Transaction feature map generation method and apparatus

Abstract

This invention discloses a method and apparatus for generating transaction feature maps, relating to the field of financial risk analysis and identification technology. One specific implementation of the method includes: determining multiple early warning indicators corresponding to a financial transaction; merging the parsed text corresponding to each early warning indicator to generate a summary text; extracting multiple feature words of the financial transaction from the summary text; determining the correlation between each feature word and the financial transaction; and generating a feature map corresponding to the financial transaction based on each feature word and its corresponding correlation. This implementation displays transaction features in a visual feature map format, making the feature map more intuitive and specific, which is beneficial for relevant personnel to make correct analysis and decisions regarding financial transactions.

Description

Technical Field

[0001] This invention relates to the field of financial risk analysis and identification technology, and in particular to a method and apparatus for generating transaction feature maps. Background Technology

[0002] To establish a fair and healthy financial order, it is necessary to monitor financial transactions to reduce the adverse effects of illicit transactions on the financial market. Typically, multiple warning indicators are identified for suspicious transactions and displayed in a list. Relevant personnel use this list to analyze and assess suspicious transactions. However, because there may be multiple warning indicators, and their names are complex and difficult to understand, the information available in the indicator list is limited, making it of little reference value and hindering relevant personnel from making accurate analytical decisions. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide a method and apparatus for generating transaction feature maps, which display transaction features in a visual manner. The feature maps are more intuitive and specific, which is conducive to relevant personnel making correct analysis and decisions.

[0004] In a first aspect, embodiments of the present invention provide a method for generating a transaction feature map, comprising:

[0005] Identify multiple early warning indicators corresponding to financial transactions;

[0006] Merge the parsed texts corresponding to each of the aforementioned warning indicators to generate a summary text;

[0007] Extract multiple feature words of the financial transactions from the aggregated text;

[0008] Determine the correlation between each of the aforementioned feature words and the aforementioned financial transaction;

[0009] Based on each of the aforementioned feature words and their corresponding correlation degrees, a feature map corresponding to the financial transaction is generated.

[0010] Optionally, the feature map is displayed in the form of a word cloud;

[0011] The step of generating a feature map corresponding to the financial transaction based on each of the feature words and their corresponding correlation degrees includes:

[0012] Based on the correlation degree corresponding to each feature word, the display information of each feature word is determined, and the display information includes: size, color, position and drawing shape;

[0013] Based on the display information of each feature word, a text label corresponding to each feature word is generated, and the text label is used to display the feature word;

[0014] Combine the text tags corresponding to each of the aforementioned feature words to generate a feature word cloud corresponding to the financial transaction.

[0015] Optionally, determining the display information of each feature word based on its relevance includes:

[0016] Obtain configuration information, which includes: the mapping relationship between the correlation degree range and the display information;

[0017] Determine the target correlation range of the feature word in the configuration information;

[0018] Based on the mapping relationship, determine the target display information corresponding to the target relevance interval;

[0019] The target display information is determined as the display information of the feature word.

[0020] Optionally, the extraction of multiple feature words of the financial transaction from the aggregated text includes:

[0021] The summarized text is preprocessed;

[0022] The preprocessed summary text is segmented to obtain multiple candidate words;

[0023] Based on the frequency of each candidate word in the aggregated text, a plurality of feature words are determined from the plurality of candidate words.

[0024] Optionally, determining the correlation between each of the feature words and the financial transaction includes:

[0025] The TF-IDF algorithm is used to determine the correlation between each feature word and the financial transaction.

[0026] Optionally, the determination of multiple early warning indicators corresponding to financial transactions includes:

[0027] Obtain the transaction information corresponding to the financial transaction;

[0028] Identify at least one early warning model corresponding to the financial transaction;

[0029] For each of the aforementioned early warning models, the transaction information is input into the early warning model to obtain the output of the early warning model;

[0030] By combining the outputs of each of the aforementioned early warning models, multiple early warning indicators corresponding to the financial transaction are generated.

[0031] Optionally, the output of the early warning model includes: model indicators and indicator scores;

[0032] The combined outputs of each of the aforementioned early warning models generate multiple early warning indicators corresponding to the financial transaction, including:

[0033] By combining the model indicators output by each of the aforementioned early warning models, multiple early warning indicators corresponding to the financial transaction are generated.

[0034] Optionally, determining the correlation between each of the feature words and the financial transaction includes:

[0035] The model weights of each feature word are determined based on the model indicators and indicator scores output by each of the early warning models.

[0036] Based on the frequency of the feature words and the model weights, the correlation between the feature words and the financial transactions is determined, where the frequency is the number of times the feature words appear in the summary text.

[0037] Secondly, embodiments of the present invention provide an apparatus for generating a transaction feature map, comprising:

[0038] The indicator determination module is used to determine multiple early warning indicators corresponding to financial transactions;

[0039] The text generation module is used to merge the parsed texts corresponding to each of the aforementioned warning indicators to generate a summary text;

[0040] The feature word extraction module is used to extract multiple feature words of the financial transaction from the summarized text;

[0041] The correlation determination module is used to determine the correlation between each of the aforementioned feature words and the financial transaction;

[0042] The feature map generation module is used to generate a feature map corresponding to the financial transaction based on each of the feature words and their corresponding correlation.

[0043] Optionally, the feature map is displayed in the form of a word cloud;

[0044] The feature map generation module is specifically used for:

[0045] Based on the correlation degree corresponding to each feature word, the display information of each feature word is determined, and the display information includes: size, color, position and drawing shape;

[0046] Based on the display information of each feature word, a text label corresponding to each feature word is generated, and the text label is used to display the feature word;

[0047] Combine the text tags corresponding to each of the aforementioned feature words to generate a feature word cloud corresponding to the financial transaction.

[0048] Optionally, the feature map generation module is further configured to:

[0049] Obtain configuration information, which includes: the mapping relationship between the correlation degree range and the display information;

[0050] Determine the target correlation range of the feature word in the configuration information;

[0051] Based on the mapping relationship, determine the target display information corresponding to the target relevance interval;

[0052] The target display information is determined as the display information of the feature word.

[0053] Optionally, the feature word extraction module is specifically used for:

[0054] The summarized text is preprocessed;

[0055] The preprocessed summary text is segmented to obtain multiple candidate words;

[0056] Based on the frequency of each candidate word in the aggregated text, a plurality of feature words are determined from the plurality of candidate words.

[0057] Thirdly, embodiments of the present invention provide an electronic device, including:

[0058] One or more processors;

[0059] Storage device for storing one or more programs.

[0060] When the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any of the above embodiments.

[0061] Fourthly, embodiments of the present invention provide a computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the methods described in any of the above embodiments.

[0062] Fifthly, embodiments of the present invention provide a computer program product, including a computer program that, when executed by a processor, implements the methods described in any of the above embodiments.

[0063] One embodiment of the above invention has the following advantages or beneficial effects: It merges the parsed text corresponding to various early warning indicators of financial transactions to generate a summary text; it extracts multiple feature words of financial transactions from the summary text; and it generates a feature map corresponding to the financial transaction based on the correlation between each feature word and the financial transaction. The solution of this invention can automatically and effectively extract feature words of financial transactions and display transaction features in a visual feature map. The feature map can reflect multiple feature words and their correlation, making it more intuitive and specific, and facilitating correct analysis and decision-making by relevant personnel.

[0064] The further effects of the aforementioned unconventional alternative methods will be explained below in conjunction with specific implementation methods. Attached Figure Description

[0065] The accompanying drawings are provided to better understand the invention and are not intended to unduly limit the scope of the invention. Wherein:

[0066] Figure 1 This is a flowchart illustrating a method for generating a transaction feature map according to the first embodiment of the present invention;

[0067] Figure 2 This is a flowchart illustrating a method for generating a transaction feature map according to a second embodiment of the present invention;

[0068] Figure 3 This is a flowchart illustrating a method for generating a transaction feature map according to the third embodiment of the present invention;

[0069] Figure 4 This is a schematic diagram of the structure of a transaction feature map generation device provided in an embodiment of the present invention;

[0070] Figure 5 This is a schematic diagram of the structure of a computer system suitable for implementing terminal devices or servers of the present invention. Detailed Implementation

[0071] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of the present invention, including various details to aid understanding. These details should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0072] It should be noted that the collection, analysis, use, transmission, and storage of user personal information involved in the technical solution of this invention all comply with relevant laws and regulations, are used for legitimate and reasonable purposes, and are not shared, disclosed, or sold outside of these legitimate uses, and are subject to supervision and management by regulatory authorities. Necessary measures should be taken to prevent unauthorized access to such personal information data, ensure that personnel authorized to access personal information data comply with relevant laws and regulations, and ensure the security of user personal information. Once this user personal information data is no longer needed, the risk should be minimized by restricting or even prohibiting data collection and / or deleting the data.

[0073] Figure 1 This is a flowchart illustrating a method for generating a transaction feature map according to the first embodiment of the present invention, as shown below. Figure 1 As shown, the method includes:

[0074] Step 101: Identify multiple early warning indicators corresponding to financial transactions.

[0075] Early warning indicators can be set according to specific needs. If a financial transaction meets the risk conditions of certain early warning indicators, then the financial transaction is considered suspicious and is highly likely to be an illegal transaction.

[0076] Step 102: Merge the parsed texts corresponding to each early warning indicator to generate a summary text.

[0077] All early warning indicators have been compiled with relatively accurate Chinese analytical text based on their characteristics. Alternatively, the indicator name, description, etc., can be selected as the analytical text for the corresponding early warning indicator.

[0078] Step 103: Extract multiple feature words of financial transactions from the aggregated text.

[0079] The summary text can be preprocessed first; then the preprocessed summary text can be segmented to obtain multiple candidate words; based on the frequency of each candidate word in the summary text, multiple feature words can be determined from the multiple candidate words.

[0080] Preprocessing may include: removing punctuation, removing stop words, and correcting typos. The jieba algorithm can be used to segment the preprocessed text, yielding multiple candidate words. Candidate words can be nouns or verbs that have concrete meaning.

[0081] There are many ways to determine feature words. For example, all nouns in the summary text can be used as feature words. Alternatively, you can first determine the frequency of each candidate word in the summary text, and then select a preset number of candidate words with the highest frequency as feature words. Or, you can select multiple candidate words with frequencies greater than a frequency threshold as feature words.

[0082] Step 104: Determine the correlation between each feature word and financial transactions.

[0083] There are many ways to determine the correlation between each feature term and financial transactions. For example, the correlation can be determined by the frequency of each feature term's appearance in the aggregated text. The more frequently a feature term appears in the aggregated text, the stronger its correlation with financial transactions. Algorithms such as TF-IDF can also be used to determine the correlation between each feature term and financial transactions.

[0084] Step 105: Generate a feature map corresponding to the financial transaction based on each feature word and its corresponding correlation.

[0085] The feature map should show each feature word and its corresponding correlation. Based on each feature word and its corresponding correlation, word clouds, pie charts, bar charts, etc., of the feature words corresponding to the financial transaction can be generated, and the generated word clouds, pie charts, bar charts, etc., can be used as the feature map corresponding to the financial transaction.

[0086] In this embodiment of the invention, the parsed text corresponding to each early warning indicator of a financial transaction is merged to generate a summary text; multiple feature words of the financial transaction are extracted from the summary text; and a feature map corresponding to the financial transaction is generated based on the correlation between each feature word and the financial transaction. The solution of this embodiment can automatically and effectively extract feature words of financial transactions and display transaction features using a visual feature map. The feature map can reflect multiple feature words and their correlation, making it more intuitive and specific, and facilitating correct analysis and decision-making by relevant personnel.

[0087] Figure 2 This is a flowchart illustrating a method for generating a transaction feature map according to a second embodiment of the present invention, as shown below. Figure 2 As shown, the method includes:

[0088] Step 201: Identify multiple early warning indicators corresponding to financial transactions.

[0089] Step 202: Merge the parsed texts corresponding to each early warning indicator to generate a summary text.

[0090] Step 203: Extract multiple feature words of financial transactions from the aggregated text.

[0091] Step 204: Determine the correlation between each feature word and financial transactions.

[0092] Step 205: Determine the display information for each feature word based on its correlation with the feature word. The display information includes: size, color, position, and drawing shape.

[0093] The display information for each feature word can be determined as follows: obtain configuration information, which includes the mapping relationship between the relevance interval and the display information; determine the target relevance interval of the feature word in the configuration information; determine the target display information corresponding to the target relevance interval based on the mapping relationship; and determine the target display information as the display information for the feature word.

[0094] Users can configure the information as needed. To achieve a better display effect for the feature word cloud, the higher the relevance of a feature word, the larger its size, the brighter its color, and the more centrally located it will be. Conversely, the lower the relevance of a feature word, the smaller its size, the dimmer its color, and the more peripheral its position will be.

[0095] Step 206: Based on the display information of each feature word, generate a text label corresponding to each feature word. The text label is used to display the feature word.

[0096] Step 207: Combine the text labels corresponding to each feature word to generate a feature word cloud corresponding to financial transactions.

[0097] Feature word clouds, through a visual approach, highlight highly relevant feature words within an image, making them more visually appealing to users. In this embodiment of the invention, a feature word cloud is created based on the correlation between each feature word and financial transactions, presenting the feature words of financial transactions in a vivid and concrete manner. By observing the feature word cloud, users can identify the risks and key points of financial transactions, enabling timely risk management and mitigating the adverse effects of illegal transactions.

[0098] Figure 3 This is a flowchart illustrating a method for generating a transaction feature map according to the third embodiment of the present invention, as shown below. Figure 3 As shown, the method includes:

[0099] Step 301: Obtain transaction information corresponding to the financial transaction; determine at least one early warning model corresponding to the financial transaction.

[0100] Early warning models are used to assess the potential risks of financial transactions. Different early warning models can be set according to different transaction types. First, determine the transaction type corresponding to the financial transaction, and then determine at least one early warning model corresponding to the transaction type.

[0101] Step 302: For each early warning model, input the transaction information into the early warning model to obtain the output of the early warning model.

[0102] The transaction information is input into each early warning model, and the output of each model is obtained. The output of the early warning model may include early warning indicators. If the output of an early warning model is empty, the target transaction does not meet the risk conditions corresponding to that early warning model, and the output of that model does not need to be considered.

[0103] Step 303: Combine the outputs of each early warning model to generate multiple early warning indicators corresponding to financial transactions.

[0104] Step 304: Merge the parsing texts corresponding to each early warning indicator to generate a summary text.

[0105] Step 305: Extract multiple feature words of financial transactions from the aggregated text.

[0106] Step 306: Determine the correlation between each feature word and financial transactions.

[0107] Step 307: Generate a feature map corresponding to the financial transaction based on each feature word and its corresponding correlation.

[0108] In this embodiment of the invention, multiple early warning models corresponding to financial transactions are set up, and the outputs of each early warning model are combined to generate multiple early warning indicators corresponding to financial transactions. The early warning indicators come from multiple early warning models, enabling them to more comprehensively and richly reflect the characteristics of financial transactions, resulting in a better display effect for the final generated feature map.

[0109] In one embodiment of the present invention, the output of the early warning model includes: model indicators and indicator scores; combining the outputs of each early warning model to generate multiple early warning indicators corresponding to financial transactions includes: combining the model indicators output by each early warning model to generate multiple early warning indicators corresponding to financial transactions.

[0110] Determine the correlation between each feature word and financial transactions, including: determining the model weight of each feature word based on the model indicators and indicator scores output by each early warning model; and determining the correlation between the feature word and financial transactions based on the frequency of the feature word and the model weight, where frequency is the number of times the feature word appears in the summary text.

[0111] If the output of the early warning model includes model indicators and their scores, the model weights of each feature word can be determined based on these indicators. Specifically, if a feature word originates from a single model indicator, its corresponding score is used as the model weight. If a feature word originates from multiple model indicators, the statistical value of the scores across these indicators is used as its model weight; this statistical value can be the mean, sum, maximum, or minimum value, etc.

[0112] The higher the frequency and model weight of a feature word, the greater its correlation with financial transactions. Weights can be set for both frequency and model weight, and a weighted sum of these weights can be calculated to represent the correlation between the feature word and financial transactions.

[0113] Alternatively, the TF-IDF algorithm can be used to calculate the importance of each feature word to a financial transaction. The importance of a feature word increases proportionally with its frequency of occurrence. Then, a weighted sum of the feature word importance and model weights is calculated, and this weighted sum is used as the correlation between the feature word and the financial transaction.

[0114] Figure 4 This is a schematic diagram of the structure of a transaction feature map generation device provided in one embodiment of the present invention, as shown below. Figure 4 As shown, the device includes:

[0115] The indicator determination module 401 is used to determine multiple early warning indicators corresponding to financial transactions.

[0116] Text generation module 402 is used to merge the parsed texts corresponding to each early warning indicator to generate a summary text;

[0117] Feature word extraction module 403 is used to extract multiple feature words of financial transactions from the summary text;

[0118] The correlation determination module 404 is used to determine the correlation between each feature word and financial transactions;

[0119] The feature map generation module 405 is used to generate feature maps corresponding to financial transactions based on each feature word and its corresponding correlation.

[0120] Optionally, the feature map is displayed in the form of a word cloud;

[0121] The feature map generation module 405 is specifically used for:

[0122] Based on the correlation of each feature word, determine the display information of each feature word, including: size, color, position and drawing shape;

[0123] Based on the display information of each feature word, a text label corresponding to each feature word is generated. The text label is used to display the feature word.

[0124] Combine the text labels corresponding to each feature word to generate a feature word cloud corresponding to financial transactions.

[0125] Optionally, the feature map generation module 405 is also used for:

[0126] Obtain configuration information, which includes the mapping relationship between the correlation interval and the display information;

[0127] Determine the relevance of the feature words within the target relevance range in the configuration information;

[0128] Based on the mapping relationship, determine the target display information corresponding to the target relevance interval;

[0129] The target display information is determined as the display information of the feature words.

[0130] Optionally, the feature word extraction module 403 is specifically used for:

[0131] Preprocess the summary text;

[0132] The preprocessed summary text is segmented to obtain multiple candidate words;

[0133] Based on the frequency of each candidate word in the aggregated text, multiple feature words are identified from the candidate words.

[0134] Optionally, the correlation determination module 404 is specifically used for:

[0135] The TF-IDF algorithm is used to determine the correlation between each feature word and financial transactions.

[0136] Optionally, the indicator determination module 401 is specifically used for:

[0137] Obtain transaction information corresponding to financial transactions;

[0138] Identify at least one early warning model corresponding to a financial transaction;

[0139] For each early warning model, transaction information is input into the early warning model to obtain its output;

[0140] By combining the outputs of various early warning models, multiple early warning indicators corresponding to financial transactions are generated.

[0141] Optionally, the output of the early warning model includes: model indicators and indicator scores;

[0142] The indicator determination module 401 is specifically used for:

[0143] By combining the model indicators output by various early warning models, multiple early warning indicators corresponding to financial transactions can be generated.

[0144] Optionally, the correlation determination module 404 is specifically used for:

[0145] Based on the model indicators and indicator scores output by each early warning model, determine the model weights of each feature word;

[0146] Based on the frequency of the feature words and the model weights, the correlation between the feature words and financial transactions is determined. The frequency is the number of times the feature word appears in the summary text.

[0147] This invention provides an electronic device, comprising:

[0148] One or more processors;

[0149] Storage device for storing one or more programs.

[0150] When one or more programs are executed by one or more processors, the one or more processors implement the methods of any of the above embodiments.

[0151] This invention provides a computer program product, including a computer program that, when executed by a processor, implements the enterprise risk assessment method of this invention.

[0152] The following is for reference. Figure 5 It shows a schematic diagram of the structure of a computer system 500 suitable for implementing a terminal device of the present invention. Figure 5 The terminal device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0153] like Figure 5 As shown, the computer system 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 502 or programs loaded from storage section 508 into random access memory (RAM) 503. The RAM 503 also stores various programs and data required for the operation of the system 500. The CPU 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0154] The following components are connected to I / O interface 505: an input section 506 including a keyboard, mouse, etc.; an output section 507 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to I / O interface 505 as needed. A removable medium 511, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 510 as needed so that computer programs read from it can be installed into storage section 508 as needed.

[0155] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 509, and / or installed from removable medium 511. When the computer program is executed by central processing unit (CPU) 501, it performs the functions defined above in the system of this invention.

[0156] It should be noted that the computer-readable medium shown in this invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0157] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0158] The modules described in the embodiments of the present invention can be implemented in software or hardware. The described modules can also be housed in a processor, and for example, can be described as: an indicator determination module, a text generation module, a feature word extraction module, a correlation determination module, and a feature map generation module. The names of these modules do not necessarily limit the module itself; for example, the indicator determination module can also be described as "a module for determining multiple early warning indicators corresponding to financial transactions".

[0159] In another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiments; or it may exist independently and not assembled into the device. The computer-readable medium carries one or more programs, which, when executed by the device, cause the device to include:

[0160] Identify multiple early warning indicators corresponding to financial transactions;

[0161] Merge the parsed texts corresponding to each of the aforementioned warning indicators to generate a summary text;

[0162] Extract multiple feature words of the financial transactions from the aggregated text;

[0163] Determine the correlation between each of the aforementioned feature words and the aforementioned financial transaction;

[0164] Based on each of the aforementioned feature words and their corresponding correlation degrees, a feature map corresponding to the financial transaction is generated.

[0165] According to the technical solution of this invention, the parsed text corresponding to each early warning indicator of a financial transaction is merged to generate a summary text; multiple feature words of the financial transaction are extracted from the summary text; and a feature map corresponding to the financial transaction is generated based on the correlation between each feature word and the financial transaction. The solution of this invention can automatically and effectively extract feature words of financial transactions and display transaction features using a visual feature map. The feature map can reflect multiple feature words and their correlation, making it more intuitive and specific, and facilitating correct analysis and decision-making by relevant personnel.

[0166] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for generating a transaction feature map, characterized in that, include: Determine multiple early warning indicators corresponding to financial transactions; among them, determine the transaction type corresponding to the financial transaction, determine at least one early warning model corresponding to the financial transaction based on the transaction type, and obtain multiple early warning indicators using at least one early warning model; Merge the parsed texts corresponding to each of the aforementioned warning indicators to generate a summary text; Multiple feature words of the financial transaction are extracted from the aggregated text; wherein, if the feature word comes from a single early warning indicator, the indicator score corresponding to the early warning indicator is used as the model weight of the feature word, and if the feature word comes from multiple early warning indicators, the statistical value of the indicator scores corresponding to the multiple early warning indicators is used as the model weight of the feature word. Determining the correlation between each feature word and the financial transaction includes: determining the model weight of each feature word based on the model indicators and indicator scores output by each early warning model; and determining the correlation between the feature word and the financial transaction based on the frequency of the feature word and the model weight, wherein the frequency is the number of times the feature word appears in the summary text. Based on each of the aforementioned feature words and their corresponding correlation degrees, a feature map corresponding to the financial transaction is generated; The feature map is displayed in the form of a word cloud; The step of generating a feature map corresponding to the financial transaction based on each of the aforementioned feature words and their corresponding correlation degrees includes: determining the display information of each of the aforementioned feature words based on their correlation degrees, the display information including size, color, position, and drawing shape; generating a text label corresponding to each of the aforementioned feature words based on their display information, the text label being used to display the feature word; combining the text labels corresponding to each of the aforementioned feature words to generate a feature word cloud corresponding to the financial transaction; the feature word cloud, through a view, highlights feature words with high correlation degrees in the image to identify risk points and key points, and performs risk processing.

2. The method according to claim 1, characterized in that, The step of determining the display information of each feature word based on the relevance of each feature word includes: Obtain configuration information, which includes: the mapping relationship between the correlation degree range and the display information; Determine the target correlation range of the feature word in the configuration information; Based on the mapping relationship, determine the target display information corresponding to the target relevance interval; The target display information is determined as the display information of the feature word.

3. The method according to claim 1, characterized in that, The extraction of multiple feature words of the financial transaction from the aggregated text includes: The summarized text is preprocessed; The preprocessed summary text is segmented to obtain multiple candidate words; Based on the frequency of each candidate word in the aggregated text, a plurality of feature words are determined from the plurality of candidate words.

4. The method according to claim 1, characterized in that, Determining the correlation between each of the aforementioned feature words and the financial transaction includes: The TF-IDF algorithm is used to determine the correlation between each feature word and the financial transaction.

5. The method according to claim 1, characterized in that, The multiple early warning indicators for determining financial transactions include: Obtain the transaction information corresponding to the financial transaction; Identify at least one early warning model corresponding to the financial transaction; For each of the aforementioned early warning models, the transaction information is input into the early warning model to obtain the output of the early warning model; By combining the outputs of each of the aforementioned early warning models, multiple early warning indicators corresponding to the financial transaction are generated.

6. The method according to claim 5, characterized in that, The output of the early warning model includes: model indicators and indicator scores; The combined outputs of each of the aforementioned early warning models generate multiple early warning indicators corresponding to the financial transaction, including: By combining the model indicators output by each of the aforementioned early warning models, multiple early warning indicators corresponding to the financial transaction are generated.

7. An apparatus for generating a transaction feature map, characterized in that, include: The indicator determination module is used to determine multiple early warning indicators corresponding to financial transactions; The text generation module is used to merge the parsed texts corresponding to each of the aforementioned warning indicators to generate a summary text; wherein, the transaction type corresponding to the financial transaction is determined, at least one warning model corresponding to the financial transaction is determined according to the transaction type, and multiple warning indicators are obtained using at least one warning model. The feature word extraction module is used to extract multiple feature words of the financial transaction from the summary text; wherein, if the feature word comes from a single early warning indicator, the indicator score corresponding to the early warning indicator is used as the model weight of the feature word; if the feature word comes from multiple early warning indicators, the statistical value of the indicator scores corresponding to the multiple early warning indicators is used as the model weight of the feature word. The correlation determination module is used to determine the correlation between each of the feature words and the financial transaction, including: determining the model weight of each feature word based on the model indicators and indicator scores output by each of the early warning models; and determining the correlation between the feature word and the financial transaction based on the frequency of the feature word and the model weight, wherein the frequency is the number of times the feature word appears in the summary text. The feature map generation module is used to generate a feature map corresponding to the financial transaction based on each of the feature words and their corresponding correlation. The feature map is displayed in the form of a word cloud; The feature map generation module is specifically used for: determining the display information of each feature word based on the correlation degree of each feature word, the display information including: size, color, position and drawing shape; generating a text label corresponding to each feature word based on the display information of each feature word, the text label being used to display the feature word; combining the text labels corresponding to each feature word to generate a feature word cloud corresponding to the financial transaction; the feature word cloud, through a view, highlights feature words with high correlation in the image, identifies risk points and key points, and performs risk processing.

8. The apparatus according to claim 7, characterized in that, The feature map generation module is also used for: Obtain configuration information, which includes: the mapping relationship between the correlation degree range and the display information; Determine the target correlation range of the feature word in the configuration information; Based on the mapping relationship, determine the target display information corresponding to the target relevance interval; The target display information is determined as the display information of the feature word.

9. The apparatus according to claim 7, characterized in that, The feature word extraction module is specifically used for: The summarized text is preprocessed; The preprocessed summary text is segmented to obtain multiple candidate words; Based on the frequency of each candidate word in the aggregated text, a plurality of feature words are determined from the plurality of candidate words.

10. An electronic device, characterized in that, include: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-6.

11. A computer-readable medium having a computer program stored thereon, characterized in that... When the program is executed by the processor, it implements the method as described in any one of claims 1-6.

12. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-6.

Images