Project demand document analysis method and device, medium and program product

Abstract

The embodiment of the invention discloses a project requirement document analysis method and device, a medium and a program product. The method comprises the steps of obtaining a to-be-analyzed project demand document, and performing content region division according to contents in the project demand document to obtain at least one category of document content regions; according to the category of each document content area, determining a corresponding content identification model, and performing content identification on the document content area by adopting the content identification model to obtain corresponding text information; and performing semantic analysis according to the text information corresponding to each document content area to obtain key semantics of each document content area. According to the method, the document content is subjected to category partitioning, and then content recognition under different models is performed on each document content region, so that the accuracy of content recognition can be improved, the reliability of document analysis can be improved, and omission of information analysis can be avoided.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method, device, medium and program product for parsing a project requirement document. Background Art

[0002] When adding new services or updating services, software development is often required. Software development often involves numerous requirements documents that describe and specify software functionality, performance, and interface requirements. Developers need to analyze project requirements documents to understand user needs, clarify development goals, standardize development processes, and ensure that the developed product meets user requirements.

[0003] In existing technology, project managers typically rely on manual analysis to read, analyze, and summarize requirements documents to understand the development needs of various business departments and coordinate the management of specific project content. Manual analysis wastes human resources and relies on individual human skills, resulting in inconsistent quality and prone to errors, such as missing information. Summary of the Invention

[0004] The present invention provides a method, device, medium and program product for parsing a project requirement document, so as to improve the accuracy of content recognition and the reliability of document parsing.

[0005] According to one aspect of the present invention, a method for parsing a project requirements document is provided, the method comprising:

[0006] Obtaining a project requirement document to be parsed, dividing the content area according to the content in the project requirement document, and obtaining at least one category of document content area;

[0007] Determining a corresponding content recognition model according to the category of each document content area, and performing content recognition on the document content area using the content recognition model to obtain corresponding text information;

[0008] Semantic analysis is performed based on the text information corresponding to each of the document content areas to obtain key semantics of each of the document content areas.

[0009] According to another aspect of the present invention, a device for parsing a project requirements document is provided, the device comprising:

[0010] A content area division module is used to obtain a project requirement document to be parsed, and divide the content area according to the content in the project requirement document to obtain at least one category of document content area;

[0011] a content recognition module, configured to determine a corresponding content recognition model according to the category of each document content area, and perform content recognition on the document content area using the content recognition model to obtain corresponding text information;

[0012] The semantic analysis module is used to perform semantic analysis based on the text information corresponding to each of the document content areas to obtain the key semantics of each of the document content areas.

[0013] According to another aspect of the present invention, an electronic device is provided, comprising:

[0014] at least one processor; and

[0015] a memory communicatively connected to the at least one processor; wherein,

[0016] 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 method for parsing a project requirement document described in any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for parsing a project requirement document according to any embodiment of the present invention when executed.

[0018] According to another aspect of the present invention, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the method for parsing a project requirement document according to any embodiment of the present invention.

[0019] The technical solution of the embodiment of the present invention obtains the project requirement document to be parsed, divides the content area according to the content in the project requirement document, and obtains at least one category of document content area; determines the corresponding content recognition model according to the category of each document content area, and uses the content recognition model to perform content recognition on the document content area to obtain corresponding text information; performs semantic analysis based on the text information corresponding to each document content area to obtain the key semantics of each document content area, thereby solving the technical problems of document parsing. By partitioning the document content into categories and then performing content recognition under different models for each document content area, the accuracy of content recognition can be improved, the reliability of document parsing can be improved, and omissions in information parsing can be avoided.

[0020] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 This is a flowchart of a method for parsing a project requirement document provided according to the first embodiment of the present invention;

[0023] Figure 2 This is a flowchart of a method for parsing a project requirement document according to a second embodiment of the present invention;

[0024] Figure 3 This is a schematic diagram of the structure of a device for parsing a project requirement document provided in a third embodiment of the present invention;

[0025] Figure 4 The present invention is a schematic diagram of the structure of an electronic device for implementing the method for parsing a project requirement document according to an embodiment of the present invention. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0027] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] Example 1

[0029] Figure 1This is a flowchart of a method for parsing a project requirement document according to the first embodiment of the present invention. This embodiment is applicable to the case of efficiently parsing project requirements. The method can be executed by a parsing device for a project requirement document. The parsing device for the project requirement document can be implemented in the form of hardware and / or software. The parsing device for the project requirement document can be configured in electronic devices such as mobile phones, tablet computers (PADs), and computers. Figure 1 As shown, the method includes:

[0030] Step 110: Obtain a project requirement document to be parsed, divide the content area according to the content in the project requirement document, and obtain at least one category of document content area.

[0031] The project requirements document may be a document that describes and specifies requirements for the project product's functions, performance, user interface, and other aspects. The project requirements document may be in a variety of formats. For example, the project requirements document may be in a Word document, a PDF document, or a PPT document, etc., although this is not specifically limited in the present embodiment.

[0032] A project requirement document may include a variety of content areas. For example, a project requirement document may include a document text area, a document table area, and a document image area. In an embodiment of the present invention, the content area can be divided according to the content in the project requirement document to obtain document content areas of different categories. The content area division of the project requirement document can be achieved based on the attributes of different categories of content in the document. For example, training samples can be formed by collecting and annotating different categories of content. A neural network model is trained based on the training samples to obtain a document classification model for content area division. Thus, the content area in the document is divided by the document classification model to obtain at least one category of document content area.

[0033] By dividing the project requirement document into different categories of content areas, different categories of content areas can be processed differently, thereby improving the accuracy of project requirement document parsing.

[0034] Optionally, content areas are divided according to the content in the project requirements document to obtain at least one category of document content areas, including: content areas are divided according to color information, pixel information, spacing information and arrangement information of the content in the project requirements document to obtain document text areas, document table areas, and document image areas.

[0035] When dividing the content area of the project requirement document, the pixel information, color information, gap information between text distribution, and content arrangement information in the project requirement document can be extracted. Then, the content area can be divided based on the color information, pixel information, gap information, and arrangement information. For example, in the project requirement document, the frame arrangement area can be used as the document table area. Among them, the frame arrangement area is the area where the proportion of blank pixels is higher than the preset proportion threshold. In the project requirement document, the area where the pixel color density is greater than the preset density threshold can be used as the document image area. In the project requirement document, the area with a single color and uniform distribution of blank pixels can be used as the document text area. Among them, the uniform distribution of blank pixels means that the blank space between words is equal.

[0036] Step 120: Determine a corresponding content recognition model according to the category of each document content area, and use the content recognition model to perform content recognition on the document content area to obtain corresponding text information.

[0037] Different content recognition models can be used to identify different document content regions to improve the accuracy of content recognition. The content recognition models can be generated through model training based on the corresponding document content regions, allowing for separate content recognition for different document content regions, ensuring the reliability of content recognition.

[0038] Optionally, the corresponding content recognition model is determined according to the category of each document content area, including: when the category of the document content area is a document text area, determining that the corresponding content recognition model is a recurrent neural network model; when the category of the document content area is a document table area, determining that the corresponding content recognition model is a convolutional neural network and a transformer model; when the category of the document content area is a document image area, determining that the corresponding content recognition model is an image neural network model.

[0039] In an embodiment of the present invention, a recurrent neural network (RNN) model is used to identify the content of a text region, and the correlation information between the continuous information input before and after can be considered, thereby achieving accurate extraction of the text.

[0040] Using Convolutional Neural Networks (CNN) to identify table regions can improve processing efficiency. When performing table region recognition, combined with the Transformer model, it can perform equidistant information processing on each input, accurately identifying data in the table that has no temporal or spatial order, ensuring the independence of the table data and thus accurately identifying the table region.

[0041] Using a Generative Neural Network (GNN) model to identify image regions can accurately identify text within images. RNN, CNN, Transformer, and GNN models can all be trained to adjust model parameters to improve recognition accuracy.

[0042] Optionally, a content recognition model is used to perform content recognition on the document content area to obtain corresponding text information, including: when the category of the document content area is a document table area, a convolutional neural network is used to perform table structure recognition to obtain cells filled with data; and a transformer model is used to perform content recognition on cells filled with data to obtain text information of the document table area.

[0043] Specifically, CNN can identify the table structure within a table area, specifically cells populated with data and formatted with identifiers such as the font, size, and color of the data. For cells populated with data, a Transformer model can be used to independently identify the cell content, obtaining the text information within the document's table area. By combining CNN with the Transformer model, accurate table data extraction can be achieved, ensuring both accuracy and reliability.

[0044] Step 130: Perform semantic analysis based on the text information corresponding to each document content area to obtain the key semantics of each document content area.

[0045] In an embodiment of the present invention, natural language processing technology can be used to perform semantic analysis on the text information corresponding to each document content area to obtain the corresponding key semantics. Project managers can intuitively understand the relevant content in the project requirements document through the key semantics, thereby accelerating project development.

[0046] Optionally, in order to ensure that project managers can have a comprehensive and intuitive understanding of project requirements, the document content areas can be merged and then key semantics extracted. Specifically, semantic analysis is performed based on the text information corresponding to each document content area to obtain the key semantics of each document content area, including: performing a first keyword match on the text information of the document table area with the text information of the document text area; when the first keyword match is successful, the document table area is attached to the document text area to form the text of the attached table; otherwise, the document table area is determined to be an independent table; the text information of the document image area is performed on the text information of the document text area with the text information of the document text area; when the second keyword match is successful, the document image area is attached to the document text area to form the text of the attached image; otherwise, the document image area is determined to be an independent image; semantic analysis is performed on independent tables, independent images, text of attached tables, text of attached images, and independent text without attached tables or images to obtain corresponding key semantics.

[0047] Among them, the first keyword matching can be to determine whether the table area is attached to the text area. For example, if "Table 1" appears in the text area, the table area corresponding to "Table 1" can be used as an attached table of the text area. The second keyword matching can be to determine whether the image area is an attached picture of the text area. For example, if " Figure 1 ”, you can change the Figure 1 "The corresponding image area is used as an attached image of the text area. By matching the first keyword and the second keyword, the text of the attached table, the text of the attached image, the independent text without an attached table or image, the independent table, and the independent image can be obtained.

[0048] After keyword matching and merging the document text area, document table area and document image area, semantic analysis is performed on the text of the attached table, the text of the attached image, the independent text without attached table or image, the independent table, and the independent image. This can ensure the comprehensiveness of the document analysis and enable project managers to easily understand the project requirement documents.

[0049] The technical solution of this embodiment obtains the project requirement document to be parsed, divides the content area according to the content in the project requirement document, and obtains at least one category of document content area; determines the corresponding content recognition model according to the category of each document content area, and uses the content recognition model to perform content recognition on the document content area to obtain corresponding text information; performs semantic analysis based on the text information corresponding to each document content area to obtain the key semantics of each document content area, thereby solving the technical problems of document parsing. By partitioning the document content into categories and then performing content recognition under different models for each document content area, the accuracy of content recognition can be improved, the reliability of document parsing can be improved, and omissions in information parsing can be avoided.

[0050] Example 2

[0051] Figure 2 This is a flowchart of a method for parsing a project requirement document according to the second embodiment of the present invention. This embodiment is a further addition to the above technical solution. The technical solution in this embodiment can be combined with each optional solution in one or more of the above embodiments. Figure 2 As shown, the method includes:

[0052] Step 210: Obtain the project requirement document to be parsed, divide the content area according to the content in the project requirement document, and obtain at least one category of document content area.

[0053] Optionally, content areas are divided according to the content in the project requirements document to obtain at least one category of document content areas, including: content areas are divided according to color information, pixel information, spacing information and arrangement information of the content in the project requirements document to obtain document text areas, document table areas, and document image areas.

[0054] Step 220: Determine a corresponding content recognition model according to the category of each document content area, and use the content recognition model to perform content recognition on the document content area to obtain corresponding text information.

[0055] Optionally, the corresponding content recognition model is determined according to the category of each document content area, including: when the category of the document content area is a document text area, determining that the corresponding content recognition model is a recurrent neural network model; when the category of the document content area is a document table area, determining that the corresponding content recognition model is a convolutional neural network and a transformer model; when the category of the document content area is a document image area, determining that the corresponding content recognition model is an image neural network model.

[0056] Optionally, a content recognition model is used to perform content recognition on the document content area to obtain corresponding text information, including: when the category of the document content area is a document table area, a convolutional neural network is used to perform table structure recognition to obtain cells filled with data; and a transformer model is used to perform content recognition on cells filled with data to obtain text information of the document table area.

[0057] Step 230: Perform semantic analysis based on the text information corresponding to each document content area to obtain the key semantics of each document content area.

[0058] Optionally, semantic analysis is performed based on the text information corresponding to each document content area to obtain the key semantics of each document content area, including: performing a first keyword matching on the text information of the document table area with the text information of the document text area; when the first keyword matching is successful, the document table area is attached to the document text area to form the text of the attached table; otherwise, the document table area is determined to be an independent table; a second keyword matching is performed on the text information of the document image area with the text information of the document text area; when the second keyword matching is successful, the document image area is attached to the document text area to form the text of the attached image; otherwise, the document image area is determined to be an independent image; and semantic analysis is performed on independent tables, independent images, text of attached tables, text of attached images, and independent text without attached tables or images to obtain corresponding key semantics.

[0059] Step 240: Perform transaction association based on the key semantics of the content areas of the multiple project requirement documents to determine whether the project transactions in the project requirement documents are associated.

[0060] In project management, multi-department collaboration is common. Project requirements documents from different departments may differ. Therefore, it is necessary to aggregate and organize multiple project requirements documents. In embodiments of the present invention, transaction association can be determined for multiple project requirements documents. For example, if key semantic matching determines that multiple project requirements documents are about the same transaction, it can be determined that the project transactions in the multiple project requirements documents are associated.

[0061] Step 250: When the first project requirement document and the second project requirement document are associated with target project transactions, determine whether the target project transactions conflict.

[0062] When different project requirement documents have different requirements for the same project transaction, it can be determined that a transaction conflict exists.

[0063] Optionally, when the first project requirement document and the second project requirement document are associated with target project transactions, determining whether the target project transactions conflict includes: obtaining a first data condition for the target project transaction in the first project requirement document, and obtaining a second data condition for the target project transaction in the second project requirement document; when the first data condition is inconsistent with the second data condition, determining that there is a data condition conflict in the target project transaction; determining a first processing flow for the target project transaction in the first project requirement document, and determining a second processing flow for the target project transaction in the second project requirement document; when there is a coupling relationship between the first processing flow and the second processing flow, determining that there is a processing flow conflict in the target project transaction; when the first processing flow and the second processing flow access the same processing resources, determining that there is a resource access conflict in the target project transaction.

[0064] For example, in the first project requirements document, the first data condition for the target project transaction states that the transaction duration for transaction type A1 must be controlled within 5 milliseconds. In the second project requirements document, the second data condition for the target project transaction states that the transaction duration for transaction type A1 must be controlled within 10 milliseconds. In this case, the first and second data conditions are inconsistent, i.e., contradictory, and a data condition conflict can be determined for the target project transaction.

[0065] For example, a first process flow for a target project transaction can be drawn based on a first project requirements document. A second process flow for the target project transaction can be drawn based on a second project requirements document. If the first process flow relies on the second process flow as a prerequisite, and the second process flow relies on the first process flow as a prerequisite, indicating a coupling relationship between the first and second processes, it can be determined that a process conflict exists for the target project transaction.

[0066] For another example, if both the first processing flow and the second processing flow need to access processing resource B, then the resource access time conflict problem in the first processing flow and the second processing flow needs to be resolved, and it can be determined that there is a resource access conflict in the target project transaction.

[0067] By performing various types of transaction conflict detection on multiple project requirement documents, project managers can analyze multiple project requirement documents and accurately capture problems that exist when project requirement documents are aggregated, so as to overcome problems and smoothly advance projects.

[0068] Step 260: When there is a target project transaction conflict, first transaction data related to the target project in the first project requirement document is sent to a second transaction processing party corresponding to the second project requirement document.

[0069] Step 270: Send the second transaction data associated with the target project in the second project requirement document to the first transaction processing party corresponding to the first project requirement document.

[0070] By sending conflicting data to the conflicting parties in project transactions, all parties can share the conflicting points and negotiate to reach a consensus. The first transaction processing party can review the first transaction data against the second transaction data to determine if there are any errors or if data changes are necessary. The second transaction processing party can review the second transaction data against the first transaction data to determine if there are any errors or if data changes are necessary. By using the conflicting party's data as a basis for data verification, conflicts can be resolved quickly and efficiently, ensuring rapid project progress.

[0071] When providing project managers with the parsed results of a project requirements document, key semantics can be provided for each independent table, independent image, text attached to a table, text attached to an image, and independent text without an attached table or image. Transaction conflict information can also be provided. When providing the parsed results of a project requirements document, key semantics and transaction conflict information can be annotated using different methods to help project managers focus on them and avoid information loss and errors. Transaction conflict information can include conflicting data conditions, conflict resolution procedures, and conflicting access processing resources.

[0072] The technical solution of the embodiment of the present invention is to obtain a project requirement document to be parsed, divide the content area according to the content in the project requirement document, and obtain at least one category of document content area; determine the corresponding content recognition model according to the category of each document content area, and use the content recognition model to perform content recognition on the document content area to obtain corresponding text information; perform semantic parsing according to the text information corresponding to each document content area to obtain the key semantics of each document content area; perform transaction association according to the key semantics of each document content area of multiple project requirement documents to determine whether the project transactions in each project requirement document are associated; when there is a target project transaction association between the first project requirement document and the second project requirement document, determine whether the target project transaction conflicts; when the target project transaction conflicts , sending the first transaction data related to the target project in the first project requirement document to the second transaction processing party corresponding to the second project requirement document; sending the second transaction data associated with the target project in the second project requirement document to the first transaction processing party corresponding to the first project requirement document, solving the technical problem of document parsing, by categorizing the document content and then performing content recognition under different models for each document content area, the accuracy of content recognition can be improved, the reliability of document parsing can be improved, and omissions in information parsing can be avoided; on the basis of document parsing, transaction relevance and conflict judgment are performed on multiple project requirement documents, which can timely determine the conflict events in the project requirement documents, making it easier for project managers to focus on and resolve transaction conflict problems to quickly advance project completion.

[0073] Example 3

[0074] Figure 3 Schematic diagram of a device for parsing project requirements documents according to the third embodiment of the present invention. Figure 3 As shown, the device includes: a content area division module 310, a content identification module 320 and a semantic analysis module 330. Among them:

[0075] The content area division module 310 is used to obtain a project requirement document to be parsed, and divide the content area according to the content in the project requirement document to obtain at least one category of document content area;

[0076] The content recognition module 320 is used to determine the corresponding content recognition model according to the category of each document content area, and use the content recognition model to perform content recognition on the document content area to obtain the corresponding text information;

[0077] The semantic analysis module 330 is used to perform semantic analysis based on the text information corresponding to each document content area to obtain the key semantics of each document content area.

[0078] Optionally, the content area division module 310 includes:

[0079] The content area division unit is used to divide the content area according to the color information, pixel information, spacing information and arrangement information of the content in the project requirement document to obtain the document text area, document table area and document image area.

[0080] Optionally, the content identification module 320 includes:

[0081] A first content recognition model determination unit is configured to determine, when the category of the document content region is a document text region, that the corresponding content recognition model is a recurrent neural network model;

[0082] a second content recognition model determination unit, configured to determine, when the category of the document content area is a document table area, that the corresponding content recognition model is a convolutional neural network and a transformer model;

[0083] The third content recognition model determination unit is used to determine that the corresponding content recognition model is an image neural network model when the category of the document content area is a document image area.

[0084] Optionally, the content identification module 320 includes:

[0085] A table structure recognition unit is used to use a convolutional neural network to perform table structure recognition when the category of the document content area is a document table area, and obtain cells filled with data;

[0086] The cell content recognition unit is used to use a transformer model to perform content recognition on cells filled with data to obtain text information in the document table area.

[0087] Optionally, the semantic parsing module 330 includes:

[0088] A first keyword matching unit, configured to perform first keyword matching between text information in the document table area and text information in the document text area;

[0089] a table appending unit, configured to append the document table area to the document text area to form text of an attached table when the first keyword is successfully matched; otherwise, determine that the document table area is an independent table;

[0090] A second keyword matching unit, configured to perform second keyword matching between the text information of the document image region and the text information of the document text region;

[0091] an image appending unit, configured to append the document image region to the document text region to form text of an attached image when the second keyword is successfully matched; otherwise, determine that the document image region is an independent image;

[0092] The semantic parsing unit is used to perform semantic parsing on independent tables, independent images, text attached to tables, text attached to images, and independent text without attached tables or images to obtain corresponding key semantics.

[0093] Optionally, the device further includes:

[0094] A transaction association module is used to perform semantic analysis based on the text information corresponding to each document content area to obtain the key semantics of each document content area, and then perform transaction association based on the key semantics of each document content area of multiple project requirement documents to determine whether the project transactions in each project requirement document are associated;

[0095] A transaction conflict detection module is used to determine whether a target project transaction conflicts when the first project requirement document and the second project requirement document are associated with the target project transaction;

[0096] A first transaction data sending module, configured to send the first transaction data related to the target project in the first project requirement document to the second transaction processing party corresponding to the second project requirement document when there is a target project transaction conflict;

[0097] The second transaction data sending module is configured to send the second transaction data associated with the target project in the second project requirement document to the first transaction processing party corresponding to the first project requirement document.

[0098] Optional transaction conflict detection module, including:

[0099] a data condition acquisition unit, configured to acquire a first data condition of a target project transaction from a first project requirement document, and acquire a second data condition of a target project transaction from a second project requirement document;

[0100] a data condition conflict detection unit, configured to determine that a data condition conflict exists in a target project transaction when a first data condition conflicts with a second data condition;

[0101] a processing flow obtaining unit, configured to determine a first processing flow of a target project transaction in a first project requirement document, and to determine a second processing flow of the target project transaction in a second project requirement document;

[0102] a processing flow conflict detection unit, configured to determine whether a processing flow conflict exists in a target project transaction when a first processing flow and a second processing flow are coupled;

[0103] The resource access conflict detection unit is used to determine whether a resource access conflict exists in a target project transaction when the first processing flow and the second processing flow access the same processing resource.

[0104] The project requirement document parsing device provided in the embodiment of the present invention can execute the project requirement document parsing method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0105] Example 4

[0106] Figure 4 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0107] like Figure 4As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11. The memory stores a computer program that can be executed by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 10 can also be stored in the RAM 13. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0108] Multiple components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0109] Processor 11 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors that run machine learning model algorithms, a digital signal processor (DSP), and any other suitable processor, controller, microcontroller, etc. Processor 11 executes the various methods and processes described above, such as the method for parsing a project requirements document.

[0110] In some embodiments, the method for parsing a project requirements document can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method for parsing a project requirements document described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to execute the method for parsing a project requirements document in any other appropriate manner (e.g., via firmware).

[0111] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0112] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0113] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0114] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0115] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0116] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0117] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0118] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A method for parsing a project requirement document, characterized in that: include: Obtaining a project requirement document to be parsed, dividing the content area according to the content in the project requirement document, and obtaining at least one category of document content area; Determining a corresponding content recognition model according to the category of each document content area, and performing content recognition on the document content area using the content recognition model to obtain corresponding text information; Semantic analysis is performed based on the text information corresponding to each of the document content areas to obtain key semantics of each of the document content areas.

2. The method according to claim 1, characterized in that Divide the content area according to the content of the project requirements document to obtain at least one category of document content area, including: The content area is divided according to the color information, pixel information, gap information and arrangement information of the content in the project requirement document to obtain a document text area, a document table area and a document image area.

3. The method according to claim 2, characterized in that Determining a corresponding content recognition model according to the category of each document content area includes: When the category of the document content area is a document text area, determining that the corresponding content recognition model is a recurrent neural network model; When the category of the document content area is a document table area, determining that the corresponding content recognition model is a convolutional neural network and a transformer model; When the category of the document content area is a document image area, it is determined that the corresponding content recognition model is an image neural network model.

4. The method according to claim 3, characterized in that The content recognition model is used to perform content recognition on the document content area to obtain corresponding text information, including: When the category of the document content area is a document table area, a convolutional neural network is used to identify the table structure to obtain cells filled with data; The transformer model is used to perform content recognition on cells filled with data to obtain text information in the document table area.

5. The method according to claim 2, characterized in that Performing semantic analysis based on the text information corresponding to each document content area to obtain key semantics of each document content area includes: Performing a first keyword matching on the text information of the document table area and the text information of the document text area; When the first keyword is successfully matched, the document table area is attached to the document text area to form text of an attached table; otherwise, the document table area is determined to be an independent table; Performing second keyword matching on the text information of the document image area and the text information of the document text area; When the second keyword is successfully matched, the document image area is attached to the document text area to form a text of the attached image; otherwise, the document image area is determined to be an independent image; Semantic analysis is performed on independent tables, independent images, text attached to tables, text attached to images, and independent text without attached tables or images to obtain corresponding key semantics.

6. The method according to claim 1, characterized in that After performing semantic analysis on the text information corresponding to each of the document content areas to obtain the key semantics of each of the document content areas, the method further includes: Performing transaction association based on key semantics of the document content areas of the plurality of project requirement documents to determine whether project transactions in the project requirement documents are associated; When the first project requirement document and the second project requirement document are associated with target project transactions, determining whether the target project transactions conflict; When the target project transactions conflict, first transaction data related to the target project in the first project requirement document is sent to a second transaction processing party corresponding to the second project requirement document; The second transaction data associated with the target project in the second project requirement document is sent to the first transaction processing party corresponding to the first project requirement document.

7. The method according to claim 6, characterized in that When the first project requirement document and the second project requirement document are associated with target project transactions, determining whether the target project transactions conflict includes: Acquire a first data condition of the target project transaction in a first project requirement document, and acquire a second data condition of the target project transaction in a second project requirement document; When the first data condition is inconsistent with the second data condition, determining that a data condition conflict exists in the target project transaction; Determining a first processing flow of the target project transaction in a first project requirements document, and determining a second processing flow of the target project transaction in a second project requirements document; When the first processing flow and the second processing flow are coupled, determining that a processing flow conflict exists in the target project transaction; When the first processing flow and the second processing flow access the same processing resource, it is determined that a resource access conflict exists in the target project transaction.

8. 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. The computer program is executed by the at least one processor so that the at least one processor can execute the method for parsing a project requirement document according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for parsing a project requirement document according to any one of claims 1 to 7 when executed.

10. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the method for parsing a project requirement document according to any one of claims 1 to 7.

Images