Data pushing method and related device

Abstract

The invention provides a data pushing method and a related device, and relates to the field of data pushing. According to the information pushing method and device, when the information is pushed, the fund flow direction condition of the user is considered, the fund flow direction condition can reflect the concern point of the user, pushing is carried out based on the fund flow direction, and the matching degree of the pushed information and the user can be higher. Furthermore, statistical analysis operation is performed on the fund flow direction data to obtain a characteristic value of each preset flow direction characteristic, an index value of a data pushing index is calculated based on the characteristic value of each preset flow direction characteristic, and a dimension value of a region pushing dimension corresponding to the index value of the data pushing index is determined. In other words, when the data is pushed, the situation of the region where the user is located is considered, and then adaptive pushing is carried out for the region, so that the pushed information is matched with the region, the matching degree of the pushed information and the user in the region is improved, and then the accuracy of the pushed information is improved.

Description

Technical Field

[0001] The present invention relates to the field of data push, and more specifically, to a data push method and related devices. Background Art

[0002] In order to increase the stickiness of users to the bank, information will be pushed to users regularly or irregularly. The pushed information can include user operation information, preferential activity information, consultation telephone information, etc.

[0003] Currently, bank employees determine the information to be pushed based on their experience, and the accuracy of the pushed information is low. Summary of the invention

[0004] In view of this, the present invention provides a data push method and related devices to solve the problem of low accuracy of pushed information.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] A data push method, comprising:

[0007] Obtaining fund flow data of a target account; each data field in the fund flow data is separated by a preset delimiter, and each data field is arranged in a preset order;

[0008] Performing statistical analysis on the capital flow data to obtain characteristic values ​​of each preset flow characteristic;

[0009] Calculating the index value of the data push index based on the characteristic values ​​of each preset flow direction characteristic;

[0010] Determine a dimension value of a regional push dimension corresponding to the indicator value of the data push indicator;

[0011] The regional data push operation is performed using the feature values ​​of the preset flow direction features and the dimension value of the regional push dimension.

[0012] Optionally, statistical analysis operations are performed on the capital flow data to obtain characteristic values ​​of various preset flow characteristics, including:

[0013] Analyze the total amount of capital outflow in the capital flow data to obtain a characteristic value of the total amount of outflow characteristics;

[0014] Performing a fund amount calculation operation on the fund flow data according to the bank flow dimension to obtain a characteristic value of the bank flow characteristic;

[0015] Performing a capital amount calculation operation on the capital flow data according to the regional dimension to obtain a characteristic value of the regional outflow characteristic;

[0016] Performing a capital amount calculation operation on the capital flow data according to the time dimension to obtain a characteristic value of a preset period outflow characteristic;

[0017] The preset flow direction characteristics include the total outflow amount characteristics, the bank flow characteristics, the regional outflow characteristics and the preset period outflow characteristics.

[0018] Optionally, calculating an index value of a data push index based on the characteristic values ​​of each preset flow direction characteristic includes:

[0019] Calculate the index value of the regional outflow proportion index by using the characteristic value of the total outflow characteristic and the characteristic value of the regional outflow characteristic;

[0020] Calculate the index value of the regional regression rate index based on the characteristic value of the total outflow characteristic and the characteristic value of the bank flow characteristic;

[0021] Calculate the index value of the regional change degree index corresponding to the characteristic value of the outflow characteristic of the preset period;

[0022] The data push indicators include the regional outflow proportion indicator, the regional return rate indicator and the regional change degree indicator.

[0023] Optionally, calculating the index value of the regional change degree index corresponding to the characteristic value of the preset period outflow characteristic includes:

[0024] Sort the characteristic values ​​of the preset period outflow characteristics of each target account to obtain an outflow volume sequence;

[0025] Calculating the difference between adjacent data in the outflow sequence to obtain a difference sequence;

[0026] Performing a stability analysis operation on the difference sequence to obtain a fluctuation subsequence in the difference sequence;

[0027] Calculating the variance corresponding to the volatility subsequence to obtain a volatility variance value sequence for different regions;

[0028] According to the volatility variance value series of different regions, the index values ​​of regional change degree index of different regions are calculated.

[0029] Optionally, according to the volatility variance value sequences of different regions, the index values ​​of the regional change degree index of different regions are calculated, including:

[0030] For the volatility variance value sequences of different regions, a K-means clustering operation is performed on the volatility variance value sequences to obtain multiple clusters;

[0031] Calculate the mean fluctuation degree of each cluster and the number of sample points in the cluster;

[0032] The cluster corresponding to the largest number of sample points in the cluster is taken as the final fluctuation degree cluster, and the fluctuation degree mean value of the final fluctuation degree cluster is taken as the index value of the regional change degree index of the corresponding region.

[0033] Optionally, determining a dimension value of a regional push dimension corresponding to the indicator value of the data push indicator includes:

[0034] The index value of the regional outflow proportion index, the index value of the regional regression rate index and the index value of the regional change degree index are weightedly summed to obtain the dimension value of the regional push dimension.

[0035] Optionally, using the characteristic values ​​of the preset flow direction characteristics and the dimension value of the regional push dimension to push regional data includes:

[0036] Based on the characteristic value of the bank flow characteristic, determine first push data of a target bank whose fund outflow amount is greater than a first preset threshold, and generate regional push data based on the first push data; or, based on the characteristic value of the regional outflow characteristic, determine second push data of a target region whose fund outflow amount is less than a second preset threshold, and generate regional push data based on the second push data;

[0037] According to the dimension value of the regional push dimension, a push operation is performed on the regional push data.

[0038] Optionally, according to the dimension value of the regional push dimension, the push operation on the regional push data includes:

[0039] Using the dimension value of the regional push dimension, calling the classifier to allocate the total data resources, and obtaining the data resource allocation amount of each region;

[0040] According to the data resource allocation amount, the push data is pushed to the corresponding area.

[0041] A data pushing device, comprising:

[0042] A data acquisition module is used to acquire the funds flow data of the target account; each data field in the funds flow data is separated by a preset delimiter, and each data field is arranged in a preset order;

[0043] A statistical analysis module, used to perform statistical analysis on the fund flow data to obtain characteristic values ​​of each preset flow characteristic;

[0044] An index value calculation module, used to calculate the index value of the data push index based on the characteristic values ​​of each preset flow direction characteristic;

[0045] A dimension value determination module, used to determine the dimension value of the regional push dimension corresponding to the indicator value of the data push indicator;

[0046] The data push module is used to use the characteristic values ​​of the preset flow characteristics and the dimension value of the regional push dimension to push regional data.

[0047] An electronic device comprises at least one processor and a memory connected to the processor, wherein:

[0048] The memory is used to store computer programs;

[0049] The processor is used to execute the computer program so that the electronic device can implement the above-mentioned data pushing method.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] The present invention provides a data push method and related devices. In the present invention, when pushing information, the user's fund flow situation will be considered. The fund flow situation can reflect the user's focus, and then push based on the fund flow, which can make the pushed information more compatible with the user. Further, the present invention performs statistical analysis operations on the fund flow data to obtain the characteristic values ​​of each preset flow feature, calculates the index value of the data push index based on the characteristic values ​​of each preset flow feature, and determines the dimension value of the regional push dimension corresponding to the index value of the data push index. That is, when pushing data, the present invention will consider the situation of the user's region, and then push appropriately for the region, so that the pushed information matches the region, improves the matching degree between the pushed information and the users in the region, and then improves the accuracy of the pushed information. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0053] Figure 1 A flow chart of a data push method provided by an embodiment of the present invention;

[0054] Figure 2 A flow chart of a feature calculation method provided by an embodiment of the present invention;

[0055] Figure 3A flowchart of an indicator calculation method provided by an embodiment of the present invention;

[0056] Figure 4 A flowchart of another indicator calculation method provided by an embodiment of the present invention;

[0057] Figure 5 A structural diagram of a data push system provided by an embodiment of the present invention;

[0058] Figure 6 A schematic diagram of the structure of a data push device provided by an embodiment of the present invention;

[0059] Figure 7 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0062] In order to increase the stickiness of users to the bank, information will be pushed to users regularly or irregularly. The pushed information can include user operation information, preferential activity information, consultation telephone information, etc.

[0063] At present, when determining the information to be pushed, bank employees make the decision based on their experience, and the accuracy of the information pushed is low. Taking the push of promotional information as an example, major banks across the country are competing fiercely in marketing corporate deposits. Bank marketing personnel obtaining the data on the flow of funds in the bank's corporate accounts helps branch staff conduct targeted marketing, thereby avoiding capital outflow. At present, when obtaining the flow of funds in corporate accounts within the jurisdiction of the bank, bank marketing managers often adopt the traditional model of branch marketing personnel reporting recent fund flow data and marketing managers manually counting fund flow characteristics. This will cause data lag, and at the same time, it is difficult to count the full amount of data, and the data display is not intuitive. Marketing managers cannot obtain the dynamic flow of funds in corporate deposit accounts opened by the bank intuitively and timely, so they cannot formulate appropriate marketing strategies, resulting in capital outflow and a decline in deposits, which reduces the market competitiveness of the bank.

[0064] To this end, in order to improve the accuracy of information push, the present invention will consider the user's fund flow situation when pushing information. The fund flow situation can reflect the user's focus, and then push based on the fund flow, which can make the pushed information more compatible with the user. Further, the present invention performs statistical analysis operations on the fund flow data to obtain the characteristic values ​​of each preset flow feature, calculates the index value of the data push index based on the characteristic values ​​of each preset flow feature, and determines the dimension value of the regional push dimension corresponding to the index value of the data push index. That is, when pushing data, the present invention will consider the situation of the user's region, and then push appropriately for the region, so that the pushed information matches the region, improves the matching degree between the pushed information and the users in the region, and then improves the accuracy of the pushed information.

[0065] Based on the above content, an embodiment of the present invention provides a data push method, referring to Figure 1 , which may include:

[0066] S11. Obtain the fund flow data of the target account.

[0067] In actual applications, the flow of funds can reflect the user's focus. Therefore, analyzing the flow of funds data can accurately understand the user's demand information, and then perform corresponding push operations based on the demand information. For example, after analyzing the user's fund flow, it is found that the user has invested more in the tourism field. Therefore, when pushing information, priority is given to pushing tourism-related information, such as tourism marketing information, tourism emergency contact numbers, tourism ranking data, etc.

[0068] In actual applications, since more attention is paid to the fund flow of public accounts (such as financial institutions, etc.), in this embodiment, the public account can be set as the target account to obtain the fund flow data of the target account.

[0069] In addition, for the purpose of comprehensive monitoring, both public accounts and personal accounts can be used as target accounts, and the fund flow data of public accounts and personal accounts can be obtained at the same time.

[0070] Generally, the big data platform contains bank public account opening information, capital data and other information. If all of them are extracted, the amount of data will be large, affecting the system operation efficiency. When extracting data, this embodiment only extracts information such as the public account account number, opening bank, account opening date, account opening region, capital outflow bank, outflow time, outflow amount, etc., and uses the extracted data as capital flow data.

[0071] In order to achieve standardization of data collection, standardized processing operations will be performed on the fund flow data, such as using preset delimiters to separate the various data fields in the fund flow data, and arranging the various data fields in a preset order.

[0072] To expand on this, the collected capital flow data can be concatenated into a string, with one flow data per line, and the data fields in each line are separated by |! and arranged in the agreed order.

[0073] The splicing example can be as follows: Account number of public account|! Bank of account opening|! Account opening date|! Account opening region|! Bank of fund outflow|! Outflow time|! Outflow amount|! .

[0074] The data extraction system saves the concatenated character string in a text document txt format according to the agreed naming convention, and automatically saves the target file in the specified directory of the system database server.

[0075] In actual applications, in the database, the fiscal funds flow data record table can be used to record the above-extracted funds flow data to provide basic information for subsequent data analysis. It should be noted that the field order of the fiscal funds flow data record table should be consistent with the field setting order in the above-mentioned target file to avoid data dislocation.

[0076] In this embodiment, the user can set the time according to needs. For example, at 22:30 every day, the system automatically checks the target file in the specified directory. If the target file exists, the file is imported into the financial funds flow data record table. Otherwise, wait for the next scheduled time to check the specified directory again.

[0077] In another implementation of the present invention, in addition to creating a financial funds flow data record table, data tables such as a user information table, a bank branch information table, and an authority allocation table may also be created to achieve corresponding functions.

[0078] Among them, a user information table is created in the system database to save bank personnel information (such as marketing personnel information), including employee number, affiliated organization and password, etc. The table information is used for background verification when the user logs in.

[0079] The bank branch information table is used to record all bank branches under the jurisdiction of the bank, including branch numbers and other information. This table is used to manage user permissions, aiming to assign minimum permissions to users and prevent financial risks such as data leakage.

[0080] The permission allocation table includes the permissions of different users. Assuming that the user is a provincial marketing manager, he has the highest permission and can see the capital outflow data of the entire province. The permission of marketing personnel at the second-level branch and below is set to only view the data of the outlets within his jurisdiction. For example, marketing personnel of the XX city branch can only see the capital outflow data of all outlets in XX city.

[0081] For the convenience of users, this embodiment sets a time interval selection, and users can select the fund outflow time period they want to know. Subsequently, the user-set time period and the data allowed to be viewed by the user authority are extracted from the financial fund flow data record table, and the extracted data is analyzed.

[0082] S12. Perform statistical analysis on the capital flow data to obtain characteristic values ​​of various preset flow characteristics.

[0083] In this embodiment, the preset flow direction characteristics are determined based on experience and are characteristics that are used more frequently when pushing information, such as total outflow characteristics, bank flow characteristics, regional outflow characteristics, and the preset period outflow characteristics.

[0084] In another implementation of the present invention, referring to Figure 2 , step S12 may include:

[0085] S21. Analyze the total amount of capital outflow in the capital flow data to obtain a characteristic value of the total amount of outflow characteristics.

[0086] Among them, the total outflow feature refers to the total amount of bank outflows. The total amount of bank outflows refers to the sum of the amount of funds transferred out by all branches in the region within the time period selected by the user. This feature allows users to have an overall impression of the bank's fund outflows over a period of time, and to compare the total amount of fund outflows in the time dimension, determine the peak and valley of fund outflows each year, and then push information in a targeted manner, such as increasing marketing efforts. It should be noted here that the basis for determining the branch to which the funds flow out is the bank where the public account is opened.

[0087] In this embodiment, the system analyzes the data fields in the financial funds flow data record table and obtains the total amount of bank outflow funds by statistics.

[0088] S22. Perform a fund amount calculation operation on the fund flow data according to the bank flow dimension to obtain a characteristic value of the bank flow characteristic.

[0089] In this embodiment, the bank flow characteristics refer to the characteristics of the amount of funds flowing to the bank.

[0090] When the bank transfers funds out, the system will record the outflow bank. The outflow bank refers to the bank where the funds are transferred to, and different banks are marked differently. The sum of the amount of funds flowing to each bank during the selected time period is counted as the total amount of funds flowing to the bank. This total amount of funds is the characteristic value of the feature of the amount of funds flowing to the bank.

[0091] S23. Perform a capital amount calculation operation on the capital flow data according to the regional dimension to obtain a characteristic value of the regional outflow characteristic.

[0092] In this embodiment, the regional outflow feature refers to the regional outflow amount feature. In this embodiment, the region can be when the object of analysis is the whole province, the region can be such as a prefecture-level city, and when the object of analysis is the whole city, the region can be such as various urban areas, counties, etc. Taking the analysis object as the whole province as an example, the prefecture-level city is used as the standard for different regions, and the user sets the scope of the prefecture-level city according to the needs, and then counts the outflow amount of funds in different prefecture-level cities in the province. Generally speaking, banks will set different bank outlets according to the administrative regions of prefecture-level cities, and each outlet has an outlet id (identity document, account number), that is, an institution number. The outlet institution number is strictly assigned according to the upper and lower hierarchical relationship between the outlets, and the first two digits of the institution number in each region are different. Therefore, when counting the outflow of funds in different regions, according to the rule of the outlet institution number, all the account opening outlets in the region can be obtained, and the total outflow funds of all the account opening outlets in the region are calculated according to the outflow funds corresponding to the field of the account opening bank. The total outflow funds of each prefecture-level city in the province are calculated as the outflow amount of funds in different regions.

[0093] S24. Perform a capital amount calculation operation on the capital flow data according to the time dimension to obtain a characteristic value of a preset period outflow characteristic.

[0094] Specifically, the time dimension in this embodiment may be a dimension such as month, day, etc. Generally, the frequency of using month as the time dimension is higher. At this time, the preset period outflow feature refers to the monthly outflow capital amount feature.

[0095] The monthly outflow of funds refers to the sum of the outflow of funds from all outlets in the province each month. The total monthly outflow of funds can be calculated based on the outflow time field in the fiscal funds flow data record table.

[0096] In addition, the corresponding monthly total fund outflow can be counted according to the target account, such as the monthly total fund outflow of each target account in each region.

[0097] At this time, the characteristic value of the monthly outflow capital amount characteristic may include the monthly total capital outflow of all outlets in the province each month, and the monthly total capital outflow of each target account in each region.

[0098] After obtaining the above-mentioned total outflow characteristics, bank flow characteristics, regional outflow characteristics and the preset period outflow characteristics, a front-end display page can be designed, which includes four modules: the total amount of bank outflows in the province, a bar chart of the amount of funds flowing to banks, a regional outflow map and a monthly outflow line chart.

[0099] When designing the page, it is divided into four parts, with the total amount of bank outflows in the province in units of 10,000 yuan, and a box diagram is used to display it. For the characteristics of the amount of funds flowing to the bank, the bank is classified according to several types of banks, such as the four major banks, commercial banks and other banks, and arranged in ascending order according to the amount of funds flowing out. The amount of funds flowing to the bank is displayed in the form of a bar chart, which can intuitively see the amount of funds flowing to major banks in different regions, which is convenient for marketing personnel to refer to. According to the amount of funds flowing out of different cities in the province calculated above, the provincial plug-in map developed by the bank is called up, and the amount of funds flowing out of different cities is compared with the cities in the provincial map and displayed on the map. The floating window bubble can display the amount of funds flowing out of the city and the corresponding marketing guidance indicator value when the user puts the mouse arrow on the city. According to the statistical monthly outflow of funds, a monthly line chart is drawn on the front-end page. The horizontal axis of the chart is the year and month, and the vertical axis is the amount of funds flowing out. According to the chart, users can intuitively see the changing trend of the amount of funds flowing out in the time dimension.

[0100] S13. Calculate the index value of the data push index based on the characteristic values ​​of each preset flow direction characteristic.

[0101] In this embodiment, the data push index refers to the push index of each region. When information is pushed, the pushed information content may only apply to a certain region and not to the whole country. If a certain information is pushed to the whole country, the push effect may not be good. Different regions may need different push information. For example, the southern region is more inclined to use credit cards, so information related to credit cards can be pushed, while the northern region is more inclined to use bank cards, so information related to bank cards can be pushed. The push indexes of different regions can be counted to perform corresponding information push operations by region.

[0102] It should be noted that the above-mentioned "the southern region is more inclined to use credit cards, while the northern region is more inclined to use bank cards" is only for illustrative purposes to illustrate the regional push situation, which may deviate from the actual situation. The specific situation shall prevail.

[0103] In practical applications, data push indicators can be as follows:

[0104] The outflow proportion index PF of the region i 、The regression rate index R of the region i and indicators of the degree of change in the region

[0105] S14. Determine the dimension value of the regional push dimension corresponding to the indicator value of the data push indicator.

[0106] In this embodiment, after obtaining PF i , R i as well as Afterwards, according to the PF of the i-th region i , R i , The three characteristic indicators can obtain the dimension value of the regional push dimension of the city. The regional push dimension in this embodiment can be a comprehensive information push indicator value. If marketing is carried out in the region, the comprehensive information push indicator value can be a comprehensive marketing indicator value. If warning is carried out in the region, the comprehensive information push indicator value can be a comprehensive warning indicator value.

[0107] In order to avoid the influence of scale range between different feature indicators, the detected indicators Normalize and obtain the normalized feature index: Calculate the comprehensive information push index value ρ of the city i The formula is:

[0108]

[0109] Among them, ω1, ω2, and ω3 are the weight values ​​of the characteristic indicators respectively.

[0110] As an example, in the embodiment of the present invention, ω1=0.2, ω2=0.3, and ω3=0.5.

[0111] As can be seen from the above, in this embodiment, the index value of the regional outflow proportion index, the index value of the regional regression rate index and the index value of the regional change degree index are weighted and summed to obtain the dimension value of the regional push dimension.

[0112] Taking the marketing scenario as an example, when ρ i The larger it is, the greater the guiding indicator value of the city when marketers formulate marketing strategies.

[0113] S15. Generate regional push data using at least one of the feature values ​​of the preset flow direction features and the dimension value of the regional push dimension.

[0114] In this embodiment, after the characteristic values ​​of each preset flow direction characteristic and the dimension value of the regional push dimension are calculated, these data can be used to generate regional push data.

[0115] For example, the capital outflow amount or regional outflow characteristics may be used to generate regional push data, and then the dimension value of the regional push dimension may be used to perform resource allocation operations, and the allocated resources may be used to perform data push operations.

[0116] In this embodiment, when pushing information, the user's fund flow situation will be considered. The fund flow situation can reflect the user's focus, and then push based on the fund flow, which can make the pushed information more compatible with the user. Further, the present invention performs statistical analysis operations on the fund flow data to obtain the characteristic values ​​of each preset flow feature, calculates the index value of the data push index based on the characteristic values ​​of each preset flow feature, and determines the dimension value of the regional push dimension corresponding to the index value of the data push index. That is, when pushing data, the present invention will consider the situation of the user's region, and then push appropriately for the region, so that the pushed information matches the region, improves the matching degree between the pushed information and the users in the region, and then improves the accuracy of the pushed information.

[0117] The above embodiment mentions that the indicator value of the data push indicator needs to be calculated. This embodiment introduces the specific implementation process. Figure 3 , based on the characteristic values ​​of the preset flow direction characteristics, calculating the index value of the data push index may include:

[0118] S31. Calculate the index value of the regional outflow proportion index by using the characteristic value of the total outflow characteristic and the characteristic value of the regional outflow characteristic.

[0119] Specifically, the regional outflow ratio index of the i-th region is PF i It means that the regional outflow proportion index refers to the proportion of funds outflowing from cities.

[0120] According to the regional outflow characteristics and total outflow characteristics of different regions obtained in the above steps, after dimensionless processing, the index value of the regional outflow proportion index of each prefecture-level city is obtained, and the formula is as follows:

[0121]

[0122] Among them, PF i is the outflow ratio index of the ith city, Fi is the characteristic value of the regional outflow characteristic of the i-th prefecture-level city, F S is the characteristic value of the total outflow feature.

[0123] S32. Calculate the index value of the regional regression rate index based on the characteristic value of the total outflow characteristic and the characteristic value of the bank flow characteristic.

[0124] Specifically, the regional regression rate index can be expressed as R i It specifically refers to the regional capital return rate.

[0125] According to the bank flow characteristics in the database, the amount of funds flowing to different banks in each city can be obtained, including the amount of funds flowing to this bank. According to the regional outflow characteristics, the total amount of funds flowing out of this bank in the province can be obtained, and then the capital return rate of each city can be obtained. The formula is as follows:

[0126]

[0127] Among them, R i is the regional capital return rate of the ith city, FN i is the amount of funds flowing to the bank from the ith city, FN S It is the total amount of funds flowing out of the bank in the province.

[0128] S33, calculating the index value of the regional change degree index corresponding to the characteristic value of the outflow characteristic of the preset period.

[0129] Regional change index It refers to the degree of change in funds in the region, and the degree of change generally refers to the degree of volatility.

[0130] According to prior knowledge, the outflow of funds from each target account due to monthly salary payments, normal fund transactions, etc. should be roughly at the same level, while the amount of funds flowing out due to marketing by the bank itself or other banks should increase or decrease abnormally, and the degree of fund change can be obtained accordingly.

[0131] In practical applications, refer to Figure 4 , step S33 may include:

[0132] S41. Sort the characteristic values ​​of the preset period outflow characteristics of each target account to obtain an outflow volume sequence.

[0133] Specifically, according to the above discussion, the characteristic value of the monthly outflow capital amount characteristic may include the monthly total capital outflow of all outlets in the province each month, and the monthly total capital outflow of each target account in each region.

[0134] Therefore, in this embodiment, for the i-th region, a fluctuation analysis may be performed on the total monthly fund outflow of each target account.

[0135] First, according to the time period selected by the user, the monthly total capital outflow and outflow date of all target accounts in each city are obtained, and the outflow sequence is obtained according to the monthly total capital outflow in the time period. The outflow sequence can be a monthly outflow capital sequence. For example, the monthly outflow capital sequence of a certain institution in a certain city in a year is X = [FM1, FM2, ..., FM 12 ].

[0136] S42: Calculate the difference between adjacent data in the outflow sequence to obtain a difference sequence.

[0137] Specifically, the difference between adjacent funds in the outflow sequence is calculated to obtain the difference sequence of outflow funds X1 = [ΔFM1, ΔFM2, ..., ΔFM k ], the calculation formula is:

[0138] ΔFM k =|FM k+1 -FM k ]

[0139] Where ΔFD k It is the total monthly capital outflow of the k+1th month and the total monthly capital outflow of the kth month.

[0140] S43, performing a stability analysis operation on the difference sequence to obtain a fluctuation subsequence in the difference sequence.

[0141] Specifically, a stability analysis is performed on the difference sequence in order to screen out the volatile subsequences with large fluctuations and the stable subsequences with stable fluctuations.

[0142] In practical applications, the difference sequence can be sent to the TCN (Temporal Convolutional Network) network to obtain the stable capital flow subsequence X2 and the fluctuating capital flow subsequence X3. The TCN samples can be manually labeled.

[0143] Among them, the above-mentioned capital flow fluctuation subsequence X3 is the fluctuation subsequence in this embodiment.

[0144] For example, the difference sequence includes the values ​​of a, b, c, d, c, f, and g. The stable subsequence of capital flow is the sequence composed of a, b, c, and d, and the fluctuating subsequence of capital flow is the sequence composed of c, f, and g.

[0145] S44. Calculate the variance corresponding to the volatility subsequence to obtain a volatility variance value sequence for different regions.

[0146] Specifically, the variance σ of all values ​​in the capital flow volatility subsequence X3 is calculated. The larger the variance value σ is, the greater the volatility of the capital flow is.

[0147] For the i-th region, the variance σ of all target accounts located in the region is aggregated to obtain the volatility variance value sequence [σ1, σ2, ..., σ3] of the i-th region.

[0148] S45. According to the volatility variance value sequence of different regions, the index values ​​of the regional change degree index of different regions are calculated.

[0149] After obtaining the volatility variance value series of different regions, for each region, K-means clustering can be used to obtain the index value of the regional change degree index of the corresponding region.

[0150] Specifically, step S45 may include:

[0151] 1) For the volatility variance value sequences of different regions, a K-means clustering operation is performed on the volatility variance value sequences to obtain multiple clusters.

[0152] Specifically, according to the volatility variance value sequence, K-means clustering is performed on the values ​​in the sequence, and multiple clusters are finally obtained. The σ in each cluster is similar.

[0153] 2) Calculate the mean fluctuation degree of each cluster and the number of sample points in the cluster.

[0154] Specifically, for each cluster, the average value of all σ in the cluster is calculated to obtain the mean fluctuation degree:

[0155] For the number of sample points in the cluster, we can directly count the total number of sample points in each cluster to get the number of sample points in the cluster N m The maximum number of sample points in a cluster is called N max .

[0156] 3) The cluster corresponding to the largest number of sample points in the cluster is used as the final fluctuation degree cluster, and the fluctuation degree mean of the final fluctuation degree cluster is used as the index value of the regional change degree index of the corresponding region.

[0157] In practical applications, if the value of the number of sample points in a cluster is large, it means that a large number of σ in the region are concentrated in the cluster, which means that the fluctuations in the region are mainly concentrated in the cluster. Therefore, based on this principle, the maximum number of sample points in the cluster N can be max The corresponding cluster is taken as the final volatility cluster, and the volatility mean of the final volatility cluster is taken as Index value as an indicator of the degree of regional change in the corresponding region As an example, the total number of clusters of the plurality of clusters in the embodiment of the present invention may be 5.

[0158] In this embodiment, when analyzing the regional change degree index, the analysis is performed from three perspectives: the regional outflow proportion index, the regional regression rate index, and the regional change degree index, fully considering the factors affecting the regional information recommendation and improving the reliability and accuracy of subsequent information recommendations.

[0159] In another implementation of the present invention, a specific implementation of "using the characteristic values ​​of each preset flow direction characteristic and the dimension value of the regional push dimension to perform a push operation of regional data" is provided, which may include the following steps:

[0160] 1) Based on the characteristic value of the bank flow characteristic, determine the first push data of the target bank whose fund outflow amount is greater than the first preset threshold, and generate the regional push data based on the first push data; or, based on the characteristic value of the regional outflow characteristic, determine the second push data of the target region whose fund outflow amount is less than the second preset threshold, and generate the regional push data based on the second push data.

[0161] In this embodiment, taking the marketing information push as an example, taking a certain region as an example, after the provincial marketing management personnel log in to the capital flow data analysis system, they can accurately formulate marketing strategies according to the graphics of several modules in the page. In this embodiment, examples are given, and the strategy formulation method includes but is not limited to the following examples. According to the bar chart of the amount of funds flowing to banks in different regions, provincial users can obtain target banks with high capital outflow (capital outflow greater than the first preset threshold), view their recently launched corporate products and preferential activities, that is, the first push data, and generate regional push data based on the first push data.

[0162] After obtaining the promotional activities of the target bank, the backend can adaptively adjust the promotional activities of the target bank based on its own bank attributes, such as supported credit card types, bank card types, supported transaction times and other information, and obtain regional push data.

[0163] Taking early warning information as an example, query the target bank with high capital outflow (capital outflow greater than the first preset threshold) to obtain its early warning information. The early warning information is the first push data. Since the differences between each line of early warning information are small, the first push data can be directly used as the regional push data.

[0164] In addition to using capital outflow to generate regional push data, you can also generate regional push data based on regional outflow characteristics.

[0165] Specifically, still taking marketing as an example, according to the map of capital outflow in cities, provincial users can see that city A has more capital outflow (capital outflow is not less than the second preset threshold) and city B has less capital outflow (capital outflow is less than the second preset threshold). If the city to be pushed information is city B, since the capital outflow situation is OK, no information push operation can be performed. However, for city A, the capital outflow is large, and the marketing activities of city B can be adaptively promoted to city A. For example, the marketing activities of city B are used as the second push data, and the second push data is adjusted according to the attribute information of city A, such as geographical factors and human factors, to obtain regional push data.

[0166] 2) According to the dimension value of the regional push dimension, a push operation is performed on the regional push data.

[0167] In this embodiment, data resources such as manpower, network traffic, funds, memory capacity, equipment occupancy, etc. are used when information is pushed.

[0168] In actual applications, when information is pushed, there will be multiple regions, and it is necessary to consider how much data resources to allocate to each region.

[0169] In specific implementation, the dimension value of the region push dimension can be used to call the classifier to allocate the total data resources to obtain the data resource allocation amount of each region.

[0170] Specifically, a classifier such as a SVM (Support Vector Machine) classifier can be used to classify data resources. When SVM performs classification, the larger the dimension value of the regional push dimension, the larger the amount of distributed data resources.

[0171] Then, the dimension value of the regional push dimension is used to call the classifier to allocate the total data resources, and the data resource allocation amount of each region is obtained. When allocating specific data resources, one-click allocation is supported. The input is the dimension value of the regional push dimension of multiple cities, and the output is the data resource allocation amount (in the marketing scenario, the marketing resource allocation amount) sorted from large to small. Then, according to the data resource allocation amount, the corresponding regional push data is pushed. That is, data push operations are performed according to the allocated manpower, network traffic, funds, memory, equipment occupancy, etc. If the dimension value of the regional push dimension of a certain region is large, ten processors can be used for data push, while the dimension value of the regional push dimension of another region is small, then five processors can be used for data push.

[0172] It should be noted that after obtaining the allocated data resources, support managers can adjust the marketing resource allocation for all cities and counties in the province according to this order. If the user believes that a certain area should be allocated more resources, more resources can be allocated to it.

[0173] In addition, based on the monthly line graph generated based on the monthly outflow of funds, provincial users can count the months with relatively more capital outflows in a year and carry out more intensive marketing activities in these months.

[0174] In this embodiment, when information is pushed, it is pushed from two strategic perspectives, namely, content and resources, so as to improve the matching accuracy between information push and region, and also ensure the reasonable allocation of resources.

[0175] In another implementation of the present invention, in order to implement the above-mentioned data push, a data push system is provided, such as Figure 5 As shown, it includes a big data platform for statistics and unified storage of basic data such as bank fund flow information; a data extraction system for extracting information such as financial fund flow data from the bank's big data platform and generating files with unified naming specifications in a fixed format; a data storage subsystem for periodically scanning the fund flow data files extracted by the data extraction subsystem, and if a data file is scanned, importing it into the corresponding table of the database; a data analysis subsystem for analyzing four characteristics of the total amount of bank outflow funds, the bank to which funds flow, the region to which the outflow funds belong, and the time of fund outflow according to the data fields in the database table; a data display subsystem for displaying the characteristics obtained by the data analysis subsystem on the front-end page, and conducting human-computer interaction with bank personnel. Bank personnel can intuitively analyze the dynamics of fund flows based on the page characteristics, thereby improving the richness and accuracy of the formulated information and enhancing the competitiveness of the bank.

[0176] Based on the above data push method embodiment, another embodiment of the present invention provides a data push device, referring to Figure 6 , which may include:

[0177] The data acquisition module 11 is used to acquire the fund flow data of the target account; each data field in the fund flow data is separated by a preset delimiter, and each data field is arranged in a preset order;

[0178] A statistical analysis module 12 is used to perform statistical analysis on the fund flow data to obtain characteristic values ​​of each preset flow characteristic;

[0179] An index value calculation module 13, used to calculate the index value of the data push index based on the characteristic values ​​of each preset flow direction characteristic;

[0180] A dimension value determination module 14, used to determine the dimension value of the regional push dimension corresponding to the indicator value of the data push indicator;

[0181] The data push module 15 is used to push the regional data using the characteristic values ​​of the preset flow characteristics and the dimension value of the regional push dimension.

[0182] In an optional implementation, the statistical analysis module 12 includes:

[0183] The first analysis submodule is used to analyze the total amount of capital outflow in the capital flow data to obtain a characteristic value of the total amount of outflow characteristics;

[0184] The second analysis submodule is used to perform a fund amount calculation operation on the fund flow data according to the bank flow dimension to obtain a characteristic value of the bank flow feature;

[0185] The third analysis submodule is used to perform a capital amount calculation operation on the capital flow data according to the regional dimension to obtain a characteristic value of the regional outflow characteristic;

[0186] A fourth analysis submodule is used to perform a capital amount calculation operation on the capital flow data according to the time dimension to obtain a characteristic value of a preset period outflow characteristic;

[0187] The preset flow direction characteristics include the total outflow amount characteristics, the bank flow characteristics, the regional outflow characteristics and the preset period outflow characteristics.

[0188] In an optional implementation, the indicator value calculation module 13 includes:

[0189] A first calculation submodule is used to calculate the index value of the regional outflow proportion index by using the characteristic value of the total outflow characteristic and the characteristic value of the regional outflow characteristic;

[0190] A second calculation submodule is used to calculate the index value of the regional regression rate index based on the characteristic value of the total outflow characteristic and the characteristic value of the bank flow characteristic;

[0191] A third calculation submodule is used to calculate the index value of the regional change degree index corresponding to the characteristic value of the outflow characteristic of the preset period;

[0192] The data push indicators include the regional outflow proportion indicator, the regional return rate indicator and the regional change degree indicator.

[0193] In an optional implementation, the third computing submodule includes:

[0194] A sorting unit, used to sort the characteristic values ​​of the preset period outflow characteristics of each target account to obtain an outflow volume sequence;

[0195] A difference calculation unit, used for calculating the difference between adjacent data in the outflow sequence to obtain a difference sequence;

[0196] A stability analysis unit, used for performing a stability analysis operation on the difference sequence to obtain a fluctuation subsequence in the difference sequence;

[0197] A variance calculation unit, used to calculate the variance corresponding to the volatility subsequence to obtain a volatility variance value sequence for different regions;

[0198] The index value calculation unit is used to calculate the index values ​​of the regional change degree index of different regions according to the fluctuation variance value sequence of different regions.

[0199] In an optional implementation, the indicator value calculation unit includes:

[0200] A clustering subunit is used for performing a K-means clustering operation on the volatility variance value sequences of different regions to obtain a plurality of clusters;

[0201] The data calculation subunit is used to calculate the mean value of the fluctuation degree of each cluster and the number of sample points in the cluster;

[0202] The index value calculation subunit is used to take the cluster corresponding to the largest number of sample points in the cluster as the final fluctuation degree cluster, and take the fluctuation degree mean value of the final fluctuation degree cluster as the index value of the regional change degree index of the corresponding region.

[0203] In an optional implementation, the dimension value determination module 14 is specifically used for:

[0204] The index value of the regional outflow proportion index, the index value of the regional regression rate index and the index value of the regional change degree index are weightedly summed to obtain the dimension value of the regional push dimension.

[0205] In an optional implementation, the data push module 15 includes:

[0206] a data determination submodule, configured to determine first push data of a target bank whose fund outflow is greater than a first preset threshold value based on the characteristic value of the bank-to-bank characteristic, and generate regional push data based on the first push data, or to determine second push data of a target region whose fund outflow is less than a second preset threshold value based on the characteristic value of the regional outflow characteristic, and generate regional push data based on the second push data;

[0207] The data push submodule is used to push the regional push data according to the dimension value of the regional push dimension.

[0208] In an optional implementation, the data push submodule includes:

[0209] A resource allocation unit, configured to use the dimension value of the region push dimension to call the classifier to allocate the total data resources and obtain the data resource allocation amount for each region;

[0210] The data push unit is used to push data to corresponding regions according to the data resource allocation amount.

[0211] In this embodiment, when pushing information, the user's fund flow situation will be considered. The fund flow situation can reflect the user's focus, and then push based on the fund flow, which can make the pushed information more compatible with the user. Further, the present invention performs statistical analysis operations on the fund flow data to obtain the characteristic values ​​of each preset flow feature, calculates the index value of the data push index based on the characteristic values ​​of each preset flow feature, and determines the dimension value of the regional push dimension corresponding to the index value of the data push index. That is, when pushing data, the present invention will consider the situation of the user's region, and then push appropriately for the region, so that the pushed information matches the region, improves the matching degree between the pushed information and the users in the region, and then improves the accuracy of the pushed information.

[0212] It should be noted that, for the working process of each module, sub-module, unit and sub-unit in this embodiment, please refer to the corresponding description in the above embodiment, which will not be repeated here.

[0213] The present application also provides an electronic device in an embodiment. Figure 7 As shown, it shows a schematic diagram of the structure of an electronic device suitable for implementing the embodiments of the present application. The electronic device in the embodiments of the present application may include but is not limited to fixed terminals such as mobile phones, notebook computers, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Figure 7 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0214] like Figure 7 As shown, the electronic device may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603 to implement the above-mentioned data push method. When the electronic device is powered on, various programs and data required for the operation of the electronic device are also stored in the RAM 603. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0215] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a memory card, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although Figure 7 An electronic device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0216] An embodiment of the present application also provides a computer program product including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements any one of the data push methods provided in the embodiments of the present application.

[0217] A computer-readable storage medium is also provided in an embodiment of the present application. The storage medium carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement any data push method provided in the embodiment of the present application.

[0218] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A data push method, characterized in that: include: Obtaining fund flow data of a target account; each data field in the fund flow data is separated by a preset delimiter, and each data field is arranged in a preset order; Performing statistical analysis on the capital flow data to obtain characteristic values ​​of each preset flow characteristic; Calculating the index value of the data push index based on the characteristic values ​​of each preset flow direction characteristic; Determine a dimension value of a regional push dimension corresponding to the indicator value of the data push indicator; The regional data push operation is performed using the feature values ​​of the preset flow direction features and the dimension value of the regional push dimension.

2. The data push method according to claim 1, characterized in that: Perform statistical analysis on the fund flow data to obtain characteristic values ​​of various preset flow characteristics, including: Analyze the total amount of capital outflow in the capital flow data to obtain a characteristic value of the total amount of outflow characteristics; Performing a fund amount calculation operation on the fund flow data according to the bank flow dimension to obtain a characteristic value of the bank flow characteristic; Performing a capital amount calculation operation on the capital flow data according to the regional dimension to obtain a characteristic value of the regional outflow characteristic; Performing a capital amount calculation operation on the capital flow data according to the time dimension to obtain a characteristic value of a preset period outflow characteristic; The preset flow direction characteristics include the total outflow amount characteristics, the bank flow characteristics, the regional outflow characteristics and the preset period outflow characteristics.

3. The data push method according to claim 2, characterized in that: Calculating the index value of the data push index based on the characteristic values ​​of each preset flow direction characteristic includes: Calculate the index value of the regional outflow proportion index by using the characteristic value of the total outflow characteristic and the characteristic value of the regional outflow characteristic; Calculate the index value of the regional regression rate index based on the characteristic value of the total outflow characteristic and the characteristic value of the bank flow characteristic; Calculate the index value of the regional change degree index corresponding to the characteristic value of the outflow characteristic of the preset period; The data push indicators include the regional outflow proportion indicator, the regional return rate indicator and the regional change degree indicator.

4. The data push method according to claim 3, characterized in that: Calculating the index value of the regional change degree index corresponding to the characteristic value of the preset period outflow characteristic includes: Sort the characteristic values ​​of the preset period outflow characteristics of each target account to obtain an outflow volume sequence; Calculating the difference between adjacent data in the outflow sequence to obtain a difference sequence; Performing a stability analysis operation on the difference sequence to obtain a fluctuation subsequence in the difference sequence; Calculating the variance corresponding to the volatility subsequence to obtain a volatility variance value sequence for different regions; According to the volatility variance value series of different regions, the index values ​​of regional change degree index of different regions are calculated.

5. The data push method according to claim 4, characterized in that: According to the volatility variance value sequence of different regions, the index values ​​of regional change degree indicators of different regions are calculated, including: For the volatility variance value sequences of different regions, a K-means clustering operation is performed on the volatility variance value sequences to obtain multiple clusters; Calculate the mean fluctuation degree of each cluster and the number of sample points in the cluster; The cluster corresponding to the largest number of sample points in the cluster is taken as the final fluctuation degree cluster, and the fluctuation degree mean value of the final fluctuation degree cluster is taken as the index value of the regional change degree index of the corresponding region.

6. The data push method according to claim 5, characterized in that: Determining the dimension value of the regional push dimension corresponding to the indicator value of the data push indicator includes: The index value of the regional outflow proportion index, the index value of the regional regression rate index and the index value of the regional change degree index are weightedly summed to obtain the dimension value of the regional push dimension.

7. The data push method according to claim 2, characterized in that: Using the characteristic values ​​of each of the preset flow direction characteristics and the dimension value of the regional push dimension, a regional data push operation is performed, including: Based on the characteristic value of the bank flow characteristic, determine first push data of a target bank whose fund outflow amount is greater than a first preset threshold, and generate regional push data based on the first push data; or, based on the characteristic value of the regional outflow characteristic, determine second push data of a target region whose fund outflow amount is less than a second preset threshold, and generate regional push data based on the second push data; According to the dimension value of the regional push dimension, a push operation is performed on the regional push data.

8. The data push method according to claim 7, characterized in that: According to the dimension value of the regional push dimension, a push operation is performed on the regional push data, including: Using the dimension value of the regional push dimension, calling the classifier to allocate the total data resources, and obtaining the data resource allocation amount of each region; According to the data resource allocation amount, the push data is pushed to the corresponding area.

9. A data push device, characterized in that: include: A data acquisition module is used to acquire the funds flow data of the target account; each data field in the funds flow data is separated by a preset delimiter, and each data field is arranged in a preset order; A statistical analysis module, used to perform statistical analysis on the fund flow data to obtain characteristic values ​​of each preset flow characteristic; An index value calculation module, used to calculate the index value of the data push index based on the characteristic values ​​of each preset flow direction characteristic; A dimension value determination module, used to determine the dimension value of the regional push dimension corresponding to the indicator value of the data push indicator; The data push module is used to use the characteristic values ​​of the preset flow characteristics and the dimension value of the regional push dimension to push regional data.

10. An electronic device, characterized in that: The method comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the electronic device can implement the data push method as described in any one of claims 1 to 8.

Images