CN111652738A - Insurance product recommendation method based on user behavior weight - Google Patents
Insurance product recommendation method based on user behavior weight Download PDFInfo
- Publication number
- CN111652738A CN111652738A CN202010307239.9A CN202010307239A CN111652738A CN 111652738 A CN111652738 A CN 111652738A CN 202010307239 A CN202010307239 A CN 202010307239A CN 111652738 A CN111652738 A CN 111652738A
- Authority
- CN
- China
- Prior art keywords
- commodity
- user
- recommendation
- weight
- 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
- 238000000034 method Methods 0.000 title claims abstract description 21
- 230000006399 behavior Effects 0.000 claims abstract description 54
- 238000004364 calculation method Methods 0.000 claims abstract description 6
- 238000012163 sequencing technique Methods 0.000 claims abstract description 4
- 230000003542 behavioural effect Effects 0.000 claims 2
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 208000003443 Unconsciousness Diseases 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
-
- 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
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Engineering & Computer Science (AREA)
- Marketing (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Technology Law (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention provides a user behavior weight-based insurance product recommendation method, which comprises the following steps: the first step is as follows: according to the current operation of a user on a platform for a commodity, collecting user behaviors related to the commodity, expressing the commodity as characteristic contents, and quantizing the collected user behaviors into a label; the second step is as follows: forming a multi-dimensional recommendation association relation according to the collected user behaviors; the third step: dividing different recommendation rules for a user, and selecting the recommendation rules according to the multi-dimensional recommendation association relation to obtain an initial recommendation result; the fourth step: and sequencing the initial recommendation results through weighting calculation to generate a final recommendation result of the commodity based on the operation of the user on the platform for the commodity in a preset time period.
Description
Technical Field
The invention relates to the field of insurance, in particular to a method for recommending insurance products based on user behavior weight.
Background
For each large insurance platform, the recommendation rules based on the basic information of the user, such as the basic information of region, age, gender, occupation, etc., are taken as the basis. And finding the similarity degree between other users according to the basic information between the users, and recommending other articles favored by the similar users to the current user. The rules do not relate to the attributes of the commodities and the user behavior data, and meanwhile, because the amount of the basic information data of the user is too small, the recommendation accuracy based on the data is low, and therefore the recommendation according to the user tags is realized. The label is the most important data basis of all recommendation systems, and the label effectively standardizes non-standard commodities.
The user tag is established for the mobile application, and is generally judged according to the operation behavior of the user. Such as user browsing, sharing, collecting, searching, ordering, etc. Each behavior has a certain degree of proportion, the operation behavior of the user on the commodity reflects the interest degree of the user on the commodity containing the labels, and finally, certain labels of the commodity are given to the user according to a certain algorithm rule, so that the labels of the user are established.
Due to scoring of indexes such as collection conditions, sales conditions and good evaluation rates of the commodities of the amount of the interview, a proper personalized intervention and business pressing rule is added, but the phenomenon that the specific gravity of characteristic scoring of the commodities is possibly greater than that of actual behavior scoring of the users due to the fact that a certain dimension of the commodities is over-prominent, the commodity recommendation result presented by the users is finally normalized with the behavior scoring of the users may occur.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a method for recommending insurance products based on user behavior weight, which can better reflect the self needs of users, aiming at the defects in the prior art.
According to the invention, a method for recommending insurance products based on user behavior weight is provided, which comprises the following steps:
the first step is as follows: according to the current operation of a user on a platform for a commodity, collecting user behaviors related to the commodity, expressing the commodity as characteristic contents, and quantizing the collected user behaviors into a label;
the second step is as follows: forming a multi-dimensional recommendation association relation according to the collected user behaviors;
the third step: dividing different recommendation rules for a user, and selecting the recommendation rules according to the multi-dimensional recommendation association relation to obtain an initial recommendation result;
the fourth step: and sequencing the initial recommendation results through weighting calculation to generate a final recommendation result of the commodity based on the operation of the user on the platform for the commodity in a preset time period.
Preferably, in the second step, the commodities with the correlation degree are determined according to the collected user behaviors, the commodities with the correlation characteristics are determined according to the user tags, and the commodities concerned by the users with the similar behaviors are determined, so that the multi-dimensional recommendation association relationship is formed.
Preferably, in the third step, after the initial recommendation result is obtained, the operator may perform manual intervention on the initial recommendation result to decide that the initial recommendation result is filtered.
Preferably, in the fourth step, the first-class commodity operation records in the preset time period of the user are collected, and the user behavior weight value of the commodity is calculated according to the first-class commodity operation records of the commodity; collecting second commodity operation records of a user in a preset time period, and calculating a commodity attribute weight value of the commodity according to the second commodity operation records of the commodity; and then adding the user behavior weight value and the commodity attribute weight value to obtain a total weight value of the commodity.
Preferably, the first-class commodity operation record comprises a search operation record, a browsing operation record, an ordering operation record and a payment operation record for the commodity.
Preferably, the second type of commodity operation record includes a collection operation record, a sales record and a good rating record.
Preferably, the search operation record, the browse operation record and the order placing operation record are subjected to weight assignment in an increasing manner, the weight of the payment operation record is assigned to be 0 in the case that the payment operation record exists, the weight of the payment operation record is assigned to be a positive number in the case that the payment operation record does not exist, then the sum of the product of the search operation record weight multiplied by the search operation times, the product of the browse operation record weight multiplied by the browse operation times and the product of the order placing operation record weight multiplied by the search operation times is calculated, and the sum is multiplied by the payment operation record weight assignment, so that the user behavior weight value of the commodity is calculated.
Preferably, the collection operation record, the sales record and the goodness record are respectively subjected to weight assignment, and the sum of the product of the collection operation record weight value multiplied by the collection operation times, the product of the sales record weight value multiplied by the sales volume and the product of the goodness record weight value multiplied by the goodness is calculated as the commodity attribute weight value of the commodity.
The recommendation method based on the user behaviors and the commodity attributes can reflect the self requirements of the user and help the user to make recommendations and decisions in a profound and impersonal manner in operation, so that the order conversion rate of commodities is improved, and the income is increased for a company.
Drawings
A more complete understanding of the present invention, and the attendant advantages and features thereof, will be more readily understood by reference to the following detailed description when considered in conjunction with the accompanying drawings wherein:
FIG. 1 schematically illustrates an overall flow chart of a method for user behavior weight-based insurance product recommendation, in accordance with a preferred embodiment of the present invention.
Fig. 2 schematically shows an example of a recommendation relationship of a user with an article.
FIG. 3 schematically shows an example of a user behavior and product attribute calculation process.
It is to be noted, however, that the appended drawings illustrate rather than limit the invention. It is noted that the drawings representing structures may not be drawn to scale. Also, in the drawings, the same or similar elements are denoted by the same or similar reference numerals.
Detailed Description
In order that the present disclosure may be more clearly and readily understood, reference will now be made in detail to the present disclosure as illustrated in the accompanying drawings.
The invention proposes that different behaviors of the user should be given different weights; for the condition that the weight has no clear data for guiding the weight rule in cold start, firstly, the scene analysis and the psychological needs and experience of the user are carried out (at the moment, the division of the user behavior into weights is wrong), then, the users are clustered according to the behavior data and the result of the users, and then, the weights are corrected and optimized.
In particular, FIG. 1 schematically illustrates an overall flow diagram of a method for user behavior weight-based insurance product recommendation, in accordance with a preferred embodiment of the present invention. As shown in fig. 1, the method for user behavior weight-based insurance product recommendation according to a preferred embodiment of the present invention includes:
first step S1: according to the current operation of a user on a platform for a commodity, collecting user behaviors related to the commodity, expressing the commodity as characteristic contents, and quantizing the collected user behaviors into a label;
second step S2: forming a multi-dimensional recommendation association relation according to the collected user behaviors;
for example, in the second step, according to the collected user behaviors, the commodities with the correlation degree are determined, the commodities with the correlation characteristics are determined according to the user tags, and the commodities, which are concerned by the users with the similar behaviors, are determined, so that a multi-dimensional recommendation association relationship is formed, for example, as shown in fig. 2.
Third step S3: dividing different recommendation rules for a user, and selecting the recommendation rules according to the multi-dimensional recommendation association relation to obtain an initial recommendation result;
preferably, in the third step, after the initial recommendation result is obtained, the operator may perform manual intervention on the initial recommendation result to decide that the initial recommendation result is filtered.
Fourth step S4: and sequencing the initial recommendation results through weighting calculation to generate a final recommendation result of the commodity based on the operation of the user on the platform for the commodity in a preset time period.
Specifically, for example, a first-class commodity operation record in a predetermined time period of a user is collected (for example, the first-class commodity operation record includes a search operation record, a browsing operation record, an ordering operation record and a payment operation record for the commodity), and a user behavior weight value of the commodity is calculated according to the first-class commodity operation record of the commodity; collecting a second type of commodity operation record (for example, the second type of commodity operation record comprises a collection operation record, a sales record and a good rating record) in a preset time period of the user, and calculating a commodity attribute weight value of the commodity according to the second type of commodity operation record of the commodity; and then adding the user behavior weight value and the commodity attribute weight value to obtain a total weight value of the commodity.
More specifically, for example, weight assignment is performed incrementally on the search operation record, the browsing operation record, and the placing operation record, the weight of the payment operation record is assigned to 0 in the case where the payment operation record exists, the weight of the payment operation record is assigned to a positive number (e.g., 1) in the case where the payment operation record does not exist, then the sum of the product of the search operation record weight multiplied by the number of search operations, the product of the browsing operation record weight multiplied by the number of browsing operations, and the product of the placing operation record weight multiplied by the number of search operations is calculated, and the sum is multiplied by the payment operation record weight, thereby calculating the user behavior weight value of the commodity.
Further, more specifically, for example, weight assignment is performed on each of the collection operation record, the sales record, and the rating score record, and the sum of the product of the collection operation record weight value multiplied by the number of collection operations, the product of the sales record weight value multiplied by the sales volume, and the product of the rating score record weight value multiplied by the rating score is calculated as the product attribute weight value of the commodity.
For example, if browsing is 1 point, the interested intention can be represented by clicking to enter detailed browsing; the search is 2 points, and if the product is searched, the intention is strong; the order placing is 3 minutes, and the user is willing to apply insurance when arriving at the order placing environment; the payment needs to be taken by 0, and the same type of product is underwritten because the user has already applied the insurance and does not need to be recommended again.
The method comprises the steps of performing score quantification on the attributes of commodities in a database, calculating reference indexes of the commodities such as collection times, sales volume, good rate and the like, then calculating user behavior scores, combining with a ranking strategy (such as a RANK function), normalizing the scores, and then presenting final results to a user, wherein for example, fig. 3 shows weight calculation examples for five commodities A, B, C, D, E (wherein M1, N1, M2, N2, M3, N3, M4, N4, M5 and N5 respectively represent collection and sales proportions).
Therefore, the recommendation method based on the user behaviors and the commodity attributes can reflect the self requirements of the user and help the user to make recommendations and decisions in a latent and unconscious manner in operation, so that the order conversion rate of commodities is improved, and the income is increased for companies.
It should be noted that the terms "first", "second", "third", and the like in the description are used for distinguishing various components, elements, steps, and the like in the description, and are not used for indicating a logical relationship or a sequential relationship between the various components, elements, steps, and the like, unless otherwise specified.
It is to be understood that while the present invention has been described in conjunction with the preferred embodiments thereof, it is not intended to limit the invention to those embodiments. It will be apparent to those skilled in the art from this disclosure that many changes and modifications can be made, or equivalents modified, in the embodiments of the invention without departing from the scope of the invention. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention are still within the scope of the protection of the technical solution of the present invention, unless the contents of the technical solution of the present invention are departed.
Claims (8)
1. A method for user behavior weight-based insurance product recommendation, comprising:
the first step is as follows: according to the current operation of a user on a platform for a commodity, collecting user behaviors related to the commodity, expressing the commodity as characteristic contents, and quantizing the collected user behaviors into a label;
the second step is as follows: forming a multi-dimensional recommendation association relation according to the collected user behaviors;
the third step: dividing different recommendation rules for a user, and selecting the recommendation rules according to the multi-dimensional recommendation association relation to obtain an initial recommendation result;
the fourth step: and sequencing the initial recommendation results through weighting calculation to generate a final recommendation result of the commodity based on the operation of the user on the platform for the commodity in a preset time period.
2. The method for recommending insurance products based on user behavior weights according to claim 1, wherein in the second step, commodities with correlation degrees are determined according to the collected user behaviors, commodities with correlation characteristics are determined according to the user tags, and commodities concerned by users with similar behaviors are determined, so that a multi-dimensional recommendation association relationship is formed.
3. The method for user behavior weight-based insurance product recommendation according to claim 1 or 2, wherein in the third step, after the initial recommendation result is obtained, the operator may perform manual intervention on the initial recommendation result to decide the initial recommendation result to filter.
4. The method for recommending insurance products based on user's behavioral weights according to claim 1 or 2, characterized in that, at the fourth step, the first-class commodity operation records within a predetermined time period of the user are collected, and the user's behavioral weight values of the commodities are calculated according to the first-class commodity operation records of the commodities; collecting second commodity operation records of a user in a preset time period, and calculating a commodity attribute weight value of the commodity according to the second commodity operation records of the commodity; and then adding the user behavior weight value and the commodity attribute weight value to obtain a total weight value of the commodity.
5. The method for user behavior weight-based insurance product recommendation according to claim 1 or 2, wherein the first-class commodity operation records comprise a search operation record, a browse operation record, an order placing operation record and a payment operation record for the commodity.
6. The method for user behavior weight-based insurance product recommendation according to claim 1 or 2, wherein the second type merchandise operation records include a collection operation record, a sales record and a good rate record.
7. The method for user behavior weight-based insurance product recommendation according to claim 1 or 2, wherein weight assignment is incrementally performed on the search operation record, the browsing operation record, and the placing operation record, the weight of the payment operation record is assigned to 0 in the case where the payment operation record exists, the weight of the payment operation record is assigned to a positive number in the case where the payment operation record does not exist, and then a sum of a product of the search operation record weight multiplied by the number of search operations, a product of the browsing operation record weight multiplied by the number of browsing operations, and a product of the placing operation record weight multiplied by the number of search operations is calculated, and the sum is multiplied by the payment operation record weight assignment, thereby calculating the user behavior weight value of the product.
8. The method of user behavior weight-based insurance product recommendation according to claim 1 or 2, wherein weight assignment is performed for each of the collection operation record, the sales record and the goodness record, and a sum of a product of a collection operation record weight value multiplied by the collection operation times, a product of a sales record weight value multiplied by the sales volume, and a product of a goodness record weight value multiplied by the goodness is calculated as the commodity attribute weight value of the commodity.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010307239.9A CN111652738A (en) | 2020-04-17 | 2020-04-17 | Insurance product recommendation method based on user behavior weight |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010307239.9A CN111652738A (en) | 2020-04-17 | 2020-04-17 | Insurance product recommendation method based on user behavior weight |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111652738A true CN111652738A (en) | 2020-09-11 |
Family
ID=72348478
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010307239.9A Pending CN111652738A (en) | 2020-04-17 | 2020-04-17 | Insurance product recommendation method based on user behavior weight |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111652738A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090164453A1 (en) * | 2007-12-21 | 2009-06-25 | Glyde Corporation | System and method for providing real-time search results on merchandise |
| JP2010073170A (en) * | 2008-09-22 | 2010-04-02 | Yahoo Japan Corp | Recommended commodity selection device, recommended commodity selection program and commodity retrieval device |
| JP2012027757A (en) * | 2010-07-26 | 2012-02-09 | Dainippon Printing Co Ltd | Commodity recommendation device, commodity recommendation system, commodity recommendation method, and program |
| WO2016058485A2 (en) * | 2014-10-15 | 2016-04-21 | 阿里巴巴集团控股有限公司 | Methods and devices for calculating ranking score and creating model, and product recommendation system |
| CN105809475A (en) * | 2016-02-29 | 2016-07-27 | 南京大学 | Commodity recommendation method compatible with O2O applications in internet plus tourism environment |
| CN106021337A (en) * | 2016-05-09 | 2016-10-12 | 房加科技（北京）有限公司 | A big data analysis-based intelligent recommendation method and system |
| CN107239495A (en) * | 2017-04-28 | 2017-10-10 | 杭州集盒网络技术有限公司 | Commodity and searching order methods of exhibiting based on relation chain weight |
| JP2018063484A (en) * | 2016-10-11 | 2018-04-19 | 凸版印刷株式会社 | User's evaluation prediction system, user's evaluation prediction method and program |
| CN109598588A (en) * | 2018-12-04 | 2019-04-09 | 广州拓飞商贸有限公司 | A kind of online merchandise display method based on big data analysis |
-
2020
- 2020-04-17 CN CN202010307239.9A patent/CN111652738A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090164453A1 (en) * | 2007-12-21 | 2009-06-25 | Glyde Corporation | System and method for providing real-time search results on merchandise |
| JP2010073170A (en) * | 2008-09-22 | 2010-04-02 | Yahoo Japan Corp | Recommended commodity selection device, recommended commodity selection program and commodity retrieval device |
| JP2012027757A (en) * | 2010-07-26 | 2012-02-09 | Dainippon Printing Co Ltd | Commodity recommendation device, commodity recommendation system, commodity recommendation method, and program |
| WO2016058485A2 (en) * | 2014-10-15 | 2016-04-21 | 阿里巴巴集团控股有限公司 | Methods and devices for calculating ranking score and creating model, and product recommendation system |
| CN105809475A (en) * | 2016-02-29 | 2016-07-27 | 南京大学 | Commodity recommendation method compatible with O2O applications in internet plus tourism environment |
| CN106021337A (en) * | 2016-05-09 | 2016-10-12 | 房加科技（北京）有限公司 | A big data analysis-based intelligent recommendation method and system |
| JP2018063484A (en) * | 2016-10-11 | 2018-04-19 | 凸版印刷株式会社 | User's evaluation prediction system, user's evaluation prediction method and program |
| CN107239495A (en) * | 2017-04-28 | 2017-10-10 | 杭州集盒网络技术有限公司 | Commodity and searching order methods of exhibiting based on relation chain weight |
| CN109598588A (en) * | 2018-12-04 | 2019-04-09 | 广州拓飞商贸有限公司 | A kind of online merchandise display method based on big data analysis |
Non-Patent Citations (2)
| Title |
|---|
| 谢人强等: "基于共同评分项和权重计算的推荐算法研究", 《计算机技术与发展》 * |
| 马汉达等: "一种基于贝叶斯分类的个性化导购推荐算法", 《信息技术》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7805431B2 (en) | System and method for generating a display of tags | |
| CN108629665B (en) | Personalized commodity recommendation method and system | |
| CN106844787B (en) | Recommendation method for searching target users and matching target products for automobile industry | |
| CN109064285B (en) | Commodity recommendation sequence and commodity recommendation method | |
| CN111080398B (en) | Commodity recommendation method, commodity recommendation device, computer equipment and storage medium | |
| US8548996B2 (en) | Ranking content items related to an event | |
| CN109033101B (en) | Label recommendation method and device | |
| CN109582876B (en) | Tourist industry user portrait construction method and device and computer equipment | |
| US20150310528A1 (en) | Distinguishing accessories from products for ranking search results | |
| Wang et al. | A strategy-oriented operation module for recommender systems in E-commerce | |
| CN105183727A (en) | Method and system for recommending book | |
| CN103473354A (en) | Insurance recommendation system framework and insurance recommendation method based on e-commerce platform | |
| US20220245706A1 (en) | Methods and apparatuses for providing search results using embedding-based retrieval | |
| CN105913290A (en) | Commodity matching recommending method and recommending system | |
| Badriyah et al. | Recommendation system for property search using content based filtering method | |
| CN115131101A (en) | Individualized intelligent recommendation system for insurance products | |
| CN111310038A (en) | Information recommendation method and device, electronic equipment and computer-readable storage medium | |
| US20190065611A1 (en) | Search method and apparatus | |
| CN110033324A (en) | Data processing method, device, electronic equipment and computer readable storage medium | |
| CN113052651A (en) | Intelligent retail management method and system based on big data | |
| CN104615741A (en) | Cloud computing based cold start item recommending method and device | |
| CN115760202A (en) | Product operation management system and method based on artificial intelligence | |
| CN111861679A (en) | Commodity recommendation method based on artificial intelligence | |
| CN117495482B (en) | Secondhand mobile phone sales recommendation method and system based on user portrait | |
| US6868299B2 (en) | Generating a sampling plan for testing generated content |
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: 20200911 |