approach | |
Langley et al. | The platform political economy of FinTech: Reintermediation, consolidation and capitalisation | |
Athique | Integrated commodities in the digital economy | |
CN110400202A (en) | Bank product recommended method and system | |
CN107341173A (en) | A kind of information processing method and device | |
US9390168B1 (en) | Customer keyword-based item recommendations | |
McStay | 3.1 Micro-moments, liquidity, intimacy and automation: Developments in programmatic Ad-tech | |
CN108122153A (en) | Personalized recommendation method based on cloud computing tupe under e-commerce environment | |
US20190303494A1 (en) | Node linkage in entity graphs | |
US9460163B1 (en) | Configurable extractions in social media | |
Kuah et al. | Fast‐Expanding “Online” Markets in South Korea and China: Are They Worth Pursuing? | |
Bu et al. | Mining analysis of customer perceived value of online customisation experience under social commerce | |
US9967297B1 (en) | Generating item suggestions from a profile-based group | |
Sheriff | Big Data Revolution: Is It a Business Disruption? | |
Edi Surya et al. | Recommendation System with Content-Based Filtering in NFT Marketplace | |
Enahoro et al. | GSM and the Nigerian Economy: The Journey from 2004 to 2019 | |
Shen et al. | Big data overview | |
CN110020118B (en) | Method and device for calculating similarity between users | |
Prashar et al. | Promoting businesses beyond borders: Multinationals’ perspectives | |
CN114996579A (en) | Information pushing method and device, electronic equipment and computer readable medium | |
Truong et al. | Machine learning optimization in computational advertising—A systematic literature review | |
Khan et al. | Information technology adaptation in Indian small and medium sized enterprises: opportunities and challenges ahead | |
Deshmukh et al. | Transformative effect of e-commerce and m-commerce on national productivity in India' | |
Ragab et al. | Embedding Based Recommender systems, a review and comparison. | |
Han | The design and implementation of E-commerce personalized services based on collaborative filtering recommendation system |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191101 |