Business intervention method, apparatus, device, storage medium and program product

Abstract

The application relates to a business intervention method and device, equipment, a storage medium and a program product, and relates to the field of artificial intelligence. The method comprises the following steps: obtaining target transaction data containing a target transaction code used for identifying a target transaction corresponding to the target transaction data; then, querying a preset database according to the target transaction code; if the target transaction code is found in the preset database, obtaining an intervention measure corresponding to the target transaction code from the preset database; finally, performing intervention processing on the transaction corresponding to the target transaction data according to the intervention measure. The preset database is constructed according to transaction error information and stores a plurality of corresponding relationships between affected transaction codes of transactions that are affected by day cut delay and intervention measures. The method can automatically intervene in the business affected by day cut delay.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a business intervention method, apparatus, device, storage medium, and program product. Background Technology

[0002] In a bank system based on a distributed platform, operations that were originally centered on a single mainframe are now deployed in a distributed manner using a mainframe + platform architecture. During year-end closing, due to high mainframe load, there are unique production scenarios where mainframe day-to-day switching is delayed while platform day-to-day switching occurs normally. This inconsistency between mainframe and platform day-to-day switching times leads to date discrepancies between the mainframe and platform. In this situation, when business transactions involve cross-mainframe and cross-platform processing and require date consistency verification, the date inconsistency between the mainframe and platform can cause transaction errors.

[0003] Currently, in order to clarify the scope of business impact caused by the delayed daily switch of the mainframe, manual simulation verification is required in advance, which is inefficient. Summary of the Invention

[0004] Therefore, it is necessary to provide a business intervention method, device, equipment, storage medium, and program product that can automatically intervene in response to the above-mentioned technical problems.

[0005] Firstly, this application provides a business intervention method. The method includes:

[0006] The process involves: acquiring target transaction data, which includes target transaction codes to identify the transactions corresponding to the target transaction data; querying a pre-defined database based on the target transaction codes, which stores multiple sets of correspondences between affected transaction codes and intervention measures, whereby the affected transaction codes are those of transactions that have encountered errors due to daily cut-off delays; and constructing the pre-defined database based on transaction error information. If the target transaction code is found in the pre-defined database, the corresponding intervention measures are retrieved from the database, and the transactions corresponding to the target transaction data are then intervened based on the retrieved intervention measures.

[0007] In one embodiment, the process of constructing the preset database includes: obtaining an initial database containing multiple initial correspondences of affected transaction codes, intervention measures, and transaction types; obtaining target transaction error information triggered by the daily cut-off delay, the target transaction error information including a target erroneous transaction code, a target transaction description, and the original error information of the target business; determining the target transaction type corresponding to the target transaction error information based on the target transaction error information and the initial database, and determining the target intervention measures based on the target transaction type; and updating the initial database using the target erroneous transaction code, the target transaction type, and the target intervention measures determined based on the target transaction type to obtain the preset database.

[0008] In one embodiment, the process of constructing the initial database includes: obtaining a set of transaction error information; extracting the error transaction code, transaction description, and original business error information of each transaction error information in the set of transaction error information through business field tags to obtain a first aggregation result; filtering out the error transaction code, transaction description, and original business error information of transaction error information triggered by the daily cut-off delay from the first aggregation result to obtain a second aggregation result; and determining the transaction type and intervention measures corresponding to the transaction error information in the second aggregation result to construct the initial database.

[0009] In one embodiment, determining the target transaction type corresponding to the target transaction error information based on the target transaction error information and the initial database includes: performing word segmentation on the target transaction description and the original error information of the target business to obtain the word segmentation result, and merging the word segmentation result with the target error transaction code to obtain the features of the target transaction error information; matching the features with features of multiple initial correspondences in the initial database to obtain a first matching result; matching the word segmentation result with a pre-set transaction type thesaurus to obtain a second matching result; and determining the target transaction type based on the first matching result and the second matching result.

[0010] In one embodiment, the feature is matched with features of multiple initial correspondences in the initial database, including: calculating the Euclidean distance between the feature and the features of the multiple initial correspondences in the initial database to obtain the transaction types in the n initial correspondences with the highest similarity; and calculating the proportion of each of the m transaction types in the n transaction types to obtain a set S. i S i This is the first matching result, where i takes values ​​greater than or equal to 1 and less than or equal to m.

[0011] In one embodiment, the word segmentation result is matched with a pre-defined transaction type lexicon to obtain a second matching result. This includes: matching the word segmentation result with m pre-defined transaction type lexicons, and obtaining a first hit score set P of the word segmentation result in the m transaction type lexicons based on the matching results and corresponding scores. i P i This is the second matching result.

[0012] In one embodiment, determining the target transaction type based on the first matching result and the second matching result includes: setting S i With P i Multiply the elements in the set to obtain a second hit score set of m transaction types. Take the transaction type corresponding to the maximum value in the second hit score set as the target transaction type corresponding to the target transaction error message.

[0013] In one embodiment, obtaining the target transaction error information triggered by the day-cutting delay includes: obtaining a set of newly added transaction error information; extracting the error transaction code, transaction description and original business error information of each transaction error information in the set of newly added transaction error information through business field tags to obtain a third aggregation result; and filtering out the transaction error information triggered by the day-cutting delay from the third aggregation result to obtain the target transaction error information.

[0014] Secondly, this application also provides a business apparatus. The apparatus includes:

[0015] The acquisition module is used to acquire target transaction data, which includes a target transaction code that identifies the transaction corresponding to the target transaction data.

[0016] The query module is used to query the preset database based on the target transaction code. The preset database stores multiple sets of correspondences between affected transaction codes and intervention measures. The affected transaction codes are the transaction codes of transactions that have errors due to the daily cut-off delay. The preset database is constructed based on the transaction error information.

[0017] The intervention module is used to retrieve the intervention measures corresponding to the target transaction code from the preset database if the target transaction code is found, and to intervene in the transaction corresponding to the target transaction data according to the retrieved intervention measures.

[0018] In one embodiment, the device further includes a first construction module, which is used to acquire an initial database containing multiple initial correspondences of affected transaction codes, intervention measures, and transaction types; acquire target transaction error information triggered by the daily cut-off delay, the target transaction error information including a target erroneous transaction code, a target transaction description, and original error information of the target business; determine the target transaction type corresponding to the target transaction error information based on the target transaction error information and the initial database, and determine the target intervention measures based on the target transaction type; update the initial database using the target erroneous transaction code, the target transaction type, and the target intervention measures determined based on the target transaction type to obtain a preset database.

[0019] In one embodiment, the device further includes a second construction module, which is used to acquire a set of transaction error information, extract the erroneous transaction code, transaction description, and original business error information of each transaction error information in the set of transaction error information through business field tags, and obtain a first aggregation result; filter out the erroneous transaction code, transaction description, and original business error information of transaction error information triggered by daily cut-off delay from the first aggregation result, and obtain a second aggregation result; determine the transaction type and intervention measures corresponding to the transaction error information in the second aggregation result, so as to construct an initial database.

[0020] In one embodiment, the first construction module is specifically used to perform word segmentation on the target transaction description and the original error information of the target business to obtain the word segmentation result, and merge the word segmentation result with the target error transaction code to obtain the features of the target transaction error information; match the features with the features of multiple initial correspondences in the initial database to obtain the first matching result; match the word segmentation result with a pre-set transaction type thesaurus to obtain the second matching result; and determine the target transaction type based on the first matching result and the second matching result.

[0021] In one embodiment, the first construction module is specifically used to calculate the Euclidean distance between the features and the features of multiple initial correspondences in the initial database, to obtain the transaction types among the n initial correspondences with the highest similarity; and to calculate the proportion of each of the m transaction types among the n transaction types, to obtain a set S. i S i This is the first matching result, where i takes values ​​greater than or equal to 1 and less than or equal to m.

[0022] In one embodiment, the first construction module is specifically used to perform matching in a pre-defined lexicon of m transaction types based on the word segmentation results, and to obtain a first hit score set P of the word segmentation results in the lexicon of m transaction types based on the matching results and the corresponding scores. i P iThis is the second matching result.

[0023] In one embodiment, the first building module is specifically used to transfer S i With P i Multiply the elements in the set to obtain a second hit score set of m transaction types. Take the transaction type corresponding to the maximum value in the second hit score set as the target transaction type corresponding to the target transaction error message.

[0024] In one embodiment, the first construction module is specifically used to obtain a set of newly added transaction error information, extract the error transaction code, transaction description and original business error information of each transaction error information in the set of newly added transaction error information through business field tags, and obtain a third aggregation result; filter out the transaction error information triggered by the daily cut-off delay in the third aggregation result to obtain the target transaction error information.

[0025] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the business intervention method as described in any of the first aspects above.

[0026] Fourthly, this application also provides a computer-readable storage medium. This computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the business intervention method as described in any of the first aspects above.

[0027] Fifthly, this application also provides a computer program product. This computer program product includes a computer program that, when executed by a processor, implements the business intervention method described in any of the first aspects above.

[0028] The aforementioned business intervention methods, devices, equipment, storage media, and program products first acquire target transaction data containing target transaction codes that identify the transactions corresponding to the target transaction data. Then, they query a preset database based on the target transaction codes. If the target transaction code is found in the preset database, the corresponding intervention measures are retrieved from the database. Finally, intervention processing is applied to the transactions corresponding to the target transaction data according to the intervention measures. The preset database is constructed based on transaction error information and stores multiple sets of correspondences between affected transaction codes and intervention measures for transactions that have malfunctioned due to daily cut-off delays. In this way, during a user's transaction process, user transaction data is acquired, and the database containing transaction codes affected by daily cut-off delays is queried using the transaction codes within the transaction data. If a transaction code exists in the database, it indicates that the transaction is affected by the daily cut-off delay. Intervention processing is then applied to the user's transaction based on the intervention measures corresponding to the transaction codes in the database. This approach eliminates the need for manual verification of which transactions are affected when daily cut-off delays occur; the judgment and intervention are automatically performed during the user's transaction process, resulting in higher efficiency.

[0029] Furthermore, the system determines whether to intervene based on the transaction code when the user begins a transaction, avoiding errors only after the user has performed multiple steps, thus improving the user experience. Attached Figure Description

[0030] Figure 1 This is a flowchart illustrating a business intervention method in one embodiment;

[0031] Figure 2 This is a flowchart illustrating the preset database construction steps in another embodiment;

[0032] Figure 3 This is a flowchart illustrating the initial database construction steps in another embodiment;

[0033] Figure 4 This is a flowchart illustrating a method for determining the target transaction type in another embodiment;

[0034] Figure 5 This is a flowchart illustrating a method for determining the target transaction type in another embodiment;

[0035] Figure 6 This is a schematic diagram of a method for determining the target transaction type in another embodiment;

[0036] Figure 7 This is a flowchart illustrating the business intervention method in another embodiment;

[0037] Figure 8 This is a flowchart illustrating the business intervention method in another embodiment;

[0038] Figure 9 This is a structural block diagram of the business intervention device in another embodiment;

[0039] Figure 10 This is a structural block diagram of the business intervention device in another embodiment;

[0040] Figure 11 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0042] In one embodiment, such as Figure 1 As shown, a business intervention method is provided. Taking the application of this method to a terminal as an example, it can be understood that this method can also be applied to a server, and also to a system including both a terminal and a server, and is implemented through the interaction between the terminal and the server. The method includes the following steps:

[0043] Step 101: The terminal acquires the target transaction data.

[0044] The target transaction data includes a target transaction code that identifies the transaction corresponding to the target transaction data. The target transaction data is the data corresponding to the transaction the user is currently conducting, and the target transaction code is the transaction code corresponding to the transaction the user is currently conducting. The transaction code is unique and can be represented by numbers, such as 13056, 12378, etc.

[0045] Step 102: The terminal queries the preset database based on the target transaction code.

[0046] The pre-built database stores multiple sets of correspondences between affected transaction codes and intervention measures. Affected transaction codes are those of transactions that malfunctioned due to daily cut-off delays. The pre-built database is constructed based on transaction error information. Intervention measures include business-aggregated error information and intervention plans. Business-aggregated error information is updated after aggregating and unifying error information based on the impact of daily cut-off delays, and is associated with the transaction type. Intervention plans determine whether to allow the customer to proceed with the current transaction and provide a notification to the customer based on the business-aggregated error information, and are associated with the transaction type. Optionally, transaction types can be divided into three categories, represented by a dictionary: 0 - Query type that allows the customer to adjust the time interval; 1 - Query type that does not allow the customer to adjust the time interval; 2 - Maintenance type. Correspondingly, business-aggregated error information is represented by a dictionary: 0 - Currently being processed by the system, please adjust the query date to the previous day (when the transaction type value is 0); 1 - Currently being processed by the system, please query this business later (when the transaction type value is 1); 2 - Currently being processed by the system, please operate this business later (when the transaction type value is 2). The dictionary values ​​for the intervention plan include: 0 - support the transaction and update the original error information based on the business aggregation error information (when the transaction type is 0 or 1), 1 - reject the transaction and update the original error information based on the business aggregation error information (when the transaction type is 2).

[0047] Step 103: If the terminal finds that the target transaction code is located in the preset database, it retrieves the intervention measures corresponding to the target transaction code from the preset database and performs intervention processing on the transaction corresponding to the target transaction data according to the retrieved intervention measures.

[0048] If the terminal finds the target transaction code in the preset database, it means that the target transaction code is an affected transaction code. At this time, the terminal retrieves the intervention measures corresponding to the target transaction code from the preset database, and intervenes in the user's ongoing transaction according to the intervention measures, such as whether to allow the user to continue the transaction, and outputs business aggregation error message prompts.

[0049] The aforementioned business intervention method involves several steps. First, target transaction data containing target transaction codes that identify the transactions corresponding to the target transaction data is acquired. Then, a pre-defined database is queried based on the target transaction codes. If the target transaction code is found in the pre-defined database, the corresponding intervention measures are retrieved. Finally, intervention is applied to the transactions corresponding to the target transaction data according to the intervention measures. The pre-defined database is constructed based on transaction error information and stores multiple sets of correspondences between affected transaction codes and intervention measures for transactions that have malfunctioned due to daily cut-off delays. In this way, during a user's transaction, user transaction data is acquired, and the database containing transaction codes affected by daily cut-off delays is queried using the transaction codes within the transaction data. If a transaction code exists in the database, it indicates that the transaction is affected by the daily cut-off delay. Intervention is then applied to the user's transaction based on the intervention measures corresponding to the transaction codes in the database. This approach eliminates the need for manual verification of which transactions are affected when daily cut-off delays occur; the judgment and intervention are automatically performed during the user's transaction process, resulting in higher efficiency.

[0050] Furthermore, the system determines whether to intervene based on the transaction code when the user begins a transaction, avoiding errors only after the user has performed multiple steps, thus improving the user experience.

[0051] In one embodiment, the preset database is constructed based on transaction error information, such as... Figure 2 As shown, the specific construction process includes:

[0052] Step 201: The terminal obtains the initial database.

[0053] The initial database contains multiple initial mappings of affected transaction codes, intervention measures, and transaction types. Affected transaction codes are those corresponding to transactions impacted by the daily cut-off delay. Intervention measures include business aggregation error information and intervention plans. Each affected transaction code corresponds to a transaction type and an intervention measure.

[0054] Step 202: The terminal obtains the target transaction error information triggered by the daily cut-off delay.

[0055] The target transaction error information includes the target error transaction code, the target transaction description, and the target business's original error information. The target transaction error information pertains to transactions affected by daily cut-off delays, and the target error transaction code is the transaction code corresponding to the target transaction error information. The target transaction description is a description of the transaction corresponding to the target transaction error information; the transaction description can be a description of the transaction, such as a query. The target business's original error information is the original business error information for the transaction corresponding to the target transaction error information. This original business error information can be error information from before it was consolidated and unified; for example, a dictionary value like 96314025 - "Order is not within the valid time range" could be used.

[0056] Step 203: The terminal determines the target transaction type corresponding to the target transaction error information based on the target transaction error information and the initial database, and determines the target intervention measures based on the target transaction type.

[0057] The terminal matches the target transaction error information with the initial database, and automatically classifies the target transaction type corresponding to the target transaction error information through semantic analysis. Since the target intervention measures are associated with the target transaction type, the target intervention measures are then determined based on the classified target transaction type.

[0058] Step 204: The terminal updates the initial database using the target error transaction code, the target transaction type, and the target intervention measures determined based on the target transaction type to obtain the preset database.

[0059] The target error transaction code, target transaction description, and original error information of the target business, along with the automatically classified target transaction type and target intervention measures, are added to the initial database to obtain the preset database. Optionally, for subsequently added target transaction error information, the target transaction type is automatically classified again, and then the preset database is updated based on the target transaction error information, target transaction type, and target intervention measures.

[0060] In the above embodiments, by constructing a preset database, which stores multiple sets of correspondences between the affected transaction codes and intervention measures for transactions delayed by daily cut-off, when a user makes a transaction, the preset database can be directly queried based on the user's transaction code to determine whether intervention is required, thus improving efficiency.

[0061] In the embodiments of this application, the initial database construction process is as follows: Figure 3 As shown, it includes:

[0062] Step 301: The terminal obtains the transaction error information set, and extracts the error transaction code, transaction description and original business error information of each transaction error information in the transaction error information set through the business field tags to obtain the first collection result.

[0063] Optionally, the transaction error information set includes multiple transaction error messages. By using business field tags, the error transaction code, transaction description, and original business error information of the transaction error messages are extracted. Among them, the error transaction code is the transaction code of the transaction error message, which is the unique identifier of the transaction. After the multiple transaction error messages are extracted, the first aggregation result is obtained.

[0064] Step 302: The terminal filters out the erroneous transaction codes, transaction descriptions and original business error information triggered by the daily cut-off delay from the first collection result to obtain the second collection result.

[0065] The first aggregated results are then filtered according to preset rules, including rules such as 96314025 - "Order not within the valid time range," which means the original business error message indicates that the transaction was affected by the day-cut delay and triggered the transaction error message. This process filters out the erroneous transaction codes, transaction descriptions, and original business error messages from the first aggregated results that were triggered by the day-cut delay, resulting in the second aggregated results.

[0066] Step 303: The terminal determines the transaction type and intervention measures corresponding to the transaction error information in the second collection result in order to construct the initial database.

[0067] The transaction type and intervention measures corresponding to each transaction error message in the second aggregation result are determined and used as known sample data to construct an initial database. The initial database stores multiple initial correspondences between the affected transaction codes, intervention measures, and transaction types of the existing transaction error messages.

[0068] In the above embodiments, an initial database is constructed as sample data to facilitate the automatic classification of transaction types for newly added error information.

[0069] In one embodiment, the KNN (K-Nearest Neighbor) algorithm can be used for classification. If a sample's K most similar samples in the feature space mostly belong to a certain category, then the sample also belongs to that category. That is, the category of the sample to be classified is determined based on the category of its one or more nearest neighbors. Therefore, by determining the transaction types of the initial correspondences with the highest similarity to the target transaction error information from multiple initial correspondences in the initial database, the transaction type of the target transaction error information can be obtained. Figure 4 As shown, the steps for automatically classifying the target transaction type based on the target transaction error information include:

[0070] Step 401: The terminal performs word segmentation on the target transaction description and the original error information of the target business to obtain the word segmentation result, and merges the word segmentation result with the target error transaction code to obtain the features of the target transaction error information.

[0071] Optionally, the target transaction description and the original error message of the target business are combined to form an error message description for word segmentation, and the segmented words are represented using word vectors. The word vectors of all words in an error message description are concatenated to form a matrix, with each row representing a word. Assuming the error message has *a* words and the dimension of each word vector is *b*, we obtain an *a*b matrix. The word vectors of each word can be randomly initialized. If some words already have pre-trained word vectors, they can be used directly. For untrained words, their word vectors can be filled with 0 or random small positive numbers. Each error message description can be specified as having a fixed length; for example, all error message descriptions can be composed of 10 words, with padding added if necessary. The target error transaction code is used as a numerical feature. The word segmentation results are merged with the target error transaction, resulting in each target transaction error message description having 11 features, thus obtaining the features of the target transaction error message.

[0072] Step 402: The terminal performs matching processing on the features with the features of multiple initial correspondences in the initial database to obtain the first matching result.

[0073] The first matching result is the proportion of each transaction type among multiple initial correspondences that have high feature similarity to the target transaction error information. For example... Figure 5 As shown, the calculation steps for the first matching result include:

[0074] Step 501: The terminal calculates the Euclidean distance between the features and the features of multiple initial correspondences in the initial database to obtain the transaction types in the n initial correspondences with the highest similarity.

[0075] Euclidean distance reflects the similarity between the characteristics of the target transaction error information and the initial correspondence in the initial database, i.e., the characteristics of the existing error information. Among the similarity calculated with all existing error information in the initial database, the n existing error information with the highest similarity are selected. Optionally, n can be 10, and the transaction types corresponding to the n existing error information are obtained.

[0076] Step 502: The terminal calculates the percentage of each of the m transaction types out of the n transaction types, obtaining set S. i S i This is the first matching result, where i takes values ​​greater than or equal to 1 and less than or equal to m.

[0077] Calculate the percentage of each of the m transaction types out of the n transaction types obtained above. m can be 3, meaning there are 3 types of transactions. For example, the transaction types mentioned above can take values ​​of 0, 1, and 2, resulting in the first matching result S. i At this point, the value of i is 1, 2, or 3. Therefore, set S i S1, S2, and S3 represent the percentage of each of the three transaction types among the 10 existing error messages.

[0078] Step 403: The terminal performs matching processing on the word segmentation results with the pre-set transaction type thesaurus to obtain the second matching result.

[0079] Based on the word segmentation results, matching is performed in a pre-defined dictionary of m transaction types. The first hit score set P of the word segmentation results in the dictionary of m transaction types is obtained based on the matching results and corresponding scores. i P i This is the second matching result.

[0080] The system includes m pre-defined transaction type thesauruses. When m is 3, these include Transaction Type Thesaurus 1, Transaction Type Thesaurus 2, and Transaction Type Thesaurus 3, corresponding to the three transaction types mentioned above. Each transaction type thesaurus contains multiple typical keywords related to that transaction type and their corresponding scores. Based on the word segmentation results, keyword matching is performed sequentially in the three transaction type thesauruses. The first hit scores P1, P2, and P3 are calculated based on the matching results and corresponding scores. P1, P2, and P3 represent the second matching results.

[0081] Step 404: Determine the target transaction type based on the first matching result and the second matching result.

[0082] S i With P i Multiply the elements in the set to obtain a second hit score set of m transaction types. Take the transaction type corresponding to the maximum value in the second hit score set as the target transaction type corresponding to the target transaction error message.

[0083] The percentage of each transaction type is multiplied by the first hit score of the transaction type vocabulary to obtain the second hit score for each of the three transaction types. For example, the second hit score for each transaction type is represented by Q. i This means Q1 = S1 * P1, Q2 = S2 * P2, and Q3 = S3 * P3. The transaction type with the highest second-highest hit score is selected as the target transaction type corresponding to the target transaction error message. Accordingly, the target intervention measures can be determined based on the target transaction type. Optional, please refer to... Figure 6The target transaction code is used as feature 1. The target transaction description and the original error information of the target business are combined for word segmentation to obtain features 2 to 11. Based on these 11 features, the target transaction type is automatically determined, which can be 0, 1, or 2. The target intervention measures are then determined based on the target transaction type. These intervention measures include the target business's collected error information and the target intervention plan. The values ​​for the target intervention measures can be 00, 10, or 21, depending on the target transaction type.

[0084] In the above embodiments, newly added transaction error information can be automatically classified into transaction types, which saves time and manpower compared to manual judgment and recording, and is more efficient.

[0085] In one embodiment, such as Figure 7 As shown, the error information of the target transaction triggered by the day-end delay is obtained, including:

[0086] Step 701: The terminal obtains the set of newly added transaction error information, and extracts the error transaction code, transaction description and original business error information of each transaction error information in the set of newly added transaction error information through the business field tags to obtain the third aggregation result.

[0087] For the newly added set of transaction error information, the same method as above is used to extract the error transaction code, transaction description and original business error information of each transaction error information according to the business fields, and obtain the third aggregation result.

[0088] Step 702: The terminal filters out the transaction error information triggered by the daily cut-off delay from the third collection result to obtain the target transaction error information.

[0089] In the third set of results, transaction error information triggered by the daily cut-off delay is filtered out according to preset rules. The preset rules are the same as those mentioned above, including 96314025 - the order is not within the valid time range, etc. At this time, the target transaction error information is obtained.

[0090] In the embodiments of this application, please refer to Figure 8 The document illustrates a flowchart of a business intervention method provided in an embodiment of this application, which includes the following steps:

[0091] Step 801: The terminal obtains the transaction error information set, and extracts the error transaction code, transaction description and original business error information of each transaction error information in the transaction error information set through the business field tags to obtain the first collection result.

[0092] Step 802: The terminal filters out the erroneous transaction codes, transaction descriptions and original business error information triggered by the daily cut-off delay from the first collection result to obtain the second collection result.

[0093] Step 803: The terminal determines the transaction type and intervention measures corresponding to the transaction error information in the second collection result in order to construct the initial database.

[0094] Step 804: The terminal obtains the target transaction error information triggered by the daily cut-off delay.

[0095] Step 805: The terminal determines the target transaction type corresponding to the target transaction error information based on the target transaction error information and the initial database, and determines the target intervention measures based on the target transaction type.

[0096] Step 806: The terminal updates the initial database using the target error transaction code, the target transaction type, and the target intervention measures determined based on the target transaction type to obtain the preset database.

[0097] Step 807: The terminal acquires the target transaction data.

[0098] Step 808: The terminal queries the preset database based on the target transaction code.

[0099] Step 809: If the terminal finds that the target transaction code is located in the preset database, it retrieves the intervention measures corresponding to the target transaction code from the preset database and performs intervention processing on the transaction corresponding to the target transaction data according to the retrieved intervention measures.

[0100] To facilitate readers' understanding of the technical solutions provided in the embodiments of this application, the business intervention method of this application is illustrated with examples.

[0101] To retrieve business transaction error information, an example is as follows:

[0102] {Transaction Code: 13056; Transaction Description: Transaction 1; Name: Xiaoming; Return Code: 92001234; Original Error Message: Account cannot be empty}

[0103] {Transaction Code: 12156; Transaction Description: Transaction 2; Name: Xiaolan; Return Code: 98451574; Original Business Error Message: The working date sent by the terminal is inconsistent with the date sent by the host}

[0104] By extracting the business fields from the business transaction error information, a set of business transaction error information containing the transaction code, transaction description, and original business error information is obtained, as shown below:

[0105] {Transaction Code: 13056; Transaction Description: Transaction 1; Original Error Message: Account cannot be empty}

[0106] {Transaction Code: 12156; Transaction Description: Transaction 2; Original Business Error Message: The working date sent by the terminal is inconsistent with the host date}

[0107] From the above results, we filtered out the transaction error information triggered by the day-end delay to obtain the target transaction error information, as shown below:

[0108] {Transaction Code: 12156; Transaction Description: Transaction 2; Original Business Error Message: The working date sent by the terminal is inconsistent with the host date}

[0109] Based on the target transaction error information and the initial database, the target transaction type and intervention measures for the target transaction error information are determined. The initial database contains multiple initial correspondences of transaction codes, intervention measures, and transaction types affected by the daily cut-off delay. Then, a preset database is constructed using the target transaction error information and the initial database. The preset database includes transaction data affected by the daily cut-off delay, as shown in the table below:

[0110] Table 1 Preset Database Tables

[0111]

[0112] Once the preset database is built, during the user's transaction process, the system acquires the user's transaction information data and matches the transaction code with the information in the database. If the target transaction code is found in the preset database, the system retrieves the intervention measures corresponding to the target transaction code from the preset database and intervenes in the transaction corresponding to the target transaction data based on the retrieved intervention measures.

[0113] For example, if a user's transaction code is 12156 and it matches in the preset database, it means that the transaction is affected by the daily cut-off delay scenario. Intervention measures corresponding to this transaction code in the preset database will be implemented, namely, the transaction will be rejected and the error message will be updated to: "Currently processing, please operate this service later." If the user's transaction code is 11221 and it does not match in the preset database, it means that the transaction is not affected by the daily cut-off delay scenario, and no intervention plan needs to be executed.

[0114] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0115] Based on the same inventive concept, this application also provides a business intervention device for implementing the business intervention method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more business intervention device embodiments provided below can be found in the limitations of the business intervention method described above, and will not be repeated here.

[0116] In one embodiment, such as Figure 9 As shown, a business intervention device 900 is provided, including: an acquisition module 901, a query module 902, and an intervention module 903, wherein:

[0117] The acquisition module 901 is used to acquire target transaction data, which includes a target transaction code that identifies the transaction corresponding to the target transaction data.

[0118] The query module 902 is used to query a preset database based on the target transaction code. The preset database stores multiple sets of correspondences between affected transaction codes and intervention measures. The affected transaction codes are the transaction codes of transactions that have errors due to the daily cut-off delay. The preset database is constructed based on the transaction error information.

[0119] The intervention module 903 is used to retrieve the intervention measures corresponding to the target transaction code from the preset database if the target transaction code is found to be in the preset database, and to intervene in the transaction corresponding to the target transaction data according to the retrieved intervention measures.

[0120] In the embodiments of this application, such as Figure 10As shown, the device also includes a first construction module 1001, which is used to obtain an initial database containing multiple initial correspondences of affected transaction codes, intervention measures, and transaction types; obtain target transaction error information triggered by the daily cut-off delay, the target transaction error information including target error transaction code, target transaction description, and target business original error information; determine the target transaction type corresponding to the target transaction error information based on the target transaction error information and the initial database, and determine the target intervention measures based on the target transaction type; update the initial database using the target error transaction code, target transaction type, and target intervention measures determined based on the target transaction type to obtain a preset database.

[0121] In the embodiments of this application, please continue to refer to Figure 10 The device also includes a second construction module 1002, which is used to acquire a set of transaction error information, extract the error transaction code, transaction description and original business error information of each transaction error information in the set of transaction error information through business field tags, and obtain a first collection result; filter out the error transaction code, transaction description and original business error information of transaction error information triggered by daily cut-off delay from the first collection result, and obtain a second collection result; determine the transaction type and intervention measures corresponding to the transaction error information in the second collection result to construct an initial database.

[0122] In the embodiments of this application, the first construction module 1001 is specifically used to perform word segmentation processing on the target transaction description and the original error information of the target business to obtain the word segmentation result, and merge the word segmentation result with the target error transaction code to obtain the features of the target transaction error information; match the features with the features of multiple initial correspondences in the initial database to obtain the first matching result; match the word segmentation result with a pre-set transaction type thesaurus to obtain the second matching result; and determine the target transaction type based on the first matching result and the second matching result.

[0123] In the embodiments of this application, the first construction module 1001 is specifically used to calculate the Euclidean distance between the features and the features of multiple initial correspondences in the initial database, to obtain the transaction types among the n initial correspondences with the highest similarity; and to calculate the proportion of each of the m transaction types among the n transaction types, to obtain a set S. i S i This is the first matching result, where i takes values ​​greater than or equal to 1 and less than or equal to m.

[0124] In the embodiments of this application, the first construction module 1001 is specifically used to perform matching in a pre-set m transaction type word library based on the word segmentation results, and obtain the first hit score set P of the word segmentation results in the m transaction type word library based on the matching results and the corresponding scores. i P i This is the second matching result.

[0125] In embodiments of this application, the first construction module 1001 is specifically used to determine the target transaction type based on the first matching result and the second matching result, including: setting S i With P i Multiply the elements in the set to obtain a second hit score set of m transaction types. Take the transaction type corresponding to the maximum value in the second hit score set as the target transaction type corresponding to the target transaction error message.

[0126] In the embodiments of this application, the first construction module 1001 is specifically used to obtain a set of newly added transaction error information, extract the error transaction code, transaction description and original business error information of each transaction error information in the set of newly added transaction error information through business field tags, and obtain a third collection result; filter out the transaction error information triggered by the daily cut-off delay in the third collection result to obtain the target transaction error information.

[0127] Each module in the aforementioned business intervention device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can invoke and execute the operations corresponding to each module.

[0128] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 11As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a business intervention method. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0129] Those skilled in the art will understand that Figure 11 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0130] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0131] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0132] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0133] 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, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0134] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0135] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0136] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A business intervention method, characterized in that, The method, applied to a distributed banking system, includes: In the processing flow after a user initiates a transaction request but before the transaction is completed, target transaction data is obtained. The target transaction data includes a target transaction code used to identify the transaction corresponding to the target transaction data. The target transaction code is used to query a preset database, which stores multiple sets of correspondences between affected transaction codes and intervention measures. The affected transaction code is the transaction code of a transaction that malfunctions in the bank's distributed system due to a date inconsistency between the host and platform dates caused by host date-cutting delays. The preset database is constructed based on transaction error information triggered by date-cutting delays. The intervention measures include business aggregation error information and an intervention plan, the intervention plan being used to indicate whether to refuse to execute the transaction or allow its execution. If the target transaction code is found to be located in the preset database, the intervention measures corresponding to the target transaction code are retrieved from the preset database, and the transaction is intervened in accordance with the intervention measures before the transaction is completed, including: refusing to execute the transaction or allowing the execution of the transaction according to the intervention plan, and returning the business aggregation error information to the user.

2. The method according to claim 1, characterized in that, The process of constructing the preset database includes: Obtain an initial database, which contains multiple initial correspondences between the affected transaction codes, the intervention measures, and the transaction types; Obtain the target transaction error information triggered by the daily cut-off delay, wherein the target transaction error information includes the target error transaction code, the target transaction description and the original error information of the target business; Based on the target transaction error information and the initial database, the target transaction type corresponding to the target transaction error information is determined, and the target intervention measures are determined based on the target transaction type; The initial database is updated using the target erroneous transaction code, the target transaction type, and the target intervention measures determined based on the target transaction type to obtain the preset database.

3. The method according to claim 2, characterized in that, The process of constructing the initial database includes: Obtain the transaction error information set, and extract the error transaction code, transaction description and original business error information of each transaction error information in the transaction error information set through business field tags to obtain the first aggregation result; The error transaction codes, transaction descriptions, and original business error information triggered by the daily cut-off delay are filtered out from the first collection results to obtain the second collection results; The transaction type and intervention measures corresponding to the transaction error information in the second aggregation result are determined in order to construct the initial database.

4. The method according to claim 2, characterized in that, The step of determining the target transaction type corresponding to the target transaction error information based on the target transaction error information and the initial database includes: The target transaction description and the original error information of the target business are segmented into words to obtain the segmentation results. The segmentation results are then merged with the target error transaction code to obtain the features of the target transaction error information. The feature is matched with the features of multiple sets of initial correspondences in the initial database to obtain a first matching result; The word segmentation results are matched with a pre-defined transaction type thesaurus to obtain a second matching result; The target transaction type is determined based on the first matching result and the second matching result.

5. The method according to claim 4, characterized in that, The step of matching the feature with features of multiple sets of initial correspondences in the initial database includes: Calculate the Euclidean distance between the feature and the features of multiple sets of initial correspondences in the initial database to obtain the transaction types in the n initial correspondences with the highest similarity; By calculating the percentage of each of the m transaction types out of the n transaction types, we obtain the set S. i The S i Let i be the first matching result, where i is greater than or equal to 1 and less than or equal to m.

6. The method according to claim 5, characterized in that, The step of matching the word segmentation results with a pre-defined transaction type thesaurus to obtain a second matching result includes: Based on the word segmentation results, matching is performed in the pre-defined m transaction type word libraries. The first hit score set P of the word segmentation results in the m transaction type word libraries is obtained based on the matching results and corresponding scores. i The P i This is the second matching result.

7. The method according to claim 6, characterized in that, Determining the target transaction type based on the first matching result and the second matching result includes: The S i With the P i The elements in the set are multiplied together to obtain the second hit score set of the m transaction types. The transaction type corresponding to the maximum value in the second hit score set is taken as the target transaction type corresponding to the target transaction error information.

8. The method according to claim 2, characterized in that, The process of obtaining target transaction error information triggered by the day-end delay includes: Obtain the set of newly added transaction error information, and extract the error transaction code, transaction description and original business error information of each transaction error information in the set of newly added transaction error information through business field tags to obtain the third aggregation result; The transaction error information triggered by the daily cut-off delay is filtered out from the third collection result to obtain the target transaction error information.

9. A business intervention device, characterized in that, The device, applied to a distributed banking system, includes: The acquisition module is used to acquire target transaction data in the processing flow after the user initiates a transaction request and before the transaction is completed. The target transaction data includes a target transaction code used to identify the transaction corresponding to the target transaction data. The query module is used to query a preset database based on the target transaction code. The preset database stores multiple sets of correspondences between affected transaction codes and intervention measures. The affected transaction code is the transaction code of a transaction that malfunctions in the bank's distributed system due to a date inconsistency between the host and the platform caused by a host date cut-off delay. The preset database is constructed based on transaction error information triggered by the date cut-off delay. The intervention measures include business aggregation error information and an intervention plan. The intervention plan is used to indicate whether to refuse to execute the transaction or allow the execution of the transaction. An intervention module is used to retrieve the intervention measures corresponding to the target transaction code from the preset database if the target transaction code is found to be located in the preset database, and to intervene in the transaction according to the intervention measures before the transaction is completed, including: refusing to execute the transaction or allowing the execution of the transaction according to the intervention plan, and returning the business aggregation error information to the user.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

12. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

Images