CN113962323A - Hot account identification method and device - Google Patents

Hot account identification method and device Download PDF

Info

Publication number
CN113962323A
CN113962323A CN202111286675.3A CN202111286675A CN113962323A CN 113962323 A CN113962323 A CN 113962323A CN 202111286675 A CN202111286675 A CN 202111286675A CN 113962323 A CN113962323 A CN 113962323A
Authority
CN
China
Prior art keywords
transaction
account
time period
transaction account
transaction amount
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
CN202111286675.3A
Other languages
Chinese (zh)
Inventor
王晓旭
戈星晨
张骁
朱江波
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.)
Bank of China Ltd
Original Assignee
Bank of China 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 Bank of China Ltd filed Critical Bank of China Ltd
Priority to CN202111286675.3A priority Critical patent/CN113962323A/en
Publication of CN113962323A publication Critical patent/CN113962323A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
    • G06F18/24133—Distances to prototypes
    • G06F18/24137—Distances to cluster centroïds
    • G06F18/2414—Smoothing the distance, e.g. radial basis function networks [RBFN]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2462—Approximate or statistical queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2477—Temporal data queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/02—Banking, e.g. interest calculation or account maintenance

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Mathematical Physics (AREA)
  • Computational Linguistics (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Business, Economics & Management (AREA)
  • Probability & Statistics with Applications (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Fuzzy Systems (AREA)
  • Biophysics (AREA)
  • Finance (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Biomedical Technology (AREA)
  • Health & Medical Sciences (AREA)
  • Databases & Information Systems (AREA)
  • Accounting & Taxation (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Technology Law (AREA)
  • General Business, Economics & Management (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)

Abstract

The invention discloses a hot spot account identification method and a device, which relate to the technical field of artificial intelligence, wherein the method comprises the following steps: inputting the transaction account in the current time period and the transaction characteristic data corresponding to the transaction account into a transaction amount prediction model to obtain the transaction amount of the transaction account in the next time period; judging whether the transaction account is a hot account or not according to the transaction amount of the transaction account in the next time period and the system performance of the banking system; and when the transaction account is the hotspot account, adding the transaction account into the hotspot account list. The invention can autonomously identify the hot account under the condition that the transaction frequency of the transaction account in a certain time period is greater than the preset transaction frequency and the transaction account becomes the hot account, thereby ensuring the stable and efficient operation of a banking system and ensuring the user experience.

Description

Hot account identification method and device
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a hot spot account identification method and device.
Background
This section is intended to provide a background or context to the embodiments of the invention that are recited in the claims. The description herein is not admitted to be prior art by inclusion in this section.
In a banking system, in order to make transactions as unaffected as possible by hot accounts, it is necessary to accurately identify hot accounts. In the existing hot spot account identification method, data of whether a common account belongs to a hot spot account (whether the common account belongs to the hot spot account or not means that a transaction account has a rapidly increased transaction amount in a certain time period, is higher than a preset value and is about to become the hot spot account) is pre-embedded data or manually added data, but a time delay or an error is likely to occur in manual judgment, if one common account becomes the hot spot account with a huge transaction amount under the condition that the common account is not set as the hot spot account, the transaction pressure born by a bank system is huge, so that service cannot be provided, and customer experience is influenced.
Disclosure of Invention
The embodiment of the invention provides a hot account identification method, which is used for automatically identifying a hot account under the condition that the transaction frequency of the transaction account in a certain time period is greater than the preset transaction frequency and the transaction account is to become the hot account, so that the stable and efficient operation of a banking system and the user experience can be ensured, and the method comprises the following steps:
inputting the transaction account in the current time period and the transaction characteristic data corresponding to the transaction account into a transaction amount prediction model to obtain the transaction amount of the transaction account in the next time period, wherein the current time period is a time period from a preset starting time to the current time, and the transaction amount prediction model is obtained by training a neural network model according to the historical transaction account in a banking system, the transaction characteristic data corresponding to the historical transaction account and the transaction amount in the next time period of the historical transaction account;
judging whether the transaction account is a hot account or not according to the transaction amount of the transaction account in the next time period and the system performance of the banking system;
and when the transaction account is the hotspot account, adding the transaction account into the hotspot account list.
The embodiment of the present invention further provides a hot spot account identification apparatus, configured to identify a hot spot account autonomously when a transaction frequency of a transaction account within a certain time period is greater than a preset transaction frequency and the transaction account is to become a hot spot account, so as to ensure stable and efficient operation of a banking system and ensure user experience, where the apparatus includes:
the prediction module is used for inputting the transaction account in the current time period and the transaction characteristic data corresponding to the transaction account into a transaction amount prediction model to obtain the transaction amount of the transaction account in the next time period, wherein the current time period is a time period from a preset starting time to the current time, and the transaction amount prediction model is obtained by training a neural network model according to the historical transaction account in a banking system, the transaction characteristic data corresponding to the historical transaction account and the transaction amount in the next time period of the historical transaction account;
the judging module is used for judging whether the transaction account is a hot account or not according to the transaction amount of the transaction account in the next time period and the system performance of the banking system;
and the adding module is used for adding the transaction account into the hot account list when the transaction account is the hot account.
The embodiment of the invention also provides computer equipment which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the hot spot account identification method is realized when the processor executes the computer program.
The embodiment of the invention also provides a computer readable storage medium, and the computer readable storage medium stores a computer program for executing the hot spot account identification method.
In the embodiment of the invention, a transaction account in the current time period and transaction characteristic data corresponding to the transaction account are input into a transaction amount prediction model to obtain the transaction amount of the transaction account in the next time period, wherein the current time period is a time period from a preset starting time to the current time, and the transaction amount prediction model is obtained by training a neural network model according to a historical transaction account in a banking system, the transaction characteristic data corresponding to the historical transaction account and the transaction amount in the next time period of the historical transaction account; judging whether the transaction account is a hot account or not according to the transaction amount of the transaction account in the next time period and the system performance of the banking system; and when the transaction account is the hotspot account, adding the transaction account into the hotspot account list. Compared with the existing mode of manually judging the hot account, the method and the system have the advantages that the transaction amount of the transaction account in the next time period is predicted according to the transaction data in the current time period, and then whether the transaction account is the hot account or not can be automatically judged according to the transaction amount of the transaction account in the next time period and the system performance of the banking system, so that the stable and efficient operation of the banking system can be ensured, and the user experience can be ensured.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts. In the drawings:
fig. 1 is a first flowchart of a hot spot account identification method provided in an embodiment of the present invention;
FIG. 2 is a flow chart of a method for training a transaction prediction model according to an embodiment of the present invention;
fig. 3 is a flowchart of a hot spot account identification method according to an embodiment of the present invention;
fig. 4 is a flowchart three of a hot spot account identification method provided in the embodiment of the present invention;
fig. 5 is a fourth flowchart of a hot spot account identification method provided in the embodiment of the present invention;
FIG. 6 is a flow chart of a method for updating a transaction prediction model in accordance with an embodiment of the present invention;
fig. 7 is a first schematic diagram of a hot spot account identification apparatus according to an embodiment of the present invention;
fig. 8 is a second schematic diagram of a hot spot account identification apparatus according to an embodiment of the present invention;
fig. 9 is a third schematic diagram of a hot spot account identification apparatus provided in the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention.
In the description of the present specification, the terms "comprising," "including," "having," "containing," and the like are used in an open-ended fashion, i.e., to mean including, but not limited to. Reference to the description of the terms "one embodiment," "a particular embodiment," "some embodiments," "for example," etc., means that a particular feature, structure, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. The sequence of steps involved in the embodiments is for illustrative purposes to illustrate the implementation of the present application, and the sequence of steps is not limited and can be adjusted as needed.
Interpretation of terms:
hotspot account: it is meant that the transaction frequency is maintained at a higher number of accounts for a certain period of time during the transaction. In the transaction process, account balance data corresponding to the hot account is frequently updated, so that a bank system may encounter a database processing bottleneck, and the transaction pressure is increased suddenly.
In order to solve the problem that a common account becomes a hot spot account with a huge transaction amount when not set as the hot spot account, which may cause that a banking system bears a huge transaction pressure so as not to provide a service, an embodiment of the present invention provides a hot spot account identification method, as shown in fig. 1, the method includes the following steps:
step 101, inputting a transaction account in a current time period and transaction characteristic data corresponding to the transaction account into a transaction amount prediction model to obtain a transaction amount of the transaction account in a next time period, wherein the current time period is a time period from a preset starting time to the current time, and the transaction amount prediction model is obtained by training a neural network model according to a historical transaction account in a banking system, the transaction characteristic data corresponding to the historical transaction account and the transaction amount in the next time period of the historical transaction account;
step 102, judging whether the transaction account is a hot account or not according to the transaction amount of the transaction account in the next time period and the system performance of the banking system;
and 103, adding the transaction account into the hot account list when the transaction account is the hot account.
Compared with the existing mode of manually judging the hot account, the method and the device have the advantages that the transaction amount of the transaction account in the next time period is predicted according to the transaction data in the current time period through the transaction amount prediction model, and then whether the transaction account is the hot account or not can be autonomously judged according to the transaction amount of the transaction account in the next time period and the system performance of the banking system, so that the stability and the high efficiency of the banking system can be guaranteed, and the user experience can be guaranteed.
In step 101, the transaction information of the banking system may be monitored by the daily operation and maintenance monitoring module, and a transaction account in the banking system in the current time period and transaction characteristic data corresponding to the transaction account in the current time period are obtained.
Specifically, a starting time may be preset, and transaction data in a time period from the starting time to the current time (i.e., the current time period) may be acquired, where the current time period may be an integer of minutes. And then, extracting the transaction account in the transaction data in the current time period and the transaction characteristic data corresponding to the transaction account. The transaction characteristic data may include a product type, a transaction channel, and a transaction lending direction.
And inputting the extracted transaction account and the transaction characteristic data corresponding to the transaction account into a transaction amount prediction model to obtain the transaction amount of the transaction account in the next time period.
Before step 101, a transaction amount prediction model obtained by training a neural network model may be obtained according to a historical transaction account in a banking system, transaction characteristic data corresponding to the historical transaction account, and a transaction amount in a next time period of the historical transaction account.
As shown in fig. 2, an embodiment of the present invention provides a flowchart of a method for training a transaction amount prediction model, which may include the following steps:
step 201, acquiring a historical transaction account, transaction characteristic data corresponding to the historical transaction account and a transaction amount of the historical transaction account in a next time period in a banking system as sample data, and constructing a training set and a test set, wherein the historical transaction data comprises the transaction account and the historical transaction characteristic data corresponding to the transaction account;
202, training a neural network model by using a training set to obtain a transaction amount prediction model;
step 203, the transaction amount prediction model is tested by using the test set.
In an embodiment of the present invention, in the step 201, the method specifically includes:
desensitizing the historical transaction account, transaction characteristic data corresponding to the historical transaction account and transaction amount in the next time period of the historical transaction account to be used as sample data; dividing sample data according to a sliding window algorithm by taking a preset fixed time length as the size of a sliding window to obtain a plurality of sample data sets; and constructing a training set and a test set according to a plurality of sample data sets.
In specific implementation, the historical transaction account in the banking system, the transaction characteristic data corresponding to the historical transaction account and the transaction amount of the transaction account of the historical transaction account in the next time period are acquired, and in order to enable sample data to meet the data security standard, the sample data is taken as the sample data after desensitization, cleaning and other processing is carried out on the historical transaction data and the transaction amount of the transaction account in the next time period. Then, presetting a fixed time length T, taking the T as the size of a sliding window, and dividing the sample data according to a sliding window algorithm to obtain a plurality of sample data sets with the time length of T. And counting sample data in each sample data set, extracting a transaction account in the sample data and historical transaction characteristic data corresponding to the transaction account, and carrying out standardized processing on the historical transaction characteristic data. Then, sample data is divided into a training set and a test set according to a preset proportion.
For example, T is 5 minutes, 10 sample data sets with 5 minutes as a unit are obtained after sample data is divided, 8 sample data sets are used as a training set, and 2 sample data sets are used as a test set.
In addition, the neural network model may be a Long Short-Term Memory network (LSTM), and the LSTM is trained by using a training set, so that a cyclic neural network model for predicting the transaction amount of the transaction account in the next time period, that is, a transaction amount prediction model, may be obtained.
Therefore, through desensitization, cleaning, feature extraction, standardization processing and the like of the historical transaction data in the banking system and the transaction amount in the next time period of the corresponding transaction account, the selected historical transaction data are more representative, and the prediction result of the finally trained transaction amount prediction model is more accurate.
It should be noted that, in the step 101, the duration of the next time period is the same as the duration of the current time period, and the next time period is the starting time at the current time.
In the above step 102, it is determined whether the transaction account is a hot account according to the predicted transaction amount of the transaction account in the next time period and the system performance of the banking system in the step 101.
It should be noted that, in different banking systems, the conditions for determining hot accounts are also different. Under the condition that the system performance of the banking system is good, the efficiency of processing the transaction is also high, and then the requirement on the hot account in a certain time period is also high.
In an embodiment of the present invention, as shown in fig. 3, the step 102 may specifically include:
step 301, obtaining historical transaction amounts of a transaction account in a current time period and a plurality of time periods before the current time period.
Here, the time lengths of the plurality of time periods before the current time period are the same as the time length of the current time period, and the plurality of time periods are a plurality of time periods that are continuous with the current time period. For example, the current time period is 10-11 o 'clock, and 3 time periods before 10 o' clock may be: 9-10 points, 8-9 points, and 7-8 points. Then, the historical transaction amounts of transaction account a are 10-11 points, respectively: n1; 9 points-10 points: n2; 8 points-9 points: n3; 7 point-8 point: n4.
Step 302, a transaction amount set is constructed according to the transaction amount of the transaction account in the next time period and the historical transaction amount.
Here, based on the description of step 301 above, for example, the next time period is 11 points-12 points, and the predicted value of the 11 points-12 points transaction amount of the transaction account a is R. Then, the resulting transaction amount set is Y ═ { R, N1, N2, N3, N4 }.
Step 303, judging whether the transaction account is a hot account according to the system performance and the transaction amount set of the banking system.
Here, in an embodiment of the present invention, as shown in fig. 4, step 303 may specifically include:
step 401, obtaining performance parameters of system performance, where the performance parameters include the number of concurrent transactions processed by the banking system, the highest threshold percentage of the number of concurrent transactions, the lowest threshold percentage of the number of concurrent transactions, the highest threshold percentage of the average number of concurrent transactions, and the lowest threshold percentage of the average number of concurrent transactions.
The banking system concurrently processes the transaction amount refers to the transaction amount that can be successfully processed per second by a single account.
It should be noted that the number of concurrent transactions processed by the banking system is not fixed, and according to a specific application scenario, the number of concurrent transactions may float up or down to a certain extent. Thus, the performance parameters for system performance also include a highest threshold percentage of the number of concurrently processed transactions, a lowest threshold percentage of the number of concurrently processed transactions, a mean highest threshold percentage of the number of concurrently processed transactions, and a mean lowest threshold percentage of the number of concurrently processed transactions.
Step 402, determining the transaction account as a hot account if the transaction amount of the transaction account in the next time period is greater than or equal to the product of the number of concurrent transactions and the highest threshold percentage of the number of concurrent transactions, and the mean value of the transaction amount set is greater than the product of the number of concurrent transactions and the average highest threshold percentage of the number of concurrent transactions.
In step 103, when the transaction account is a hot account, the transaction account is added to the hot account list.
After the transaction account is added into the hot spot account list, when the transaction account is transacted again, the transaction of the transaction account can be processed by using the hot spot account processing branch, so that the risk that the common account becomes the hot spot account with huge transaction amount under the condition that the common account is not set as the hot spot account, and the transaction pressure born by a banking system is huge, and the service cannot be provided is reduced.
In another embodiment of the present invention, as shown in fig. 5, after step 401, the method may further include:
step 501, when the transaction amount of a transaction account in the next time period is smaller than the product of the concurrent transaction amount and the minimum threshold percentage of the concurrent transaction amount, and the mean value of the transaction amount set is smaller than the product of the concurrent transaction amount and the average minimum threshold percentage of the concurrent transaction amount, and the transaction account is in a hot account list, determining that the transaction account is a non-hot account;
step 502, delete transaction account from hotspot account list.
After the transaction account is deleted from the hotspot account list and when the transaction account is transacted again, the transaction of the transaction account is processed by utilizing a common account processing branch of the banking system.
For example, let the transaction amount of the transaction account in the next time period be R; the number of concurrently processed transactions by the banking system may be X, the highest threshold percentage P of the number of concurrently processed transactions1120 percent; a minimum threshold percentage P of the number of concurrent transactions2Is 50 percent; a mean highest threshold percentage P of the number of concurrently processed transactions3Is 90%; average minimum threshold percentage P of the number of concurrently processed transactions4Is 60 percent; then:
when R is larger than or equal to X by 120% and MEAN (Y) is larger than X by 90%, judging A as a hot spot account;
and when R is less than X by 50%, MEAN (Y) is less than X by 60% and A is in the hot spot account list, judging A is not a hot spot account.
Mean (Y) is the mean of the transaction amount set Y.
Therefore, according to the transaction amount in the next time period of the transaction account, the current time period of the transaction account, the historical transaction amounts in a plurality of time periods before the current time period and the system performance of the banking system, whether the transaction account is changed from the common account to the hot account or from the hot account to the common account in the next time period can be accurately judged, and then a corresponding processing mode can be configured for the transaction of the transaction account according to the judgment result, so that the stable and efficient operation of the banking system and the user experience can be ensured.
In addition, in an embodiment of the present invention, the transaction account in the hot account list of the current day, the transaction characteristic data corresponding to the transaction account, and the transaction amount of the transaction account corresponding to each time period of the current day may also be added to the sample data, and the transaction amount prediction model is continuously trained and tuned, so as to ensure that the generated prediction result is more accurate.
Fig. 6 is a flowchart of a method for updating a transaction prediction model according to an embodiment of the present invention. When the transaction account is a hotspot account, after the transaction account is added to the hotspot account list, the method further comprises the following steps:
step 601, updating sample data by using a transaction account in a hot spot account list of the day, transaction characteristic data corresponding to the transaction account and a transaction amount corresponding to the transaction account in each time period of the day;
and step 602, updating the training set and the test set by using the updated sample data, and continuing to train and test the transaction amount prediction model.
The embodiment of the invention also provides a hot spot account identification device, which is described in the following embodiment. Because the principle of the device for solving the problems is similar to that of the hot spot account identification method, the implementation of the device can refer to the implementation of the hot spot account identification method, and repeated details are not repeated.
As shown in fig. 7, which is a schematic diagram of a hotspot account identification device provided in an embodiment of the present invention, the device includes:
the prediction module 701 is used for inputting the transaction account in the current time period and the transaction characteristic data corresponding to the transaction account into a transaction amount prediction model to obtain the transaction amount of the transaction account in the next time period, wherein the current time period is a time period from a preset starting time to the current time, and the transaction amount prediction model is obtained by training a neural network model according to the historical transaction account in the banking system, the transaction characteristic data corresponding to the historical transaction account and the transaction amount in the next time period of the historical transaction account;
the determining module 702 is configured to determine whether the transaction account is a hot account according to the transaction amount of the transaction account in the next time period and the system performance of the banking system;
the adding module 703 is configured to add the transaction account to the hot account list when the transaction account is a hot account.
In an embodiment of the present invention, the determining module is specifically configured to:
acquiring historical transaction amounts of a transaction account in a current time period and a plurality of time periods before the current time period;
constructing a transaction amount set according to the transaction amount of the transaction account in the next time period and the historical transaction amount;
and judging whether the transaction account is a hot account or not according to the system performance and the transaction amount set of the banking system.
In an embodiment of the present invention, the determining module is further specifically configured to:
acquiring performance parameters of system performance, wherein the performance parameters comprise the quantity of concurrent transactions processed by a banking system, the highest threshold percentage of the quantity of concurrent transactions, the lowest threshold percentage of the quantity of concurrent transactions, the average highest threshold percentage of the quantity of concurrent transactions and the average lowest threshold percentage of the quantity of concurrent transactions;
determining that the transaction account is a hotspot account if the transaction amount of the transaction account in the next time period is greater than or equal to the product of the number of concurrently processed transactions and the highest threshold percentage of the number of concurrently processed transactions, and the mean of the set of transaction amounts is greater than the product of the number of concurrently processed transactions and the average highest threshold percentage of the number of concurrently processed transactions.
In an embodiment of the present invention, the determining module is further specifically configured to:
determining that the transaction account is a non-hotspot account when the transaction amount of the transaction account in the next time period is less than the product of the concurrent transaction amount and the lowest threshold percentage of the concurrent transaction amount, and the mean value of the transaction amount set is less than the product of the concurrent transaction amount and the average lowest threshold percentage of the concurrent transaction amount, and the transaction account is in a hotspot account list;
the transaction account is deleted from the hotspot account list.
In an embodiment of the present invention, as shown in fig. 8, the model training and testing module 801 is further included, configured to, before the transaction account and the transaction characteristic data corresponding to the transaction account in the current time period are input into the transaction amount prediction model in the prediction module:
acquiring a historical transaction account, transaction characteristic data corresponding to the historical transaction account and transaction amount of the historical transaction account in a next time period in a banking system as sample data, and constructing a training set and a testing set, wherein the historical transaction data comprises the transaction account and the historical transaction characteristic data corresponding to the transaction account;
training a neural network model by using a training set to obtain a transaction amount prediction model;
and testing the transaction amount prediction model by using the test set.
In an embodiment of the present invention, the model training and testing module shown in fig. 8 is further specifically configured to:
desensitizing the historical transaction account, transaction characteristic data corresponding to the historical transaction account and transaction amount in the next time period of the historical transaction account to be used as sample data;
dividing sample data according to a sliding window algorithm by taking a preset fixed time length as the size of a sliding window to obtain a plurality of sample data sets;
and constructing a training set and a test set according to a plurality of sample data sets.
In an embodiment of the present invention, as shown in fig. 9, the present invention further includes a model optimization module 901, which is specifically configured to:
updating sample data by using a transaction account in a hot spot account list of the current day, transaction characteristic data corresponding to the transaction account and a transaction amount corresponding to the transaction account in each time period of the current day;
and updating the training set and the test set by using the updated sample data, and continuing to train and test the transaction amount prediction model.
In an embodiment of the present invention, the transaction characteristic data includes a product type, a transaction channel and a transaction lending direction.
The embodiment of the invention also provides computer equipment which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the hot spot account identification method is realized when the processor executes the computer program.
The embodiment of the invention also provides a computer readable storage medium, and the computer readable storage medium stores a computer program for executing the hot spot account identification method.
In the embodiment of the invention, a transaction account in the current time period and transaction characteristic data corresponding to the transaction account are input into a transaction amount prediction model to obtain the transaction amount of the transaction account in the next time period, wherein the current time period is a time period from a preset starting time to the current time, and the transaction amount prediction model is obtained by training a neural network model according to a historical transaction account in a banking system, the transaction characteristic data corresponding to the historical transaction account and the transaction amount in the next time period of the historical transaction account; judging whether the transaction account is a hot account or not according to the transaction amount of the transaction account in the next time period and the system performance of the banking system; and when the transaction account is the hotspot account, adding the transaction account into the hotspot account list. Compared with the existing mode of manually judging the hot account, the method and the system have the advantages that the transaction amount of the transaction account in the next time period is predicted according to the transaction data in the current time period, and then whether the transaction account is the hot account or not can be automatically judged according to the transaction amount of the transaction account in the next time period and the system performance of the banking system, so that the stable and efficient operation of the banking system can be ensured, and the user experience can be ensured.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims (18)

1. A hotspot account identification method is characterized by comprising the following steps:
inputting the transaction account in the current time period and the transaction characteristic data corresponding to the transaction account into a transaction amount prediction model to obtain the transaction amount of the transaction account in the next time period, wherein the current time period is a time period from a preset starting time to the current time, and the transaction amount prediction model is obtained by training a neural network model according to the historical transaction account in a banking system, the transaction characteristic data corresponding to the historical transaction account and the transaction amount in the next time period of the historical transaction account;
judging whether the transaction account is a hot account or not according to the transaction amount of the transaction account in the next time period and the system performance of the banking system;
and when the transaction account is the hotspot account, adding the transaction account into the hotspot account list.
2. The method of claim 1, wherein determining whether the transaction account is a hot account based on the transaction amount of the transaction account in the next time period and the system performance of the banking system comprises:
acquiring historical transaction amounts of a transaction account in a current time period and a plurality of time periods before the current time period;
constructing a transaction amount set according to the transaction amount of the transaction account in the next time period and the historical transaction amount;
and judging whether the transaction account is a hot account or not according to the system performance and the transaction amount set of the banking system.
3. The method of claim 2, wherein determining whether the transaction account is a hotspot account based on the system performance and the transaction amount set of the banking system comprises:
acquiring performance parameters of system performance, wherein the performance parameters comprise the quantity of concurrent transactions processed by a banking system, the highest threshold percentage of the quantity of concurrent transactions, the lowest threshold percentage of the quantity of concurrent transactions, the average highest threshold percentage of the quantity of concurrent transactions and the average lowest threshold percentage of the quantity of concurrent transactions;
determining that the transaction account is a hotspot account if the transaction amount of the transaction account in the next time period is greater than or equal to the product of the number of concurrently processed transactions and the highest threshold percentage of the number of concurrently processed transactions, and the mean of the set of transaction amounts is greater than the product of the number of concurrently processed transactions and the average highest threshold percentage of the number of concurrently processed transactions.
4. The method of claim 3, wherein obtaining performance parameters for system performance further comprises:
determining that the transaction account is a non-hotspot account when the transaction amount of the transaction account in the next time period is less than the product of the concurrent transaction amount and the lowest threshold percentage of the concurrent transaction amount, and the mean value of the transaction amount set is less than the product of the concurrent transaction amount and the average lowest threshold percentage of the concurrent transaction amount, and the transaction account is in a hotspot account list;
the transaction account is deleted from the hotspot account list.
5. The method of claim 1, wherein before entering the transaction account and the transaction characteristic data corresponding to the transaction account in the current time period into the transaction amount prediction model, further comprising:
acquiring a historical transaction account, transaction characteristic data corresponding to the historical transaction account and transaction amount of the historical transaction account in a next time period in a banking system as sample data, and constructing a training set and a testing set, wherein the historical transaction data comprises the transaction account and the historical transaction characteristic data corresponding to the transaction account;
training a neural network model by using a training set to obtain a transaction amount prediction model;
and testing the transaction amount prediction model by using the test set.
6. The method of claim 5, wherein the historical transaction account, the transaction characteristic data corresponding to the historical transaction account, and the transaction amount in the next time period of the historical transaction account in the banking system are obtained as sample data, and the training set and the test set are constructed, including:
desensitizing the historical transaction account, transaction characteristic data corresponding to the historical transaction account and transaction amount in the next time period of the historical transaction account to be used as sample data;
dividing sample data according to a sliding window algorithm by taking a preset fixed time length as the size of a sliding window to obtain a plurality of sample data sets;
and constructing a training set and a test set according to a plurality of sample data sets.
7. The method of claim 5, further comprising:
updating sample data by using a transaction account in a hot spot account list of the current day, transaction characteristic data corresponding to the transaction account and a transaction amount corresponding to the transaction account in each time period of the current day;
and updating the training set and the test set by using the updated sample data, and continuing to train and test the transaction amount prediction model.
8. The method of claim 1, wherein the transaction characteristic data includes a product type, a transaction channel, and a transaction lending direction.
9. A hotspot account identification device, comprising:
the prediction module is used for inputting the transaction account in the current time period and the transaction characteristic data corresponding to the transaction account into a transaction amount prediction model to obtain the transaction amount of the transaction account in the next time period, wherein the current time period is a time period from a preset starting time to the current time, and the transaction amount prediction model is obtained by training a neural network model according to the historical transaction account in a banking system, the transaction characteristic data corresponding to the historical transaction account and the transaction amount in the next time period of the historical transaction account;
the judging module is used for judging whether the transaction account is a hot account or not according to the transaction amount of the transaction account in the next time period and the system performance of the banking system;
and the adding module is used for adding the transaction account into the hot account list when the transaction account is the hot account.
10. The apparatus of claim 9, wherein the determining module is specifically configured to:
acquiring historical transaction amounts of a transaction account in a current time period and a plurality of time periods before the current time period;
constructing a transaction amount set according to the transaction amount of the transaction account in the next time period and the historical transaction amount;
and judging whether the transaction account is a hot account or not according to the system performance and the transaction amount set of the banking system.
11. The apparatus of claim 10, wherein the determining module is further specifically configured to:
acquiring performance parameters of system performance, wherein the performance parameters comprise the quantity of concurrent transactions processed by a banking system, the highest threshold percentage of the quantity of concurrent transactions, the lowest threshold percentage of the quantity of concurrent transactions, the average highest threshold percentage of the quantity of concurrent transactions and the average lowest threshold percentage of the quantity of concurrent transactions;
determining that the transaction account is a hotspot account if the transaction amount of the transaction account in the next time period is greater than or equal to the product of the number of concurrently processed transactions and the highest threshold percentage of the number of concurrently processed transactions, and the mean of the set of transaction amounts is greater than the product of the number of concurrently processed transactions and the average highest threshold percentage of the number of concurrently processed transactions.
12. The apparatus of claim 11, wherein the determining module is further specifically configured to:
determining that the transaction account is a non-hotspot account when the transaction amount of the transaction account in the next time period is less than the product of the concurrent transaction amount and the lowest threshold percentage of the concurrent transaction amount, and the mean value of the transaction amount set is less than the product of the concurrent transaction amount and the average lowest threshold percentage of the concurrent transaction amount, and the transaction account is in a hotspot account list;
the transaction account is deleted from the hotspot account list.
13. The apparatus of claim 9, further comprising a model training and testing module to, prior to entering the transaction account and transaction characteristic data corresponding to the transaction account for the current time period into the transaction amount prediction model in the prediction module:
acquiring a historical transaction account, transaction characteristic data corresponding to the historical transaction account and transaction amount of the historical transaction account in a next time period in a banking system as sample data, and constructing a training set and a testing set, wherein the historical transaction data comprises the transaction account and the historical transaction characteristic data corresponding to the transaction account;
training a neural network model by using a training set to obtain a transaction amount prediction model;
and testing the transaction amount prediction model by using the test set.
14. The apparatus of claim 13, wherein the model training and testing module is further configured to:
desensitizing the historical transaction account, transaction characteristic data corresponding to the historical transaction account and transaction amount in the next time period of the historical transaction account to be used as sample data;
dividing sample data according to a sliding window algorithm by taking a preset fixed time length as the size of a sliding window to obtain a plurality of sample data sets;
and constructing a training set and a test set according to a plurality of sample data sets.
15. The apparatus of claim 13, further comprising a model optimization module, specifically configured to:
updating sample data by using a transaction account in a hot spot account list of the current day, transaction characteristic data corresponding to the transaction account and a transaction amount corresponding to the transaction account in each time period of the current day;
and updating the training set and the test set by using the updated sample data, and continuing to train and test the transaction amount prediction model.
16. The apparatus of claim 9, wherein the transaction characteristic data includes a product type, a transaction channel, and a transaction lending direction.
17. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any one of claims 1 to 8 when executing the computer program.
18. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program for executing the method of any one of claims 1 to 8.
CN202111286675.3A 2021-11-02 2021-11-02 Hot account identification method and device Pending CN113962323A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111286675.3A CN113962323A (en) 2021-11-02 2021-11-02 Hot account identification method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111286675.3A CN113962323A (en) 2021-11-02 2021-11-02 Hot account identification method and device

Publications (1)

Publication Number Publication Date
CN113962323A true CN113962323A (en) 2022-01-21

Family

ID=79468846

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111286675.3A Pending CN113962323A (en) 2021-11-02 2021-11-02 Hot account identification method and device

Country Status (1)

Country Link
CN (1) CN113962323A (en)

Similar Documents

Publication Publication Date Title
CN111798312B (en) Financial transaction system anomaly identification method based on isolated forest algorithm
CN113837596B (en) Fault determination method and device, electronic equipment and storage medium
CN111796957B (en) Transaction abnormal root cause analysis method and system based on application log
CN109063433B (en) False user identification method and device and readable storage medium
CN110310114A (en) Object classification method, device, server and storage medium
CN111967521B (en) Cross-border active user identification method and device
CN112884569A (en) Credit assessment model training method, device and equipment
CN113268403A (en) Time series analysis and prediction method, device, equipment and storage medium
CN112529319A (en) Grading method and device based on multi-dimensional features, computer equipment and storage medium
CN115730947A (en) Bank customer loss prediction method and device
CN116739742A (en) Monitoring method, device, equipment and storage medium of credit wind control model
CN112783508B (en) File compiling method, device, equipment and storage medium
CN112819069B (en) Event grading method and device
CN113971495A (en) Daytime batch processing method and device
CN110458383B (en) Method and device for realizing demand processing servitization, computer equipment and storage medium
CN114938339B (en) Data processing method and related device
CN113962323A (en) Hot account identification method and device
CN115099928A (en) Method and device for identifying lost customers
CN114386980A (en) Fraud identification method and device based on time sequence data driving
CN111429257B (en) Transaction monitoring method and device
CN115185649A (en) Resource scheduling method, device, equipment and storage medium
CN112685610A (en) False registration account identification method and related device
CN111311086A (en) Capacity monitoring method and device and computer readable storage medium
CN115758135B (en) Track traffic signal system function demand tracing method and device and electronic equipment
CN113743532B (en) Abnormality detection method, abnormality detection device, abnormality detection apparatus, and computer storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination