Financial asset attachment classification method and device based on supply chain, and medium

Abstract

The invention relates to a data processing technology, and discloses a supply chain-based financial asset attachment classification method and device and a medium, and the method comprises the steps: obtaining a scanning text of an asset attachment, carrying out the text segmentation of the scanning text, and obtaining the segmentation data; extracting key information of the segmented data, and generating key page data of the asset attachment according to the key information; performing file type matching on the key page data according to the key information to obtain a file type; and generating an asset attachment classification result according to the file type and the key page data. According to the invention, the efficiency of asset attachment classification can be improved, and the accuracy of classification is reduced.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method, device and medium for classifying asset attachments based on supply chain finance. Background Art

[0002] In the current supply chain finance field, as the business scale continues to expand, the number of financial asset attachments involved has increased dramatically, including a variety of contract documents, transaction vouchers, valuation reports, etc.

[0003] Traditional asset attachment classification is carried out through manual review. However, when manual review is faced with a large number of files of various types, it is necessary to flip through the key information of each file one by one and record the corresponding page number, and then classify the files. This method is not only time-consuming and labor-intensive but also prone to errors. Therefore, the traditional asset attachment classification method still has the problems of low efficiency and high error rate. Summary of the invention

[0004] The present invention provides a method, device and medium for classifying asset attachments based on supply chain finance, the main purpose of which is to solve the problem of low efficiency and accuracy when classifying asset attachments.

[0005] To achieve the above purpose, the present invention provides a method for classifying supply chain financial asset attachments, comprising:

[0006] Obtaining a scanned text of the asset attachment, and performing text segmentation on the scanned text to obtain segmented data;

[0007] Extracting key information of the segmented data, and generating key page data of the asset attachment according to the key information;

[0008] Perform file type matching on the key page data according to the key information to obtain the file type;

[0009] Generate an asset attachment classification result according to the file type and the key page data.

[0010] Optionally, the performing text segmentation on the scanned text to obtain segmentation data includes:

[0011] Performing paragraph segmentation on the scanned text to obtain paragraph data;

[0012] Perform sentence boundary recognition on the paragraph data to obtain sentence data;

[0013] Structuring the sentence data to obtain format data;

[0014] The format data is cleaned to obtain segmented data.

[0015] Optionally, the performing sentence boundary recognition on the paragraph data to obtain sentence data includes:

[0016] Performing boundary symbol detection on the paragraph data to obtain symbol boundaries;

[0017] generating sentence rules according to the symbol boundaries and the paragraph data;

[0018] The paragraph data is divided using the sentence boundaries and sentence rules to obtain sentence data.

[0019] Optionally, the performing text cleaning on the format data to obtain segmented data includes:

[0020] Deleting irrelevant characters from the formatted data to obtain a cleaned file;

[0021] Performing data standardization on the cleaning file to obtain a standardized text;

[0022] Performing grammar checking on the standardized text to obtain a checked text;

[0023] Data integration is performed on the inspection text to obtain segmented data.

[0024] Optionally, extracting key information of the segmented data includes:

[0025] Performing entity recognition on the segmented data to obtain entity data;

[0026] Extracting keywords from the segmented data to obtain keyword data;

[0027] Using the segmented data, semantically analyzing the entity data and the keyword data to obtain structure data;

[0028] The structural data is converted to obtain key information.

[0029] Optionally, generating key page data of the asset attachment according to the key information includes:

[0030] Using the scanned text to divide the key information into data sets to obtain key data sets;

[0031] Using the scanned text to locate the data in the key data set to obtain location information;

[0032] Structurally sorting the data in the key data set using the scanned text and the positioning information to sort the data;

[0033] Filtering summary data from the sorted data;

[0034] The sorting data is digitally displayed using a preset template to obtain key page data.

[0035] Optionally, performing file type matching on the key page data according to the key information to obtain the file type includes:

[0036] Performing vector conversion on the key information to obtain a key vector;

[0037] Obtain a preset file data set, and calculate the conditional probability and prior probability of the key vector according to the file data set;

[0038] Calculate the file type probability of the key page data according to the conditional probability, the prior probability, the file data set and the key vector;

[0039] The key page data is classified into key pages according to the file type probability to obtain the file type.

[0040] Optionally, generating an asset attachment classification result according to the file type and the key page data includes:

[0041] Establishing a corresponding relationship between the scanned text and the key page data;

[0042] Construct a multi-level directory tree according to the file type and the key page data;

[0043] Classifying the scanned text using the file type and key page data based on the multi-level directory tree to generate an asset attachment tree;

[0044] The scanned text and key page data are classified and stored using the asset attachment tree and the corresponding relationship to obtain an asset attachment classification result.

[0045] In order to solve the above problem, the present invention further provides an electronic device, the electronic device comprising:

[0046] at least one processor; and,

[0047] a memory communicatively connected to the at least one processor; wherein,

[0048] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the above-mentioned supply chain financial asset attachment classification method.

[0049] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one computer program is stored. The at least one computer program is executed by a processor in an electronic device to implement the above-mentioned supply chain financial asset attachment classification method.

[0050] The embodiment of the present invention can achieve fine segmentation of the scanned text of the asset attachment by segmenting the text of the asset attachment, so as to improve the efficiency of subsequent classification analysis; further, the key page data of the asset attachment is generated according to the key information of the segmented data, which can achieve data dimensionality reduction. While improving the efficiency of subsequent analysis, the key information of the asset attachment is generated to facilitate the improvement of the accuracy of subsequent classification, and then the asset attachment is classified according to the key page data to achieve accurate attachment category division. Therefore, the asset attachment classification method, device and medium based on supply chain finance proposed by the present invention can solve the problem of low efficiency and accuracy when classifying asset attachments. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A schematic diagram of a flow chart of a supply chain financial asset attachment classification method provided in accordance with an embodiment of the present invention;

[0052] Figure 2 A schematic diagram of the structure of an electronic device for implementing the supply chain financial asset attachment classification method provided by an embodiment of the present invention.

[0053] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0054] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0055] The embodiment of the present application provides a method for classifying attachments of supply chain financial assets. The execution subject of the method for classifying attachments of supply chain financial assets includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for classifying attachments of supply chain financial assets can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.

[0056] Reference Figure 1 FIG. 1 is a flow chart of a method for classifying attachments of supply chain financial assets provided by an embodiment of the present invention. In this embodiment, the method for classifying attachments of supply chain financial assets includes:

[0057] S1. Obtain a scanned text of an asset attachment, and perform text segmentation on the scanned text to obtain segmented data.

[0058] In the embodiment of the present invention, the asset attachment refers to an additional document or record related to a specific asset that can provide additional information or certification documents, including invoices, receipts, asset registration forms, appraisal reports, contract agreements, insurance policies, etc.; the scanned text refers to editable and searchable electronic text data that is identified and converted from paper or image files of asset attachments through OCR technology.

[0059] In the embodiment of the present invention, the scanned text is segmented to obtain smaller paragraphs or fields, so that the segmented paragraphs and fields are easier to be accurately analyzed, thereby reducing the error rate caused by the complexity of the overall text.

[0060] In the embodiment of the present invention, the step of segmenting the scanned text to obtain segmented data includes:

[0061] Performing paragraph segmentation on the scanned text to obtain paragraph data;

[0062] Perform sentence boundary recognition on the paragraph data to obtain sentence data;

[0063] Structuring the sentence data to obtain format data;

[0064] The format data is cleaned to obtain segmented data.

[0065] In the embodiment of the present invention, the paragraph segmentation refers to searching the scanned text for continuous line breaks (e.g., two or more line breaks) as paragraph segmentation points, identifying common paragraph start patterns, such as first line indentation, and segmenting the scanned text into independent paragraphs according to the paragraph segmentation points. Each paragraph should be saved as a separate string or part of a data structure.

[0066] In the embodiment of the present invention, the step of performing sentence boundary recognition on the paragraph data to obtain sentence data includes:

[0067] Performing boundary symbol detection on the paragraph data to obtain symbol boundaries;

[0068] generating sentence rules according to the symbol boundaries and the paragraph data;

[0069] The paragraph data is divided using the sentence boundaries and sentence rules to obtain sentence data.

[0070] Specifically, by obtaining a table of commonly used punctuation marks, such as a period, a comma, etc., the positions of all symbols in the table symbols are located in the paragraph data using regular expressions, and the positions of the symbols are used as symbol boundaries.

[0071] In detail, by performing symbol rule recognition on the symbol boundary and connecting word recognition on the paragraph data, sentence rules are generated according to the recognition results; in detail, the sentence rules refer to defined rules to determine which periods actually mark the end of a sentence, such as the content within quotation marks is considered a sentence, and connecting words indicate that the sentence has not ended.

[0072] In an embodiment of the present invention, the structural segmentation refers to analyzing the text structure of the scanned text using a preset parser, for example, searching for patterns that match a specific format (such as * for list items), and constructing a data structure containing text structure information, such as a nested list or dictionary, to reflect the hierarchy and format of the original document.

[0073] In the embodiment of the present invention, the text cleaning of the format data to obtain segmented data includes:

[0074] Deleting irrelevant characters from the formatted data to obtain a cleaned file;

[0075] Performing data standardization on the cleaning file to obtain a standardized text;

[0076] Performing grammar checking on the standardized text to obtain a checked text;

[0077] Data integration is performed on the inspection text to obtain segmented data.

[0078] In detail, the irrelevant characters refer to characters such as HTML tags, special symbols (such as @, #, etc.), extra spaces and line breaks, etc. The format data is string matched using regular expressions, and the matched strings are deleted to obtain a cleaned file.

[0079] In detail, the text in the cleaning file is encoded uniformly, and the numbers or dates in the text are converted into a consistent format, for example, all numbers are expressed in a form with a thousand separator, or unified as floating point numbers with two decimal places, or the date is converted into a standard format (YYYY-MM-DD).

[0080] In detail, a grammar checking tool (such as LanguageTool or Grammarly API, etc.) is used to perform grammar checking on the standardized text. For each standardized text, an HTTP request is sent, and the standardized text is sent to the grammar checking tool as part of the request body, and finally all grammatical and spelling errors and correction plans are returned.

[0081] In the embodiment of the present invention, by segmenting the text and breaking the document into smaller units, each part of the data can be processed more accurately, thereby improving the efficiency of data processing and the probability of error.

[0082] S2. Extract key information of the segmented data, and generate key page data of the asset attachment according to the key information.

[0083] In an embodiment of the present invention, by extracting key information such as date, amount, account number, signature, etc. from the segmented data and then generating key page data based on the key information, it can be ensured that all important information is correctly captured and recorded, thereby improving the accuracy of subsequent document decisions.

[0084] In the embodiment of the present invention, the step of extracting key information of the segmented data includes:

[0085] Performing entity recognition on the segmented data to obtain entity data;

[0086] Extracting keywords from the segmented data to obtain keyword data;

[0087] Using the segmented data, semantically analyzing the entity data and the keyword data to obtain structure data;

[0088] The structural data is converted to obtain key information.

[0089] In an embodiment of the present invention, the entity data refers to specific entities identified from the segmented data, such as names of people, places, names of organizations, etc., and entity recognition refers to identifying entity sequences from the segmented data, that is, entities composed of consecutive words in the segmented data, by calculating the label probabilities of words in the segmented data and entity labels and the transition probabilities between labels (that is, the possibility of forming entities between words), and then finding the optimal entity sequence in the segmented data based on the label probabilities and transition probabilities.

[0090] In an embodiment of the present invention, keyword extraction is to extract words representing the main content from a text; by calculating the word frequency and the word frequency-inverse document frequency value in the segmented data, constructing a word co-occurrence matrix according to the word frequency and the word frequency-inverse document frequency value, calculating the importance weight of each word in the segmented data according to the co-occurrence matrix, sorting the words according to the importance weight, and screening out a preset number of words with the highest weights as keywords.

[0091] In an embodiment of the present invention, dependency syntactic analysis is performed on the entity data and the keywords to obtain a dependency relationship, and the entity data is extracted one by one from the dependency relationship to obtain structural data, for example, "In the 2023 Q1 inventory report, the income is a1 and the expenditure is a2." The entity data is "2023, annual, Q1, inventory report, income, a1, expenditure, a2", and the keyword data is finally determined to be "2023, Q1, income, a1, expenditure, a2". Dependency analysis is performed on the keyword data to obtain "2023"-"Q1", "income"-"a1", "expenditure"-"a2".

[0092] In the embodiment of the present invention, the step of generating key page data of the asset attachment according to the key information includes:

[0093] Using the scanned text to divide the key information into data sets to obtain key data sets;

[0094] Using the scanned text to locate the data in the key data set to obtain location information;

[0095] Structurally sorting the data in the key data set using the scanned text and the positioning information to sort the data;

[0096] Filtering summary data from the sorted data;

[0097] The sorting data is digitally displayed using a preset template to obtain key page data.

[0098] In the embodiment of the present invention, the key information is divided into different subsets using the scanned text, so that each document in the scanned text can generate a key data set.

[0099] In the embodiment of the present invention, the position where each keyword appears is located in the scanned text, including the line number, paragraph number, etc., and a data location index is created based on the line number and paragraph number. The location index is the location information.

[0100] In an embodiment of the present invention, the data in the key data set are sorted in a positioning order according to the positioning information in the scanned text. For example, the scanned text is "In the 2023 Q1 inventory report, the income is a1 and the expenditure is a2.", wherein the key data set is {"2023"-"Q1", "Expenditure"-"a2", "Income"-"a1"}, and the key information in the key data set is positioned and sorted using the scanned text to obtain {"2023"-"Q1", "Income"-"a1", "Expenditure"-"a2"}.

[0101] For example, if the sorted data is {"2023"-"Q1", "Income"-"a1", "Expenditure"-"a2"}, the summary data obtained is "2023Q1" and the content is "Income"-"a1", "Expenditure"-"a2".

[0102] In the embodiment of the present invention, key data is extracted and then key page data is generated from the key data. The obtained key page data can be directly applied to subsequent files, thereby reducing manual intervention and improving classification efficiency.

[0103] S3. Perform file type matching on the key page data according to the key information to obtain the file type.

[0104] In the embodiment of the present invention, by identifying the document type of key page data, the file type of the document is automatically identified, thereby improving the efficiency of document classification.

[0105] In the embodiment of the present invention, the step of performing file type matching on the key page data according to the key information to obtain the file type includes:

[0106] Performing vector conversion on the key information to obtain a key vector;

[0107] Obtain a preset file data set, and calculate the conditional probability and prior probability of the key vector according to the file data set;

[0108] Calculate the file type probability of the key page data according to the conditional probability, the prior probability, the file data set and the key vector;

[0109] The key page data is classified into key pages according to the file type probability to obtain the file type.

[0110] In the embodiment of the present invention, a dictionary containing all key information words is constructed, and for each key page information, a vector of the same length as the dictionary is created, wherein the value of each position represents the number of times the corresponding word appears in the document, thereby obtaining a key vector.

[0111] In an embodiment of the present invention, the preset file data set refers to a set of documents whose file types have been marked; the conditional probability refers to the probability that the key vector D in the file data set belongs to category C, and the conditional probability is calculated by training a classifier (such as a naive Bayes classifier, a support vector machine or a neural network); the prior probability refers to the probability that category C appears in the file data set, which is obtained by calculating the number of documents belonging to category C in the data set divided by the total number of documents.

[0112] In the embodiment of the present invention, calculating the file type probability of the key page data according to the conditional probability, the prior probability, the file data set, and the key vector includes:

[0113] The file type probability is calculated using the following formula:

[0114]

[0115] Among them, P(C|D) is the probability that the type of the key vector D is file type C, P(D|C) is the probability of the key vector D appearing under the condition that the file type is C, that is, the conditional probability, P(C) is the prior probability of file type C, and P(D) is the total probability of the key vector D appearing.

[0116] In the embodiment of the present invention, the highest file type probability corresponding to the key page is screened out, and the file type corresponding to the file type probability is used as the file type of the key page.

[0117] In the embodiment of the present invention, the enclosure and management type of the document are automatically determined to ensure that the same type of files can be processed consistently and efficiently, and the type of the file can be quickly located, saving the time of searching in the document.

[0118] S4. Generate an asset attachment classification result according to the file type and the key page data.

[0119] In the embodiment of the present invention, by generating asset attachments according to the file type and the key page data, different types of asset attachments can be stored in a suitable database, thereby achieving orderly storage, reducing unnecessary data redundancy, and achieving cost savings.

[0120] In the embodiment of the present invention, generating the asset attachment classification result according to the file type and the key page data includes:

[0121] Establishing a corresponding relationship between the scanned text and the key page data;

[0122] Construct a multi-level directory tree according to the file type and the key page data;

[0123] Classifying the scanned text using the file type and key page data based on the multi-level directory tree to generate an asset attachment tree;

[0124] The scanned text and key page data are classified and stored using the asset attachment tree and the corresponding relationship to obtain an asset attachment classification result.

[0125] In the embodiment of the present invention, the corresponding relationship refers to the mapping relationship between the key information in the scanned text and the preset key page data; in detail, by utilizing the positioning information in the key page data, each key data in the key page data can be found in the scanned text.

[0126] For example, first, create top-level nodes based on the file type, such as "Contract", "Invoice", "Financial Report", etc. Then, under each top-level node, further subdivide it based on the key page data. For example, under the "Contract" node, there are sub-nodes such as "Contract Number", "Signing Date", and "Total Amount". Each sub-node points to a specific scanned text fragment or key page data.

[0127] In the embodiment of the present invention, based on the structure of the asset attachment tree, the scanned text and key page data are classified and stored according to the hierarchical relationship, that is, files of the same type are classified together, and different key page data of the same file are subdivided into smaller categories. For example, all "signing date" information under the "contract" file will be stored centrally, realizing the orderly storage of asset attachments.

[0128] In an embodiment of the present invention, by generating asset attachment classification results based on file types and key page data, not only can a hierarchical structure of asset attachments be provided, but also a logical association between key page data and the scanned text of asset attachments can be ensured, so that data can be stored in an orderly manner, thereby shortening retrieval time and improving classification and search efficiency.

[0129] like Figure 2 , which is a schematic diagram of the structure of an electronic device for implementing a method for classifying asset attachments based on supply chain finance provided by an embodiment of the present invention.

[0130] The electronic device may include a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a supply chain financial asset attachment classification program.

[0131] In some embodiments, the processor 10 may be composed of an integrated circuit, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips. The processor 10 is the control core (ControlUnit) of the electronic device, and uses various interfaces and lines to connect various components of the entire electronic device, and executes or executes programs or modules stored in the memory 11 (for example, executing a classification program based on supply chain financial asset attachments, etc.), and calls data stored in the memory 11 to execute various functions of the electronic device and process data.

[0132] The memory 11 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (for example: SD or DX memory, etc.), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of an electronic device, such as a mobile hard disk of the electronic device. In other embodiments, the memory 11 can also be an external storage device of an electronic device, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device. Further, the memory 11 can also include both an internal storage unit of the electronic device and an external storage device. The memory 11 can not only be used to store application software and various types of data installed in the electronic device, such as code based on the supply chain financial asset attachment classification program, but also can be used to temporarily store data that has been output or is to be output.

[0133] The communication bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is configured to realize connection and communication between the memory 11 and at least one processor 10, etc.

[0134] The communication interface 13 is used for communication between the above-mentioned electronic device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a display (Display), an input unit (such as a keyboard (Keyboard)), and optionally, the user interface may also be a standard wired interface, a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode, organic light-emitting diode) touch device, etc. Among them, the display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device and to display a visual user interface.

[0135] The figure only shows an electronic device with components. Those skilled in the art will understand that the structure shown in the figure does not constitute a limitation on the electronic device, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0136] For example, although not shown, the electronic device may also include a power source (such as a battery) for supplying power to each component. Preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that the power management device can realize functions such as charging management, discharging management, and power consumption management. The power source may also include one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0137] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0138] The supply chain finance asset attachment classification program stored in the memory 11 in the electronic device is a combination of multiple instructions. When running in the processor 10, it can achieve:

[0139] Obtaining a scanned text of the asset attachment, and performing text segmentation on the scanned text to obtain segmented data;

[0140] Extracting key information of the segmented data, and generating key page data of the asset attachment according to the key information;

[0141] Perform file type matching on the key page data according to the key information to obtain the file type;

[0142] Generate an asset attachment classification result according to the file type and the key page data.

[0143] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to the description of the relevant steps in the corresponding embodiment of the accompanying drawings, which will not be repeated here.

[0144] Furthermore, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0145] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, the computer program can implement:

[0146] Obtaining a scanned text of the asset attachment, and performing text segmentation on the scanned text to obtain segmented data;

[0147] Extracting key information of the segmented data, and generating key page data of the asset attachment according to the key information;

[0148] Perform file type matching on the key page data according to the key information to obtain the file type;

[0149] Generate an asset attachment classification result according to the file type and the key page data.

[0150] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, media and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0151] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0152] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0153] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0154] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is not limited only according to the above description, and it is intended that all changes within the meaning and scope of equivalent elements within the scope of protection are included in the present invention.

[0155] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0156] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in the system can also be implemented by one unit or device through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0157] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A classification method based on supply chain financial asset attachments, characterized in that: The method comprises: Obtaining a scanned text of the asset attachment, and performing text segmentation on the scanned text to obtain segmented data; Extracting key information of the segmented data, and generating key page data of the asset attachment according to the key information; Perform file type matching on the key page data according to the key information to obtain the file type; Generate an asset attachment classification result according to the file type and the key page data.

2. The method for classifying supply chain financial asset attachments according to claim 1, characterized in that: The step of segmenting the scanned text to obtain segmented data includes: Performing paragraph segmentation on the scanned text to obtain paragraph data; Perform sentence boundary recognition on the paragraph data to obtain sentence data; Structuring the sentence data to obtain format data; The format data is cleaned to obtain segmented data.

3. The method for classifying supply chain financial asset attachments according to claim 1, characterized in that: The step of performing sentence boundary recognition on the paragraph data to obtain sentence data includes: Performing boundary symbol detection on the paragraph data to obtain symbol boundaries; generating sentence rules according to the symbol boundaries and the paragraph data; The paragraph data is divided using the sentence boundaries and sentence rules to obtain sentence data.

4. The method for classifying supply chain financial asset attachments according to claim 1, characterized in that: The step of performing text cleaning on the format data to obtain segmented data includes: Deleting irrelevant characters from the formatted data to obtain a cleaned file; Performing data standardization on the cleaning file to obtain a standardized text; Performing grammar checking on the standardized text to obtain a checked text; Data integration is performed on the inspection text to obtain segmented data.

5. The method for classifying supply chain financial asset attachments according to claim 1, characterized in that: The step of extracting key information of the segmented data includes: Performing entity recognition on the segmented data to obtain entity data; Extracting keywords from the segmented data to obtain keyword data; Using the segmented data, semantically analyzing the entity data and the keyword data to obtain structure data; The structural data is converted to obtain key information.

6. The method for classifying supply chain financial asset attachments according to claim 1, characterized in that: The step of generating key page data of the asset attachment according to the key information includes: Using the scanned text to divide the key information into data sets to obtain key data sets; Using the scanned text to locate the data in the key data set to obtain location information; Structurally sorting the data in the key data set using the scanned text and the positioning information to sort the data; Filtering summary data from the sorted data; The sorting data is digitally displayed using a preset template to obtain key page data.

7. The method for classifying supply chain financial asset attachments according to claim 1, characterized in that: The performing file type matching on the key page data according to the key information to obtain the file type includes: Performing vector conversion on the key information to obtain a key vector; Obtain a preset file data set, and calculate the conditional probability and prior probability of the key vector according to the file data set; Calculate the file type probability of the key page data according to the conditional probability, the prior probability, the file data set and the key vector; The key page data is classified into key pages according to the file type probability to obtain the file type.

8. The method for classifying supply chain financial asset attachments according to claim 1, characterized in that: The generating of the asset attachment classification result according to the file type and the key page data includes: Establishing a corresponding relationship between the scanned text and the key page data; Construct a multi-level directory tree according to the file type and the key page data; Classifying the scanned text using the file type and key page data based on the multi-level directory tree to generate an asset attachment tree; The scanned text and key page data are classified and stored using the asset attachment tree and the corresponding relationship to obtain an asset attachment classification result.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the supply chain financial asset attachment classification method as described in any one of claims 1 to 8.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for classifying asset attachments based on supply chain finance as described in any one of claims 1 to 8 is implemented.

Images