Movatterモバイル変換


[0]ホーム

URL:


CN119722231A - Bank information push method based on big data model - Google Patents

Bank information push method based on big data model
Download PDF

Info

Publication number
CN119722231A
CN119722231ACN202411796176.2ACN202411796176ACN119722231ACN 119722231 ACN119722231 ACN 119722231ACN 202411796176 ACN202411796176 ACN 202411796176ACN 119722231 ACN119722231 ACN 119722231A
Authority
CN
China
Prior art keywords
target user
merchant
coupon
preferential
information
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202411796176.2A
Other languages
Chinese (zh)
Inventor
侯丽群
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong Nayun Technology Co ltd
Original Assignee
Guangdong Nayun Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guangdong Nayun Technology Co ltdfiledCriticalGuangdong Nayun Technology Co ltd
Priority to CN202411796176.2ApriorityCriticalpatent/CN119722231A/en
Publication of CN119722231ApublicationCriticalpatent/CN119722231A/en
Pendinglegal-statusCriticalCurrent

Links

Landscapes

Abstract

Translated fromChinese

本发明公开一种基于大数据模型的银行信息推送方法,包括:通过在用户授权的情况下获取目标用户每天的消费历史明细并进行关键词提取以得出用户每天的消费商户信息,进而构建目标用户的消费商户特征识别模型;定时通过目标用户携带的移动终端获取目标用户当前所在的地址信息,并根据目标用户当前所在的地址信息以及目标用户消费商户特征识别模型进行识别得出目标用户感兴趣的优惠商户,进而根据目标用户感兴趣的优惠商户推送到移动终端的手机银行APP,同时将每个优惠商户的优惠信息生成优惠任务并存储到服务器中。本发明能够实现银行优惠信息的个性化推荐。

The present invention discloses a bank information push method based on a big data model, including: obtaining the daily consumption history details of a target user and extracting keywords to obtain the user's daily consumption merchant information under the authorization of the user, and then building a consumption merchant feature recognition model for the target user; regularly obtaining the current address information of the target user through a mobile terminal carried by the target user, and identifying the preferential merchants that the target user is interested in based on the current address information of the target user and the target user's consumption merchant feature recognition model, and then pushing the preferential merchants that the target user is interested in to the mobile banking APP of the mobile terminal, and generating preferential tasks for the preferential information of each preferential merchant and storing them in a server. The present invention can realize personalized recommendation of bank preferential information.

Description

Bank information pushing method based on big data model
Technical Field
The invention relates to bank information pushing, in particular to a bank information pushing method based on a big data model.
Background
At present, a banking system pushes some preferential information to a user's equipment end in a short message or APP notification mode at regular time. The push has information confusion, users need to screen out interesting notifications from more information, bad experience is brought to the users, and meanwhile, the push of the information is uniform push, and personalized requirements of the users cannot be met. Meanwhile, due to the fact that more short message fraud occurs at present, many people shield or select short message information from unknown sources, so that preferential information cannot be pushed to users with requirements.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention aims to provide a bank information pushing method based on a big data model, which can solve the problems that the bank information pushing information in the prior art is disordered and the personalized requirements of users cannot be met.
The invention adopts the following technical scheme:
The bank information pushing method based on the big data model comprises the following steps:
The method comprises the steps of obtaining historical daily consumption details of a target user under the condition of user authorization, extracting keywords from the historical daily consumption details of the target user to obtain daily consumption merchant information of the target user, and then constructing a consumption merchant feature recognition model of the target user by combining a neural network model;
The screening step is that address information of a target user is acquired through a mobile terminal carried by the target user at regular time, all preferential merchants corresponding to the authorized bank types of the target user in a preset range are obtained according to the address information of the target user, and each obtained preferential merchant is matched and identified with a consumer merchant feature identification model of the target user so as to screen out a plurality of candidate preferential merchants of the target user;
and the recommending step is to acquire the merchant coupon information of each candidate coupon merchant according to a plurality of candidate coupon merchants of the target user, generate coupon notices according to the merchant coupon information of each candidate coupon merchant and sequentially push the coupon notices to a mobile phone bank APP of a mobile terminal carried by the target user, and store the merchant coupon information of each candidate coupon merchant in a server.
The model construction step further comprises the steps of constructing a data set according to the collected daily consumption merchant information of the target user, dividing the data set into a training set and a verification set, training a pre-constructed neural network model according to the training set, verifying the trained model according to the verification set, and obtaining a consumption merchant feature identification model of the target user according to the verified model after verification is passed, wherein the consumption merchant feature identification model of the target user comprises the association relation between the target user and the consumption merchant features of the target user.
The model construction step further comprises the steps of obtaining the working days and the rest days of the target user through obtaining configuration information of the target user, dividing a data set into the working day data set and the rest day data set, and then respectively carrying out model training and verification according to the working day data set, the rest day data set and a pre-constructed neural network model to obtain a consumption merchant feature identification model of the working days of the target user and a consumption merchant feature identification model of the rest days of the target user.
Further, the screening step includes obtaining the date of the day and selecting a corresponding consuming merchant feature recognition model according to the date of the day to be matched and recognized so as to screen out a plurality of candidate preferential merchants of the target user, wherein each preferential merchant is matched and recognized with the consuming merchant feature recognition model of the target user on the working day if the date of the day is the working day, and is matched with the consuming merchant feature recognition model of the target user on the rest day if the date of the day is the rest day.
Further, the recommending step further comprises the step of displaying the latest pushed preferential notice on the main page of the mobile phone bank APP in a popup window mode when the mobile phone bank APP on the mobile terminal is detected to be opened.
Further, after the recommending step, clicking operation of the target user is obtained when the corresponding coupon is read or checked, merchant coupon information of the corresponding coupon is obtained from the server, and the merchant coupon information of the corresponding coupon is displayed to the target user, wherein the merchant coupon information of the corresponding coupon includes a coupon merchant name, a coupon merchant address, a coupon merchant type, a coupon merchant picture, a coupon amount, a coupon time, a coupon payment mode and a coupon participation brief description.
Further, the recommending step further comprises timing according to the pushing time of the corresponding preferential notice when the corresponding preferential notice is pushed, and when the timing time exceeds a first preset pushing time length and the corresponding preferential notice is not read or checked, generating the short message notice according to the short message notice template and pushing the short message notice to the short message end of the mobile terminal bound by the target user.
Further, the recommending step further comprises deleting the corresponding preferential notice from the mobile banking APP of the mobile terminal when the preferential expiration of the corresponding preferential notice or the pushing time of the corresponding preferential notice exceeds a second preset pushing time.
Further, the recommending step further comprises the steps of calculating the priority of each candidate preferential merchant of the target user according to a preset priority calculating rule, and pushing preferential notices of each candidate preferential merchant to the mobile banking APP of the mobile terminal in sequence according to the order of the priority.
Compared with the prior art, the invention has the beneficial effects that:
According to the invention, the historical consumption details of the target user are subjected to data analysis, learning and other researches by adopting a machine learning model to obtain the characteristic data of the commercial tenant consumed by the user, then the user is screened, matched with the preferential commercial tenant corresponding to the authorized bank according to the address information data of the user, and timely pushed to the mobile phone bank APP of the user, so that the preferential information of the commercial tenant with the preferential and user interested in the user is timely notified to the user for checking, consuming and the like, the user is prevented from missing the preferential, the personalized preferential recommendation is realized, the user activity of the bank is relatively increased, the user viscosity of the user to the bank is ensured, and the acceptance and experience of the user to the bank are improved.
Drawings
Fig. 1 is a flowchart of a bank information pushing method based on a big data model.
Detailed Description
The present invention will be further described with reference to the accompanying drawings and detailed description, wherein it is to be understood that, on the premise of no conflict, the following embodiments or technical features may be arbitrarily combined to form new embodiments.
Based on the defects of the existing bank information pushing, the invention provides a bank information pushing method based on a big data model, as shown in fig. 1, comprising the following steps:
Step S1, acquiring daily consumption history details of a target user under the condition of authorization of the target user, extracting keywords from the daily consumption history details of the target user to obtain daily consumption merchant information of the user, and constructing a consumption merchant feature recognition model of the target user according to the daily consumption merchant information of the user.
Specifically, in this embodiment, the message pushing is implemented in the mobile phone bank APP installed by the user, so the target user here refers to the user who opens the mobile phone bank of the corresponding bank, and these users need to install the corresponding mobile phone bank APP in the mobile terminal carried by these users.
In addition, in order to ensure privacy, the daily consumption history details of the target user are obtained under the condition of user authorization in the embodiment, for example, specific consumption details can be provided through WeChat, payment treasury and various bank cards. When the user is authorized, the daily consumption history details of the target user can be obtained and keyword extraction can be carried out on the daily consumption history details of the target user so as to obtain daily consumption merchant information of the user. The information of the consuming merchant includes information such as merchant type, merchant name, merchant address, consumption time, payment cost, payment mode, whether there is a payment offer, offer mode, and offer amount.
And carrying out data analysis and mining on the daily consumer merchant information of the user to construct a daily consumer merchant feature recognition model of the user, so that interested merchant information of the user is screened according to the consumer merchant feature recognition model, and personalized preferential recommendation of the user is realized.
More specifically, the daily consumption merchant information of the user is obtained and a data set is constructed by collecting daily consumption history details of the user in the history and extracting keywords, and the data set is divided into a training set and a verification set.
And training the neural network model according to the training set to obtain a trained neural network model through a pre-constructed neural network model, verifying the trained neural network model according to the verification set to judge whether the trained neural network model meets the requirements, if so, defining the trained neural network model as a consumer business feature recognition model of the target user, and if not, returning to retrain until the trained consumer business feature recognition model of the target user is obtained.
The invention realizes the learning and mining of data by utilizing a machine learning model, such as a neural network model, so as to obtain the characteristics of merchants consumed by the user every day, so as to recommend the interested preferential merchants for the user, and realize the pushing of the preferential merchants, the personalized pushing of the user can be realized, the user is guided to consume further, and the problems that in the prior art, preferential consumption pushing is disordered and not accurate enough, the requirement of the user can not be met, the user experiences a bank and the user viscosity is poor, the user activity is reduced and the like are solved.
And S2, acquiring the current address information of the target user through a mobile terminal carried by the target user at regular time, obtaining all preferential merchants corresponding to the authorized bank types of the target user in a preset range according to the current address information of the target user, and matching each preferential merchant with a consumer merchant feature identification model of the target user to identify the preferential merchants interested in the target user.
Specifically, in this embodiment, address information of a target user is obtained through a mobile terminal carried by the user, and then a corresponding area is defined by centering on the address information according to a preset rule to screen merchant information for the target user. More preferably, when screening the merchant information, screening can be performed according to preset keywords, for example, the merchant type is food, entertainment, furniture and the like. Meanwhile, the type of the commercial tenant which is interested by the user can be estimated according to the current address information of the user, for example, the type of the user is mostly food, entertainment and the like when the user is in a office building or a mall, and on the contrary, the type of the user is mostly furniture when the user is in a furniture mall, and the type of the user is probably food and the like, and the specific selection can be selected according to the actual interest. In addition, the present application recommends to the user merchants with the bank offers, that is, when screening the user, the merchants with the bank offers need to be screened, and for some merchants without the bank offers, the merchants with the bank offers can be ignored.
In addition, the target user authorizes the bank type, except the current bank type corresponding to the mobile phone bank APP, and simultaneously, the mobile phone bank APP can bind the bank cards of other banks under the target user name, so that the bank type of the bank card and the bank card type under the target user name can be caused by the mobile phone bank APP. Among the types of bank cards are debit cards and credit cards.
Moreover, the preset range in this embodiment refers to a certain reasonable range, such as a take-out distribution range, or an area reachable within 20 minutes of driving, etc. Normally, for a user, the user consumes the product for a merchant, firstly by taking out the product, and secondly directly consuming the product to a store. When the user of the obtained recommendation is far away, the user does not choose to run a long distance for some preferential ways to choose consumption, so that a consumption range can be set according to actual experience to improve the consumption possibility of the user. In the range, a reasonable consumption area can be obtained by mining according to the daily actual consumption requirement of the user and the distance between the household address, the work place and the like of the user and the consumption merchant.
And after obtaining all the preferential merchants corresponding to the target user authorized bank types in the preset range, carrying out matching identification on each preferential merchant and the target user consumption merchant characteristic identification model to obtain the preferential merchants interested by the target user. The consumer merchant feature recognition model of the target user is obtained by mining according to the daily consumer details of the user, so that the screened preferential merchants can be matched with the pre-trained model to obtain the target merchants interested by the user.
The characteristics of the consumers of the user are mined, so that interested preferential merchants are recommended to the user, the user is guided to consume, and better experience is provided for the user.
And step S3, acquiring the preferential information of each preferential merchant according to the preferential merchant interested by the target user and generating a preferential notice, then pushing the preferential notice to the mobile banking APP of the mobile terminal at regular time, and simultaneously generating the preferential task of the preferential information of each preferential merchant and storing the preferential task into a server.
Specifically, in this embodiment, in order to avoid that all the data of the coupon information is pushed to the mobile phone bank APP end, a jam or the like is caused to the mobile phone of the user, which affects the user experience, so that when generating the coupon notification, this embodiment generates a corresponding coupon notification according to the coupon information, for example, generates a coupon poster in a picture manner, and directly pushes the coupon poster to the mobile phone bank APP of the mobile terminal. The method can avoid equipment jamming and the like caused by storing all data in the mobile terminal, and in addition, the poster is simpler, can be more attractive to a user to view, and improves the interestingness of the user. The preferential poster is pushed to the APP end, and when a user is interested in viewing the poster, buttons such as immediate participation on the preferential poster are clicked, so that preferential data related to preferential are called from the server, and the user can further know preferential details and judge whether to participate in preferential.
That is, when the coupon notification is read, if the target user is interested, the operation button on the poster may be clicked to jump to the server, so as to obtain the coupon information corresponding to the coupon notification, which specifically includes information such as the coupon name, the coupon address, the coupon type, the coupon picture, the coupon amount, the coupon time, the coupon payment mode, the coupon participation brief description, etc., so as to facilitate the user to check whether the coupon accords with the psychological requirement of the user, whether to participate in the coupon, how to participate in the coupon, etc.
More preferably, the embodiment further obtains a working day and a rest day of the user according to the configuration information of the user, further divides the data set formed by the obtained daily consumption details of the user into a working day data set and a rest day data set, and trains the constructed neural network model according to the working day data set and the rest day data set to obtain a working day consumption merchant feature recognition model and a rest day consumption merchant feature recognition model of the target user. In particular, it is normal that for some users, especially for some office workers, the consumption of the workday is mostly based on the consumption of daily work meals, but on the contrary, the rest day may be more preferable to other consumption such as diner, family related consumption, etc., such as for some entertainment consumption, large consumption, etc., more is possible to be completed on the rest day, that is, the consumption habits on the work day and the rest day are different, so the present embodiment also classifies the consumption of the user into the work day and the rest day, so as to recommend more accurate consumption for the user.
Further, step S2 further includes, while obtaining the address information of the target user currently, obtaining a current system date, and selecting a corresponding consumer merchant feature recognition model according to the current system date to obtain the preferential merchant interested by the target user. Specifically, if the current system date is the weekday, each preferential merchant is matched with the feature recognition model of the target user's weekday consuming merchant, and if the current system date is the weekday, each preferential merchant is matched with the feature recognition model of the target user's weekday consuming merchant.
In addition, in order not to disturb the work and rest of the user, the embodiment is further realized by setting timing time, for example, setting 9 to 10 am every day, and other time can be set to be in a no-disturb mode when the address information and the system date of the user are acquired. In addition, for more accurate pushing, in this embodiment, when the address information of the user is acquired, the address information of the user needs to be timed, that is, whether the address information of the user changes greatly in a period of time is determined, if so, the user is in a moving state, and preferential pushing is not required, so that frequent pushing is avoided, system confusion is caused, and user experience is affected.
More preferably, after pushing the preferential notice to the user, if detecting that the mobile phone bank APP is opened, the latest pushed preferential notice is displayed on the main page of the mobile phone bank APP in a popup window mode. The user can be guided to view the preferential notice in a popup window mode.
Likewise, when the coupon notice is pushed, the push time of the coupon notice is started to be timed. When the timing time exceeds the preset time length or the preferential notice is not read, the corresponding preferential notice is also generated into a short message notice according to the short message notice template, and then the corresponding short message notice is pushed to the short message end of the mobile terminal of the target user. In the actual use process, some users may close the notification authority of the mobile banking APP, which may cause the target user to miss the preferential notification. Therefore, when the preferential notice is pushed and is not read or checked in a certain time, the preferential notice is prevented from being missed by the user, and the preferential notice is required to be generated into a short message notice according to a short message notice template and pushed to a short message end of a mobile terminal of a target user so as to remind the user to check the preferential notice. That is, the user can be reminded again to check the preferential notice in a short message mode, so that the preferential missing is avoided.
More preferably, when the preferential expiration of the preferential notice or the pushing time of the preferential notice exceeds a preset duration, the corresponding preferential notice is deleted from the mobile phone bank APP end. The regular deletion of the preferential notice can ensure the normal operation of the mobile banking APP, and avoid excessive cache occupation and influence on the use of users. When deleting the preferential notice, whether to delete can be judged according to whether the preferential notice preferential is expired, the pushing duration and the like.
When the offers corresponding to the offer notification are used, the time of using the offers by the target user, the merchant of the offers, the merchant address of the offers, the offer payment mode, the payment cost details and the like are recorded, and an offer use detail record of the target user is generated and stored in the system, and meanwhile keywords are set for the offer use detail record of the target user. In step S3, when a plurality of preferential merchants interested by the target user are obtained, the priority of each preferential merchant is calculated according to a priority calculation rule, and then the preferential information of the preferential merchants is pushed to the mobile banking APP of the mobile terminal according to the order of the priorities. When calculating the priority of the preferential merchant, the priority of the preferential merchant is realized in a weighted average mode, specifically, a plurality of key factors are set, each key factor has a preset weight ratio, and the weighted average sum of the plurality of key factors is calculated to calculate the priority of the preferential merchant. Meanwhile, the engineer also builds the score of each key factor in advance, for example, the preferential amount sequentially sets a plurality of scores in order of preferential size, and the like, and the scores are set according to actual experience values, which is not specifically described in the embodiment.
Preferably, the embodiment also supports the appointed pushing of the user, that is, the target user actively opens the mobile banking APP and searches the relevant preferential page, and the target user can input the data information related to the preferential page, including address, merchant type, time, bank type, consumption maximum rating, and the like, by pushing the page of the search keyword to the target user. After the relevant information input by the user is obtained, the relevant preferential information can be searched according to the information, meanwhile, the searched preferential commercial tenant is matched with a commercial tenant characteristic recognition model constructed in the system, a plurality of preferential commercial tenants are screened out, and the preferential commercial tenants are sequentially displayed to the user on a search page in a list mode, so that personalized recommendation of the user is realized.
The above embodiments are only preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, but any insubstantial changes and substitutions made by those skilled in the art on the basis of the present invention are intended to be within the scope of the present invention as claimed.

Claims (9)

2. The method for pushing bank information based on big data model according to claim 1, wherein the model construction step further comprises constructing a data set according to the collected daily consumption merchant information of the target user, dividing the data set into a training set and a verification set, training a pre-constructed neural network model according to the training set, verifying the trained model according to the verification set, and obtaining a consumption merchant feature recognition model of the target user according to the verified model after verification is passed, wherein the consumption merchant feature recognition model of the target user comprises an association relationship between the target user and the consumption merchant features of the target user.
CN202411796176.2A2024-12-092024-12-09 Bank information push method based on big data modelPendingCN119722231A (en)

Priority Applications (1)

Application NumberPriority DateFiling DateTitle
CN202411796176.2ACN119722231A (en)2024-12-092024-12-09 Bank information push method based on big data model

Applications Claiming Priority (1)

Application NumberPriority DateFiling DateTitle
CN202411796176.2ACN119722231A (en)2024-12-092024-12-09 Bank information push method based on big data model

Publications (1)

Publication NumberPublication Date
CN119722231Atrue CN119722231A (en)2025-03-28

Family

ID=95096263

Family Applications (1)

Application NumberTitlePriority DateFiling Date
CN202411796176.2APendingCN119722231A (en)2024-12-092024-12-09 Bank information push method based on big data model

Country Status (1)

CountryLink
CN (1)CN119722231A (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN106572129A (en)*2015-10-092017-04-19美的集团股份有限公司Bank-card preferential information pushing method and apparatus thereof
CN110490652A (en)*2019-08-162019-11-22阿里巴巴集团控股有限公司A kind of information-pushing method and system
KR20210123946A (en)*2020-04-062021-10-14주식회사 에스원Advertisement System by using Coupon Savinf System Based on Mobile Terminal and Method thereof
CN114117236A (en)*2021-12-072022-03-01广州道然信息科技有限公司User interaction method, device, equipment and storage medium based on intelligent terminal
CN115271041A (en)*2022-07-252022-11-01国家电网有限公司客户服务中心 A method for predicting the traffic volume of electric service

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN106572129A (en)*2015-10-092017-04-19美的集团股份有限公司Bank-card preferential information pushing method and apparatus thereof
CN110490652A (en)*2019-08-162019-11-22阿里巴巴集团控股有限公司A kind of information-pushing method and system
KR20210123946A (en)*2020-04-062021-10-14주식회사 에스원Advertisement System by using Coupon Savinf System Based on Mobile Terminal and Method thereof
CN114117236A (en)*2021-12-072022-03-01广州道然信息科技有限公司User interaction method, device, equipment and storage medium based on intelligent terminal
CN115271041A (en)*2022-07-252022-11-01国家电网有限公司客户服务中心 A method for predicting the traffic volume of electric service

Similar Documents

PublicationPublication DateTitle
CN106651424B (en) Establishment and analysis method of power user portrait based on big data technology
US12354103B2 (en)System, method, and computer program product to determine cuisine type classifiers
US9436965B2 (en)Interactive map for grouped activities within a financial and social management system
DemunterTourism statistics: Early adopters of big data
US20150051948A1 (en)Behavioral attribute analysis method and device
US8255392B2 (en)Real time data collection system and method
US20250199659A1 (en)Dynamic Sorting and Inference Using Gesture Based Machine Learning
US20160350868A1 (en)Interactive map for grouped activities within a financial and social management system
CN109840851A (en)A kind of integrated management approach, device and the equipment of data of managing money matters
US10002395B2 (en)Interactive mapping system for user experience augmentation
JP2008158823A (en) Information provision system
US9323852B2 (en)Activity list filters for a financial and social management system
US9786018B2 (en)Activity list enhanced with images for a financial and social management system
CN111028087A (en)Information display method, device and equipment
JPWO2004107232A1 (en) CONTACT POINT NAVIGATION SYSTEM AND METHOD, RECORDING MEDIUM CONTAINING THE METHOD, AND TRANSMISSION MEDIUM TRANSMITTING THE METHOD
US20250238876A1 (en)Matchmaking attendees for virtual, in-person, and hybrid events
OsaysaImproving the quality of marketing analytics systems
US20120185471A1 (en)Business intelligence and customer relationship management tools and techniques
CN110209944A (en)A kind of stock analysis teacher recommended method, device, computer equipment and storage medium
KR20230079704A (en)Gift recommendation system and method chosen by recipients
US9786019B2 (en)Grouped packages for a financial and social management system
WO2022026880A1 (en)Systems and methods for asset analysis
US9542497B2 (en)Information processing apparatus, information processing method, and information processing program
LangerThe importance of probability-based sampling methods for drawing valid inferences
CN119722231A (en) Bank information push method based on big data model

Legal Events

DateCodeTitleDescription
PB01Publication
PB01Publication
SE01Entry into force of request for substantive examination
SE01Entry into force of request for substantive examination

[8]ページ先頭

©2009-2025 Movatter.jp