Automatic project evaluation method, device, equipment and medium

Abstract

The invention discloses an automatic project evaluation method. The method comprises the steps of obtaining a demand file and a target project file of each associated object; performing classification processing on each target project file according to a machine learning algorithm to obtain structural data and non-structural data; constructing a quantization rule base according to the demand file, and performing matching processing on the quantization rule base and the structure data to obtain quantifiable data; performing semantic similarity analysis according to the quantifiable data and the demand file, and obtaining a first sorting result of each target project file according to an analysis result; inputting the non-structural data into a pre-trained sorting model to obtain a second sorting result; and generating a log file of each associated object according to the first sorting result and the second sorting result, and obtaining an evaluation result of each associated object according to the log file. According to the automatic project evaluation method, the project evaluation efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of bidding, and in particular to an automated project evaluation method, device, equipment and medium. Background Art

[0002] Project evaluation is one of the important processes in bidding projects. It helps to ensure the legality, compliance, economy and feasibility of the project, improve project quality and competitiveness, promote fair competition and transparency, reduce project risks, and provide a scientific basis for decision-making.

[0003] In the existing technology, it is mainly the review personnel who manually identify the bidding documents, calculate the scores and rankings of the bidding suppliers, and the data processing relies entirely on manual calculations, with a relatively low degree of automated calculations. Usually, the review experts need to read and review each bidding document one by one, including a large number of text descriptions, charts, and attachments, such as registered capital, turnover and other data, and need to process a large amount of table data to conduct reviews based on the data of the bidding suppliers; in addition, the bidding review often involves multiple scoring items, such as qualifications, prices, technology, services, etc., and each scoring item may have multiple sub-items, which need to be calculated according to complex rules. This process is time-consuming and labor-intensive, and as the number of bids increases, the efficiency will drop significantly. Summary of the invention

[0004] The present invention aims to provide an automated project evaluation method, device, equipment and medium to solve the technical problem of how to improve the evaluation efficiency.

[0005] In order to solve the above technical problems, an embodiment of the present invention provides an automated project evaluation method, comprising:

[0006] Obtain the requirements file and the target project file for each associated object;

[0007] Classify and process each of the target project files according to a machine learning algorithm to obtain structured data and unstructured data;

[0008] Building a quantification rule base according to the demand file, and matching the quantification rule base with the structure data to obtain quantifiable data;

[0009] Performing semantic similarity analysis on the quantifiable data and the requirement document, and obtaining a first ranking result of each of the target project files according to the analysis result;

[0010] Inputting the unstructured data into a pre-trained sorting model to obtain a second sorting result; wherein the sorting result is configured to be trained according to the historical information data of each of the associated objects;

[0011] A log file for each of the associated objects is generated according to the first sorting result and the second sorting result, and an evaluation result for each of the associated objects is obtained according to the log file.

[0012] As one of the preferred solutions, the target project files are classified and processed according to the machine learning algorithm to obtain structured data and unstructured data, including:

[0013] Convert the target project file into text format data;

[0014] Extracting features from the text format data, and manually annotating part of the extracted feature data to obtain a training data set;

[0015] Inputting the training data set into a pre-built text classification model for training to obtain a trained text classification model;

[0016] The feature data that has not been manually annotated is input into the trained text classification model to obtain the structural data and non-structural data in the target project file.

[0017] As one preferred solution, the target project file includes text format data, image format data and video format data;

[0018] The feature extraction of the text format data and manual labeling of part of the extracted feature data to obtain a training data set specifically includes:

[0019] Performing text feature extraction on the text format data to obtain numerical features;

[0020] Extracting pixels from the image format data to obtain pixel value features;

[0021] Video frame features are obtained by performing video frame extraction on the video format data.

[0022] As one of the preferred solutions, the step of building a quantization rule base according to the demand file, matching the quantization rule base with the structure data, and obtaining quantifiable data includes:

[0023] Use a loop statement to iterate each field or combination of the structure data;

[0024] Based on the quantification rule base, for each of the fields or combinations, respectively, steps of checking data type, verifying range / format, performing calculation and logical judgment are performed to implement a matching operation;

[0025] Each successfully matched field or combination is recorded to obtain quantifiable data.

[0026] As one of the preferred solutions, the performing of semantic similarity analysis on the quantifiable data and the requirement document and obtaining a first ranking result of each of the target project files according to the analysis result includes:

[0027] Using natural language processing technology, calculate the cosine similarity of the fields or combinations that successfully match the quantifiable data and the requirement document, and obtain similarity calculation results of various quantifiable indicators;

[0028] According to the requirement document, assign corresponding weights to each of the quantifiable indicators;

[0029] Based on the similarity calculation results of each quantifiable indicator and the weight corresponding to each quantifiable indicator, the comprehensive similarity between the target project file and the requirement file is obtained, and the first ranking result is obtained according to the comprehensive similarity.

[0030] As one preferred solution, the step of inputting the unstructured data into a pre-trained sorting model to obtain a second sorting result includes:

[0031] Preprocess unstructured data;

[0032] Performing feature extraction on the preprocessed unstructured data, and manually annotating a portion of the extracted unstructured data based on project demand data to obtain a training data set;

[0033] Inputting the training data set into the pre-trained sorting model for secondary training to obtain the trained sorting model;

[0034] The unstructured data that has not been manually labeled is input into the trained sorting model, and a second sorting result is output.

[0035] As one preferred solution, obtaining the evaluation result of each associated object according to the log file includes:

[0036] assigning weights to the first sorting result and the second sorting result by analyzing the requirement document;

[0037] A weighted sum is performed according to the first sorting result and the second sorting result according to their corresponding weights to obtain evaluation results of the plurality of target project files.

[0038] Another embodiment of the present invention provides an automated project evaluation device, comprising:

[0039] An acquisition module, used to acquire a requirement file and a target project file of each associated object;

[0040] A classification module, used to classify each of the target project files according to a machine learning algorithm to obtain structured data and unstructured data;

[0041] A matching module, used for building a quantification rule base according to the demand file, matching the quantification rule base with the structure data, and obtaining quantifiable data;

[0042] An analysis module, configured to perform semantic similarity analysis on the quantifiable data and the requirement document, and obtain a first ranking result of each of the target project files according to the analysis result;

[0043] An output module, used for inputting the unstructured data into a pre-trained sorting model to obtain a second sorting result; wherein the sorting result is configured to be trained according to the historical information data of each of the associated objects;

[0044] An evaluation module is used to generate a log file for each of the associated objects according to the first sorting result and the second sorting result, and obtain an evaluation result for each of the associated objects according to the log file.

[0045] Another embodiment of the present invention provides an automated project evaluation device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the automated project evaluation method as described above when executing the computer program.

[0046] Yet another embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein when a device where the computer-readable storage medium is located executes the computer program, the automated project evaluation method as described above is implemented.

[0047] Compared with the prior art, the embodiments of the present invention have the following advantages:

[0048] 1) Use the sorting model to quickly process and analyze a large number of bidding documents, significantly improving the efficiency of the review work;

[0049] 2) Reduce errors caused by human factors through automation technology and improve the accuracy and reliability of review results;

[0050] 3) The system conducts reviews according to preset review rules, reducing the impact of subjective judgment on the review results and ensuring the fairness of the review

[0051] 4) The automated recording and display of the review process and results improves the transparency of the entire review activity and enhances the trust of all stakeholders. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 is a flow chart of an automated project evaluation method in one embodiment of the present invention;

[0053] Figure 2 is a structural flow chart of an automated project evaluation device in one embodiment of the present invention;

[0054] Figure 3 It is a schematic diagram of an automated project evaluation device in one embodiment of the present invention. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] In the description of this application, the terms "first", "second", "third", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the feature. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0057] In the description of the present application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two elements. The terms "vertical", "horizontal", "left", "right", "upper", "lower" and similar expressions used herein are only for illustrative purposes, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0058] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by those skilled in the art. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood by specific circumstances.

[0059] An embodiment of the present invention provides an automated project evaluation method. For details, see Figure 1 , Figure 1 The flowchart of the automated project evaluation method provided in one embodiment of the present invention is shown. It includes steps S1 to S3:

[0060] S1. Obtain the requirement file and the target project file of each associated object;

[0061] The requirement document includes a document that records the project's objectives, scope, functional requirements, non-functional requirements, etc. in detail by analyzing and organizing the project requirements during the project initiation phase. Its purpose is to clarify the project's objectives and requirements, provide clear guidance for project development, and ensure that the project team and stakeholders have a consistent understanding of the project requirements.

[0062] The target project documents, namely the bidding documents, refer to the responsive documents prepared by the bidder in response to the requirements of the bidding documents, which generally consist of business documents, technical documents, quotation documents and other parts. Its purpose is to demonstrate the bidder's capabilities and advantages in order to win the contract or opportunity of the project.

[0063] S2. Classify and process each of the target project files according to a machine learning algorithm to obtain structured data and unstructured data;

[0064] It is understandable that the bidding documents uploaded and submitted by the bidding suppliers include various data structures, so they need to be classified to ensure the integrity of the documents and the smooth progress of the review.

[0065] Preferably, in an embodiment of the present invention, the target project files are classified according to a machine learning algorithm to obtain structured data and unstructured data, including:

[0066] Convert the target project file into text format data;

[0067] Extracting features from the text format data, and manually annotating part of the extracted feature data to obtain a training data set;

[0068] Inputting the training data set into a pre-built text classification model for training to obtain a trained text classification model;

[0069] The feature data that has not been manually annotated is input into the trained text classification model to obtain the structural data and non-structural data in the target project file.

[0070] Preferably, in an embodiment of the present invention, the target project file includes text format data, image format data and video format data;

[0071] The feature extraction of the text format data and manual labeling of part of the extracted feature data to obtain a training data set specifically includes:

[0072] Performing text feature extraction on the text format data to obtain numerical features;

[0073] Extracting pixels from the image format data to obtain pixel value features;

[0074] Video frame features are obtained by performing video frame extraction on the video format data.

[0075] S3, constructing a quantification rule base according to the demand file, and matching the quantification rule base with the structure data to obtain quantifiable data;

[0076] Specifically, a quantitative rule base is constructed according to the requirement document, and the review rule information includes rule type (business or technical), whether to veto items, review method, whether to conduct a combined review, review conditions, review elements, judgment method, scoring rules, sorting method, project number, sub-bid number, and subcontract number. If there are rules that do not meet the requirements of this review, they can be disassembled according to semantics, and recommended quantitative rules can be given based on semantic analysis. In different bidding projects, you can customize the addition or screening of scoring rules. The review rules are open and transparent to all participants to enhance the trust of the review process. Support the management of historical review rule versions to track changes and support the review of historical review processes. The rule base will consider the weight distribution of different review dimensions, such as price, technology, delivery time, etc., and how they affect the final score.

[0077] Preferably, in an embodiment of the present invention, the step of constructing a quantification rule base according to the demand file, matching the quantification rule base with the structure data, and obtaining quantifiable data includes:

[0078] Use a loop statement to iterate each field or combination of the structure data;

[0079] Based on the quantification rule base, for each of the fields or combinations, respectively, steps of checking data type, verifying range / format, performing calculation and logical judgment are performed to implement a matching operation;

[0080] Each successfully matched field or combination is recorded to obtain quantifiable data.

[0081] Specifically, checking data types is used to ensure that the data type of each field or data item meets expectations. For example, if a field should be an integer type, then in this step it is checked whether the field actually contains an integer value, rather than a floating point number, string, or other non-integer type data; validation ranges usually involve checking whether the data value falls within the expected numerical range, for example, the age field may only allow values ​​between 0 and 120; format validation focuses on whether the representation of the data conforms to specific rules, for example, email addresses should contain the "@" symbol and domain name suffix, and dates should follow a specific date format (such as YYYY-MM-DD); performing calculations is used to perform mathematical calculations or apply specific algorithms based on field values, for example, calculating the difference, percentage, or performing more complex mathematical operations between two fields, which may be used to generate new field values, verify data consistency, or perform more advanced validation logic; logical judgments evaluate conditions based on the value of a data field or the result of a calculation. This may include simple conditional judgments (such as "If A is greater than B, then...") or more complex logical expressions. The results of these logical judgments may be used to control the process (such as skipping certain steps, triggering warnings or error messages) or generate final validation results.

[0082] S4, performing semantic similarity analysis on the quantifiable data and the requirement document, and obtaining a first ranking result of each of the target project files according to the analysis result;

[0083] Preferably, in an embodiment of the present invention, performing semantic similarity analysis on the quantifiable data and the requirement document, and obtaining a first ranking result of each of the target project files according to the analysis result, includes:

[0084] Using natural language processing technology, calculate the cosine similarity of the fields or combinations that successfully match the quantifiable data and the requirement document, and obtain similarity calculation results of various quantifiable indicators;

[0085] According to the requirement document, assign corresponding weights to each of the quantifiable indicators;

[0086] Based on the similarity calculation results of each quantifiable indicator and the weight corresponding to each quantifiable indicator, the comprehensive similarity between the target project file and the requirement file is obtained, and the first ranking result is obtained according to the comprehensive similarity.

[0087] Specifically, for fields that are already numeric, directly calculate the differences between them (such as absolute value difference, relative difference, etc.). For numeric data, use statistical indicators such as the Pearson correlation coefficient and the Spearman rank correlation coefficient to measure the correlation. Numerical data can be regarded as vectors (each field value as a dimension) and cosine similarity can be calculated. If the field contains text and similarity needs to be calculated, use NLP technology (such as TF-IDF, Word2Vec, BERT, etc.) to convert the text into numeric vectors and calculate the cosine similarity between these vectors.

[0088] For each successfully matched field or combination, calculate the weighted similarity based on its similarity result and the corresponding weight. If cosine similarity is used (it may be an adjusted method for numerical data), the weight is applied directly. Add all weighted similarities (or perform other forms of aggregation as needed, such as weighted average) to obtain the comprehensive similarity between the target project file and the requirement file. Sort the target project files or related records according to the comprehensive similarity. Files or records with higher comprehensive similarity are ranked higher, indicating that they match the requirement file more closely.

[0089] S5, inputting the unstructured data into a pre-trained sorting model to obtain a second sorting result; wherein the sorting result is configured to be trained according to the historical information data of each of the associated objects;

[0090] Preferably, in an embodiment of the present invention, the step of inputting the unstructured data into a pre-trained sorting model to obtain a second sorting result includes:

[0091] Preprocess unstructured data;

[0092] Performing feature extraction on the preprocessed unstructured data, and manually annotating a portion of the extracted unstructured data based on project demand data to obtain a training data set;

[0093] Inputting the training data set into the pre-trained sorting model for secondary training to obtain the trained sorting model;

[0094] The unstructured data that has not been manually labeled is input into the trained sorting model, and a second sorting result is output.

[0095] Through the above process, unstructured data can be effectively transformed into valuable review information, providing strong support for project decision-making. At the same time, due to the use of pre-trained sorting models and secondary training technology, this method can make full use of existing language knowledge and task-related data to improve the accuracy and efficiency of review results.

[0096] S6. Generate a log file for each of the associated objects according to the first sorting result and the second sorting result, and obtain an evaluation result for each of the associated objects according to the log file.

[0097] Preferably, in an embodiment of the present invention, obtaining the evaluation result of each associated object according to the log file includes:

[0098] assigning weights to the first sorting result and the second sorting result by analyzing the requirement document;

[0099] A weighted sum is performed according to the first sorting result and the second sorting result according to their corresponding weights to obtain evaluation results of the plurality of target project files.

[0100] According to the intelligent evaluation results, candidate suppliers are given according to the three dimensions of project, bid and package. Project refers to the project to be reviewed, bid refers to the number of sub-bids included in the project, and package refers to the sub-contracts included in each bid. Because the bidding suppliers will bid for the corresponding sub-contracts according to their own situation, each bidding supplier can get a quantitative score for each package according to the business score, technical score and price score. The final score of the bid package is obtained by summing up the weighted scores set by the three, and the ranking is based on the total score of the bidders of the package, that is, the candidate suppliers are given according to the package dimension; the candidates are given according to the bid dimension, which is based on the bid dimension, and all the bidding suppliers are ranked; the candidate suppliers are given according to the project dimension, which means that all the bidding suppliers are ranked according to the project dimension.

[0101] Through this process, the review experts only need to maintain the manual review items and synchronize the data to the review system. The review system can automatically calculate the ranking of each supplier based on the configured rules and the structured data of the bidding suppliers, and give candidate suppliers. The calculation process is automatically completed by the review system according to the configured calculation rules, which can minimize the calculation error. The results are provided for reference by the review experts, further improving the accuracy of the review and finding relatively suitable suppliers for the bidding suppliers.

[0102] Compared with the prior art, the embodiments of the present invention have the following advantages:

[0103] 1) Use the sorting model to quickly process and analyze a large number of bidding documents, significantly improving the efficiency of the review work;

[0104] 2) Reduce errors caused by human factors through automation technology and improve the accuracy and reliability of review results;

[0105] 3) The system conducts reviews according to preset review rules, reducing the impact of subjective judgment on the review results and ensuring the fairness of the review;

[0106] 4) The system can adapt to large-scale bidding activities and support the expansion and growth of the bidding market;

[0107] 5) The automated recording and display of the review process and results improves the transparency of the entire review activity and enhances the trust of all stakeholders.

[0108] Another embodiment of the present invention provides an automated project evaluation device. For details, see Figure 2 , Figure 2 The structure diagram of the automatic project evaluation device provided in one embodiment of the present invention is shown. It includes: an acquisition module 11, a classification module 12, a matching module 13, an analysis module 14, an output module 15, and an evaluation module 16, wherein:

[0109] An acquisition module 11 is used to acquire a requirement file and a target project file of each associated object;

[0110] A classification module 12, used to classify each of the target project files according to a machine learning algorithm to obtain structured data and unstructured data;

[0111] A matching module 13, configured to construct a quantification rule base according to the requirement file, and perform matching processing on the quantification rule base and the structure data to obtain quantifiable data;

[0112] An analysis module 14 is used to perform semantic similarity analysis on the quantifiable data and the requirement document, and obtain a first ranking result of each of the target project files according to the analysis result;

[0113] An output module 15, configured to input the unstructured data into a pre-trained sorting model to obtain a second sorting result; wherein the sorting result is configured to be trained according to the historical information data of each of the associated objects;

[0114] The evaluation module 16 is used to generate a log file for each of the associated objects according to the first sorting result and the second sorting result, and obtain an evaluation result for each of the associated objects according to the log file.

[0115] Another embodiment of the present invention provides an automated project evaluation device. For details, see Figure 3 , which is a schematic diagram of an automated project evaluation device provided in one embodiment of the present invention. The device comprises a processor 21, a memory 22, and a computer program stored in the memory 22 and configured to be executed by the processor 21. When the processor 21 executes the computer program, the steps in the automated project evaluation method embodiment described above are implemented, for example Figure 1or, when the processor 21 executes the computer program, the functions of the modules in the above-mentioned device embodiments are implemented, such as the acquisition module 11.

[0116] Exemplarily, the computer program can be divided into one or more modules, and the one or more modules are stored in the memory 22 and executed by the processor 21 to complete the present invention. The one or more modules can be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the automated project evaluation device. For example, the computer program can be divided into an acquisition module 11, a classification module 12, a matching module 13, an analysis module 14, an output module 15, and an evaluation module 16.

[0117] The processor 21 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor 21 is the control center of the automation project evaluation device, and uses various interfaces and lines to connect various parts of the entire automation project evaluation device.

[0118] The memory 22 can be used to store the computer program and / or module. The processor 21 realizes various functions of the automated project assessment device by running or executing the computer program and / or module stored in the memory 22 and calling the data stored in the memory 22. The memory 22 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory 22 can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0119] Wherein, if the module integrated in the automated project assessment 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. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0120] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0121] Accordingly, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to perform the steps in the automated project evaluation method of the above embodiment, for example Figure 1 Steps S1 to S6 described in .

[0122] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. An automated project evaluation method, characterized in that: The following steps are involved: Obtain the requirements file and the target project file for each associated object; Classify and process each of the target project files according to a machine learning algorithm to obtain structured data and unstructured data; Building a quantification rule base according to the demand file, and matching the quantification rule base with the structure data to obtain quantifiable data; Performing semantic similarity analysis on the quantifiable data and the requirement document, and obtaining a first ranking result of each of the target project files according to the analysis result; Inputting the unstructured data into a pre-trained sorting model to obtain a second sorting result; wherein the sorting result is configured to be trained according to the historical information data of each of the associated objects; A log file for each of the associated objects is generated according to the first sorting result and the second sorting result, and an evaluation result for each of the associated objects is obtained according to the log file.

2. The automated project evaluation method according to claim 1, characterized in that: The target project files are classified and processed according to the machine learning algorithm to obtain structured data and unstructured data, including: Convert the target project file into text format data; Extracting features from the text format data, and manually annotating part of the extracted feature data to obtain a training data set; Inputting the training data set into a pre-built text classification model for training to obtain a trained text classification model; The feature data that has not been manually annotated is input into the trained text classification model to obtain the structural data and non-structural data in the target project file.

3. The automated project evaluation method according to claim 2, characterized in that: The target project file includes text format data, image format data and video format data; The feature extraction of the text format data and manual labeling of part of the extracted feature data to obtain a training data set specifically includes: Performing text feature extraction on the text format data to obtain numerical features; Extracting pixels from the image format data to obtain pixel value features; Video frame features are obtained by performing video frame extraction on the video format data.

4. The automated project evaluation method according to claim 1, characterized in that: The step of constructing a quantification rule base according to the requirement file and matching the quantification rule base with the structure data to obtain quantifiable data includes: Use a loop statement to iterate each field or combination of the structure data; Based on the quantification rule base, for each of the fields or combinations, respectively, steps of checking data type, verifying range / format, performing calculation and logical judgment are performed to implement a matching operation; Each successfully matched field or combination is recorded to obtain quantifiable data.

5. The automated project evaluation method according to claim 1, characterized in that: The performing semantic similarity analysis on the quantifiable data and the requirement document, and obtaining a first ranking result of each of the target project files according to the analysis result, includes: Using natural language processing technology, calculate the cosine similarity of the fields or combinations that successfully match the quantifiable data and the requirement document, and obtain similarity calculation results of various quantifiable indicators; According to the requirement document, assign corresponding weights to each of the quantifiable indicators; Based on the similarity calculation results of each quantifiable indicator and the weight corresponding to each quantifiable indicator, the comprehensive similarity between the target project file and the requirement file is obtained, and the first ranking result is obtained according to the comprehensive similarity.

6. The automated project evaluation method according to claim 1, characterized in that: The step of inputting the unstructured data into a pre-trained sorting model to obtain a second sorting result includes: Preprocess unstructured data; Performing feature extraction on the preprocessed unstructured data, and manually annotating a portion of the extracted unstructured data based on project demand data to obtain a training data set; Inputting the training data set into the pre-trained sorting model for secondary training to obtain the trained sorting model; The unstructured data that has not been manually labeled is input into the trained sorting model, and a second sorting result is output.

7. The automated project evaluation method according to claim 1, characterized in that: The obtaining the evaluation result of each associated object according to the log file includes: assigning weights to the first sorting result and the second sorting result by analyzing the requirement document; A weighted sum is performed on the first sorting result and the second sorting result according to their corresponding weights to obtain evaluation results of the plurality of target files.

8. An automated project evaluation device, characterized in that: include: The acquisition module is used to obtain the requirement file and multiple target project files; A classification module, used to classify each of the target project files according to a machine learning algorithm to obtain structured data and unstructured data; A matching module, configured to construct a quantification rule base according to the requirement file, and match quantifiable indicators corresponding to the structure data based on the quantification rule base to obtain quantifiable data; An analysis module, configured to perform semantic similarity analysis on the quantifiable data and the requirement document, and obtain a first ranking result of each of the target project files according to the analysis result; An output module, used for inputting the unstructured data into a pre-trained sorting model to obtain a second sorting result; An evaluation module is used to generate a log file of each of the target project files according to the first sorting result and the second sorting result, and obtain an evaluation result of each of the associated objects according to the log file.

9. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for automated project evaluation as claimed in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the automated project evaluation method according to any one of claims 1 to 7 is implemented.

Images