An automatic evaluation method and system for enterprise bidding data

Abstract

This invention relates to the field of data recognition technology and provides an automatic evaluation method and system for enterprise bidding data. The automatic evaluation method includes: extracting text data from the enterprise bidding data; obtaining keywords corresponding to each evaluation item and matching the text positions corresponding to the keywords in the text data; confirming whether the text position is a text region corresponding to the evaluation item based on the contextual relationship of the keywords in the text position; if the text position is determined to be a text region corresponding to the evaluation item, extracting project information following the keywords; and scoring the project information according to a preset scoring strategy corresponding to the evaluation item to obtain the total score corresponding to the enterprise bidding data. This solution not only improves the accuracy and reliability of the evaluation results but also significantly enhances the efficiency and fairness of the evaluation process.

Description

Technical Field

[0001] This invention belongs to the technical field of data recognition, and in particular relates to an automatic evaluation method and system for enterprise bidding data. Background Technology

[0002] With the continuous development of the social economy, competition among enterprises is becoming increasingly fierce. Bidding and tendering, as an important way for enterprises to participate in market competition, is widely used in various engineering projects, material procurement, and service outsourcing. In the traditional bidding and tendering process, manual evaluation requires a significant amount of time and effort to thoroughly review and score each bid document. This is not only susceptible to human factors, leading to subjectivity and inconsistency in scoring, but also suffers from inefficiency.

[0003] In recent years, with the rapid development of information technology and big data technology, automated bidding evaluation technology has gradually become a research and application hotspot. Automated processing and analysis of bidding data can improve the efficiency and fairness of the bidding evaluation process. However, existing automated bidding evaluation methods still have some shortcomings. For example, current technologies struggle to accurately extract effective evaluation information from complex text data, and lack flexibility and intelligence in the scoring process, thus affecting the accuracy and reliability of the evaluation results. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide an automatic evaluation method and system for enterprise bidding data to solve the technical problem of how to ensure the accuracy and reliability of bidding results.

[0005] A first aspect of this invention provides an automatic evaluation method for enterprise bidding data, the method comprising:

[0006] Extract the text data from the enterprise bidding data;

[0007] Obtain the keywords corresponding to each evaluation item, and match the text positions corresponding to the keywords in the text data;

[0008] Based on the contextual relationship of the keywords in the text location, confirm whether the text location is the text area corresponding to the evaluation item;

[0009] If the text location is determined to be the text area corresponding to the evaluation item, then extract the item information located after the keyword;

[0010] The project information is scored according to the preset scoring strategy corresponding to the evaluation items to obtain the total score corresponding to the enterprise bidding data; the total score is used to sort different enterprise bidding data.

[0011] Furthermore, the step of confirming whether the text location is the text region corresponding to the evaluation item based on the contextual relationship of the keywords in the text location includes:

[0012] If there is one and only one text location corresponding to the keyword, then the text location is confirmed as the text area corresponding to the evaluation item.

[0013] If there are multiple text locations corresponding to the keyword, then the semantic scores corresponding to the multiple text locations are calculated;

[0014] The text position corresponding to the maximum semantic score is taken as the text region corresponding to the evaluation item.

[0015] Furthermore, if there are multiple text positions corresponding to the keyword, the step of calculating the semantic scores corresponding to the multiple text positions includes:

[0016] If there are multiple text positions corresponding to the keyword, then obtain the related words and standard sentence library corresponding to the keyword;

[0017] Construct the first distribution feature matrix corresponding to the text position;

[0018] Construct the second distribution feature matrix corresponding to each of the multiple standard statements in the standard statement library;

[0019] Calculate the similarity between the first distribution feature matrix and the second distribution feature matrix;

[0020] The maximum similarity among the multiple standard statements is taken as the semantic score.

[0021] Further, the step of constructing the first distribution feature matrix corresponding to the text position includes:

[0022] Match the currently relevant words present at the text position, and calculate the first word code corresponding to the currently relevant words;

[0023] Extract the first word spacing between keywords and multiple currently related words in the text location;

[0024] A first distribution feature matrix is ​​constructed based on multiple first word codes and multiple first word intervals.

[0025] Further, the step of constructing a first distribution feature matrix based on multiple first word codes and multiple first word intervals includes:

[0026] Obtain the matrix template corresponding to the keywords; wherein, the different element positions in the matrix template correspond to the first word code and the first word spacing corresponding to different related words in turn;

[0027] The first word codes and first word spacings corresponding to multiple related words are matched to the corresponding element positions in the matrix template;

[0028] The positions of the blank elements in the matrix template are set to fixed values ​​to obtain the first distribution feature matrix.

[0029] Furthermore, the step of scoring the project information according to the preset scoring strategy corresponding to the evaluation item to obtain the total score corresponding to the enterprise bidding data includes:

[0030] Obtain a preset mapping table and match the sub-ratings corresponding to each of the multiple project information;

[0031] Each of the sub-scores is multiplied by its weight to obtain multiple numerical values;

[0032] The sum of multiple values ​​is taken as the total score corresponding to the enterprise's bidding data.

[0033] Furthermore, after the step of taking the maximum similarity among the multiple standard statements as the semantic score, the method further includes:

[0034] If the similarity is greater than the threshold, the maximum similarity among the multiple standard statements will be used as the semantic score.

[0035] If the similarity is not greater than the threshold, it is confirmed that there is no text region in the text data corresponding to the evaluation item corresponding to the similarity.

[0036] A second aspect of the present invention provides an automatic evaluation device for enterprise bidding data, comprising:

[0037] The extraction unit is used to extract text data from the enterprise bidding data;

[0038] The acquisition unit is used to acquire the keywords corresponding to each evaluation item and match the text positions corresponding to the keywords in the text data.

[0039] The confirmation unit is used to confirm whether the text location is the text area corresponding to the evaluation item based on the contextual relationship of the keywords in the text location;

[0040] The judgment unit is used to extract the item information located after the keyword if it is determined that the text position is the text area corresponding to the evaluation item;

[0041] The scoring unit is used to score the project information according to the preset scoring strategy corresponding to the evaluation item, and obtain the total score corresponding to the enterprise bidding data; the total score is used to sort different enterprise bidding data.

[0042] A third aspect of the present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the automatic evaluation method for enterprise bidding data described in the first aspect.

[0043] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the automatic evaluation method for enterprise bidding data described in the first aspect.

[0044] The beneficial effects of this invention compared to existing technologies are as follows: By extracting text data from bidding data and utilizing keyword matching technology, the text regions corresponding to evaluation items are accurately located. This method reduces errors from manual identification and improves the accuracy of information extraction. Based on keyword matching, the contextual relationships of keywords are further analyzed to ensure the accuracy of the extracted text regions. This process effectively avoids keyword misjudgment and improves the accuracy of text region confirmation. According to a preset scoring strategy, the extracted project information is automatically scored, ensuring the consistency and objectivity of the scoring process. The preset scoring strategy is based on the requirements of specific evaluation items and can flexibly adapt to different evaluation standards. By calculating the total score of enterprise bidding data, the bidding data of different enterprises are sorted, providing an objective and fair evaluation result. This function not only improves the efficiency of the evaluation work but also reduces the possibility of human intervention, ensuring the fairness and transparency of the evaluation results. The automated evaluation method significantly reduces the time and effort required for manual operation, improving the overall efficiency of the evaluation work. Evaluation personnel can focus more on optimizing the evaluation strategy and reviewing the results. Through a systematic scoring and ranking mechanism, this method minimizes potential subjective biases during the evaluation process, ensuring fair and impartial assessment of every evaluation item. The method can flexibly adjust keywords and scoring strategies according to the needs of different enterprises and projects, exhibiting strong adaptability and scalability, and can be widely applied to various bidding and tendering evaluation scenarios. In summary, the automatic evaluation method for enterprise bidding data of this invention, through precise information extraction, contextual analysis, and automated scoring and ranking, not only improves the accuracy and reliability of evaluation results but also significantly enhances the efficiency and fairness of the evaluation process. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 A schematic flowchart of an automatic evaluation method for enterprise bidding data provided by the present invention is shown;

[0047] Figure 2 This diagram illustrates an automatic evaluation device for enterprise bidding data according to an embodiment of the present invention.

[0048] Figure 3 A schematic diagram of a terminal device provided in an embodiment of the present invention is shown. Detailed Implementation

[0049] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0050] This invention provides an automatic evaluation method and system for enterprise bidding data to address the technical problem of ensuring the accuracy and reliability of bid evaluation results.

[0051] First, this invention provides an automatic evaluation method for enterprise bidding data. Please refer to [link / reference]. Figure 1 , Figure 1 This diagram illustrates an automated evaluation method for enterprise bidding data provided by the present invention. Figure 1 As shown, the automatic evaluation method for enterprise bidding data may include the following steps:

[0052] Step 101: Extract text data from the enterprise bidding data;

[0053] Extract all text data from the company's bidding documents. This text data forms the basis for subsequent analysis and processing.

[0054] Step 102: Obtain the keywords corresponding to each evaluation item, and match the text positions corresponding to the keywords in the text data;

[0055] Evaluation items include, but are not limited to, pricing, payment method, payment time, company qualifications, personnel qualifications, human resources, material resources, and financial resources.

[0056] Different evaluation items correspond to different keywords. For example, the keywords for the evaluation item "quotation" are usually price or quotation, while the keywords for the evaluation item "payment method" are payment method or settlement method, and so on for other evaluation items.

[0057] Step 103: Based on the contextual relationship of the keywords in the text location, confirm whether the text location is the text area corresponding to the evaluation item;

[0058] Since there may be multiple text locations for a keyword, but there is usually only one text location for the evaluation item corresponding to the keyword, it is necessary to analyze the context of the keyword location to confirm whether the text location is the text area corresponding to the evaluation item.

[0059] For example, in a tender document, we need to find the evaluation items related to "quotation". Suppose the word "quotation" may appear multiple times in the document, but there is only one location in the text area that actually corresponds to the evaluation item.

[0060] First, you need to search for the keyword "quotation" throughout the entire bidding document and record all instances where it appears. For example, suppose "quotation" appears in the following places:

[0061] Page 3, paragraph 2; Page 15, paragraph 3; and Page 18, paragraph 1.

[0062] Analyze the context of keywords: For each location where "quote" appears, examine the surrounding text, especially the context, to determine whether the location is relevant to the evaluation item.

[0063] For example:

[0064] The text in paragraph 2 on page 3 reads: "The final bid price for this project will be announced after the bid opening, and all bidders must ensure the accuracy and competitiveness of their bids."

[0065] The text in paragraph 3 on page 15 reads: "The tender documents shall include the following: company profile, technical solutions, quotation, project plan and other relevant documents."

[0066] The text in the first paragraph on page 18 reads: "The price of this project is xxxxx million yuan (RMB)."

[0067] The second paragraph on page 3 states that "the final price will be published after the bid opening," but this mainly describes the timing and accuracy requirements for the price publication, not specific scoring criteria or evaluation items. Therefore, this location is not a text area for evaluation items. The third paragraph on page 15 lists the contents that a bid should include, including a price, but does not detail how the price is evaluated or its specific scoring criteria in the bid evaluation. Therefore, this location is also not a text area for evaluation items. However, the first paragraph on page 18 explicitly mentions specific figures and contains other relevant vocabulary, thus this location is a text area for evaluation items.

[0068] Specifically, step 103 includes steps 1031 to 1033:

[0069] Step 1031: If there is one and only one text location corresponding to the keyword, then confirm that the text location is the text area corresponding to the evaluation item;

[0070] If only one location in the text matches the keyword, that location is directly identified as a text region relevant to the evaluation item. In this case, since there is only one match, the system can determine its relevance without further analysis.

[0071] Step 1032: If there are multiple text positions corresponding to the keyword, calculate the semantic scores corresponding to the multiple text positions;

[0072] When a keyword appears multiple times in the text, meaning there are multiple matching locations, the system needs to further analyze the contextual relationships of these locations. Specifically, this involves calculating a semantic score for each matching location, which reflects the relevance of each location to the evaluation item. The specific calculation logic for the semantic score is as follows:

[0073] Specifically, step 1032 includes steps A1 to A5:

[0074] Step A1: If there are multiple text positions corresponding to the keyword, then obtain the related words and standard sentence library corresponding to the keyword;

[0075] Related terms refer to words that typically appear alongside keywords. For example, related terms for the keyword "quotation" include "RMB," "USD," and "project," etc. Since different documents use different expressions or conventions, it is necessary to pre-define multiple standard phrases corresponding to different expressions for subsequent semantic similarity calculations.

[0076] When a keyword appears in multiple locations in the text, the system needs to obtain related words and a standard statement library. The standard statement library contains predefined standard statements related to the evaluation items, which are used for subsequent semantic similarity calculations.

[0077] Step A2: Construct the first distribution feature matrix corresponding to the text position;

[0078] Specifically, step A2 includes steps A21 to A23:

[0079] Step A21: Match the currently relevant words present at the text position, and calculate the first word code corresponding to the currently relevant words;

[0080] The process involves finding words in the text that are related to the keywords (currently relevant words) and then calculating the encoding of these words. Word encoding can be implemented using various methods, such as word embeddings or word embedding techniques (e.g., Word2Vec, GloVe, BERT, etc.). These encodings transform each word into a high-dimensional vector representing its semantic features.

[0081] Step A22: Extract the first word spacing between keywords and multiple currently related words in the text location;

[0082] Calculate the distance between the keyword and each currently related word, i.e., the word spacing. Word spacing can be the difference in position of words in the text; for example, if the keyword is in the 5th position and the related word is in the 8th position, the word spacing is 3. Word spacing reflects the relative positional relationship between words, which is crucial for capturing the local structure and semantic context of the text.

[0083] Step A23: Construct a first distribution feature matrix based on multiple first word codes and multiple first word intervals.

[0084] Specifically, step A23 includes steps A231 to A233:

[0085] Step A231: Obtain the matrix template corresponding to the keywords; wherein, the different element positions in the matrix template correspond to the first word code and the first word spacing corresponding to different related words in turn;

[0086] A matrix template is predefined for each keyword. The positions of the elements in this template will be used to store the encoding of related words and word spacing. The design aims to store the feature vectors of related words at specific positions.

[0087] Step A232: Match the first word codes and first word spacings corresponding to multiple related words to the corresponding element positions in the matrix template;

[0088] The codes and word spacings of the found related words are then filled into the corresponding positions in the matrix template. Specifically, these codes and spacings can be sequentially placed into predetermined rows or columns of the matrix. For example, one row of the template might be dedicated to coding, and another row to spacing.

[0089] Step A233: Set the positions of the blank elements in the matrix template to fixed values ​​to obtain the first distribution feature matrix.

[0090] For blank spaces in the matrix template that are not filled with the codes of related words or word spacing, a fixed value (e.g., 0 or -1) is assigned to ensure the numerical integrity and consistency of the matrix. The matrix processed in this way is the first distribution feature matrix, ready for subsequent analysis or calculation.

[0091] In this embodiment, a distribution feature matrix is ​​constructed by matching relevant words in the text location, calculating their word codes, and combining the word spacing information between keywords and relevant words. This method can comprehensively capture the semantic features of the text location, effectively improving the accuracy of semantic matching. Extracting the word spacing information between keywords and multiple relevant words allows the distribution feature matrix to reflect the detailed relationships within the text, thus achieving precise analysis and understanding of text regions at a fine-grained level. By introducing a combination of word codes and word spacing, the constructed distribution feature matrix considers not only the information of the words themselves but also the relative positional relationships between words, enhancing the system's ability to understand the text context. Based on word codes and word spacing information, the distribution feature matrix can more accurately capture the semantic environment of keywords, effectively reducing the misjudgment rate when keywords appear multiple times, ensuring more reliable text region recognition. Automated construction of the distribution feature matrix reduces the need for manual intervention, improves overall processing efficiency, and enables the system to quickly and accurately identify text regions when faced with large amounts of text. This technical solution can handle complex text environments, including long texts and structurally complex sentences, ensuring high-level recognition performance in various text scenarios through a refined distribution feature matrix construction method. In summary, this technical solution constructs a first distribution feature matrix corresponding to the text position by matching current related words, calculating word codes, and extracting word spacing information, thereby achieving accurate identification and semantic understanding of text regions and significantly improving the accuracy and efficiency of matching.

[0092] Step A3: Construct the second distribution feature matrix corresponding to each of the multiple standard statements in the standard statement library;

[0093] The construction method of the second distribution feature matrix is ​​the same as that of the first distribution feature matrix, and will not be repeated here.

[0094] Step A4: Calculate the similarity between the first distribution feature matrix and the second distribution feature matrix;

[0095] Similarity can be calculated using various methods, such as cosine similarity, Euclidean distance, and dot product. The level of similarity reflects the semantic closeness between the text and the standard statement.

[0096] Step A5: Take the maximum similarity among the multiple standard statements as the semantic score.

[0097] For each text location, the system selects the maximum value from all calculated similarity scores as the semantic score for that location. This maximum value represents the similarity between that text location and the most similar standard statement in the standard statement library, thus reflecting that the semantics of that location best meets the requirements of the evaluation criteria.

[0098] Specifically, step A5 includes: if the similarity is greater than a threshold, then the maximum value of the similarity among the multiple standard statements is taken as the semantic score; if the similarity is not greater than the threshold, then it is confirmed that there is no text region in the text data corresponding to the evaluation item corresponding to the similarity.

[0099] In this embodiment, by acquiring related words and a standard sentence library of keywords, a distribution feature matrix of text location and standard sentences is constructed, which can comprehensively capture the semantic features of the text, thereby significantly improving the accuracy of text matching. Using the distribution feature matrix, the system can better understand the meaning of keywords in different contexts and accurately identify the text location that best matches the semantics of the evaluation item by calculating similarity. By calculating the similarity between the first distribution feature matrix and multiple second distribution feature matrices and selecting the maximum similarity as the semantic score, the system can effectively avoid mismatches when keywords appear multiple times, ensuring that the selected text region best meets the actual needs. This scheme utilizes the construction and calculation of the standard sentence library and the distribution feature matrix to achieve automated recognition from keywords to text regions, reducing manual intervention and improving processing efficiency. Through matrix similarity calculation, the system can quickly find the text region that best matches the standard sentence in a large amount of text, significantly optimizing system performance and making it suitable for large-scale text processing scenarios. This technical solution enhances the system's robustness to noise and text variations through comprehensive analysis of multiple features, maintaining high accuracy even in complex or ambiguous text environments. In summary, this technical solution, by introducing keyword-related words and standard sentence libraries and combining them with distribution feature matrix similarity calculation, provides an efficient and accurate method for text region identification, significantly improving the accuracy of text matching.

[0100] Step 1033: Take the text position corresponding to the maximum semantic score as the text region corresponding to the evaluation item.

[0101] By comparing the semantic scores of all matching positions, the position with the highest score is selected as the text region corresponding to the final evaluation item. This means that the context of that position best meets the requirements of the evaluation item.

[0102] In this embodiment, when a keyword appears only once in the text, the text location can be directly identified as the text region corresponding to the evaluation item, avoiding unnecessary complex calculations and improving the accuracy and efficiency of matching. When a keyword appears multiple times in the text, by calculating the semantic score of each text location, the text region that best matches the context of the evaluation item can be effectively distinguished. The text location with the highest semantic score represents the best match, significantly improving the selection accuracy in cases of multiple occurrences. Through the calculation of contextual relationships and semantic scores, this scheme not only relies on simple keyword matching but also focuses on the semantic understanding of text content, thereby achieving accurate matching at a deeper level. This method can better handle keyword recognition tasks in complex text environments. When a keyword appears multiple times, traditional methods may misjudge, while this technical solution, through a semantic scoring mechanism, can effectively reduce the misjudge rate and ensure that the identified text region is highly consistent with the actual needs of the evaluation item. The semantic scoring-based calculation method enables the system to automatically filter out the most relevant text regions when processing large amounts of text, improving overall processing efficiency and reducing the need for manual intervention. In summary, this technical solution can effectively improve the accuracy and efficiency of text region matching through intelligent analysis of keyword contextual relationships and accurate calculation of semantic scores.

[0103] Step 104: If the text location is determined to be the text area corresponding to the evaluation item, then extract the item information located after the keyword;

[0104] The location of the item information following the keywords is relatively fixed and follows a fixed format. Therefore, the item information following the keywords can be extracted based on their relative position and format.

[0105] Step 105: Based on the preset scoring strategy corresponding to the evaluation item, score the project information to obtain the total score corresponding to the enterprise bidding data; the total score is used to sort different enterprise bidding data.

[0106] Specifically, step 105 includes steps 1051 to 1053:

[0107] Step 1051: Obtain a preset mapping table and match the sub-ratings corresponding to each of the multiple items;

[0108] Obtain a pre-defined mapping table that defines different evaluation items (such as project quality, time, cost, etc.) and their corresponding scoring criteria. Use this mapping table to map each project's information (such as the quality, time, cost, etc. of a specific project) to its corresponding sub-score.

[0109] Step 1052: Multiply each of the sub-scores by its weight to obtain multiple numerical values;

[0110] For each sub-rating, a weighted average is applied according to preset weights. These weights reflect the importance of each evaluation item in the overall score. Therefore, each sub-rating is multiplied by its corresponding weight to obtain the weighted average value.

[0111] Step 1053: The sum of the multiple values ​​is taken as the total score corresponding to the enterprise's bidding data.

[0112] The weighted values ​​are summed to obtain a total score, which is the final overall score of the company's bidding data. This overall score can be used to compare bids from different companies, helping decision-makers select the optimal bidding proposal.

[0113] As an optional embodiment of the present invention, in order to provide a more accurate score, the preset mapping table can be functionalized and calculated in the following manner:

[0114]

[0115] Where T represents the total score, w i f represents the weight of the i-th evaluation item. i (S i ) represents the evaluation function of the i-th evaluation item (obtained by functionalizing the preset mapping table), M i A represents the maximum score of the i-th evaluation item. i represents the adjustment factor for the i-th evaluation item, used to correct the weight or influence of the score, and n represents the number of evaluation items.

[0116] Adjustment factors are used to correct the score weight or proportion of each evaluation item, so as to more accurately reflect the importance of the evaluation item in different contexts. This allows the scoring system to flexibly respond to different evaluation needs. For example, if a certain evaluation item is particularly important in a specific project, its score influence can be increased by adjusting factors. Different scoring strategy functions can be used for different evaluation items, ensuring that the scoring method for each evaluation item best reflects its characteristics and key indicators. By introducing adjustment factors and nonlinear combination functions, the expressiveness and adaptability of the system are improved. Scoring strategy functions, nonlinear combination functions, and correction functions can be further added or modified according to actual needs to adapt to different application scenarios.

[0117] In this embodiment, text data is extracted from bidding data, and keyword matching technology is used to accurately locate the text regions corresponding to the evaluation items. This method reduces the error of manual identification and improves the accuracy of information extraction. Based on keyword matching, the contextual relationship of keywords is further analyzed to ensure the accuracy of the extracted text regions. This process effectively avoids keyword misjudgment and improves the accuracy of text region confirmation. According to a preset scoring strategy, the extracted project information is automatically scored to ensure the consistency and objectivity of the scoring process. The preset scoring strategy is based on the requirements of specific evaluation items and can flexibly adapt to different evaluation standards. By calculating the total score of the enterprise bidding data, the bidding data of different enterprises are sorted, providing an objective and fair evaluation result. This function not only improves the efficiency of the evaluation work but also reduces the possibility of human intervention, ensuring the fairness and transparency of the evaluation results. The automated evaluation method significantly reduces the time and effort of manual operation and improves the overall efficiency of the evaluation work. Evaluation personnel can focus more on optimizing the evaluation strategy and reviewing the results. Through a systematic scoring and ranking mechanism, this method minimizes potential subjective biases during the evaluation process, ensuring fair and impartial assessment of every evaluation item. The method can flexibly adjust keywords and scoring strategies according to the needs of different enterprises and projects, exhibiting strong adaptability and scalability, and can be widely applied to various bidding and tendering evaluation scenarios. In summary, the automatic evaluation method for enterprise bidding data of this invention, through precise information extraction, contextual analysis, and automated scoring and ranking, not only improves the accuracy and reliability of evaluation results but also significantly enhances the efficiency and fairness of the evaluation process.

[0118] like Figure 2 This invention provides an automatic evaluation device for enterprise bidding data. Please refer to [link / reference]. Figure 2 , Figure 2 This diagram illustrates an automatic evaluation device for enterprise bidding data provided by the present invention. Figure 2 The automatic evaluation device for enterprise bidding data shown includes:

[0119] Extraction unit 21 is used to extract text data from the enterprise bidding data;

[0120] Acquisition unit 22 is used to acquire the keywords corresponding to each evaluation item and match the text positions corresponding to the keywords in the text data;

[0121] The confirmation unit 23 is used to confirm whether the text location is the text area corresponding to the evaluation item based on the contextual relationship of the keywords in the text location;

[0122] The judgment unit 24 is used to extract the item information located after the keyword if it is determined that the text position is the text area corresponding to the evaluation item;

[0123] The scoring unit 25 is used to score the project information according to the preset scoring strategy corresponding to the evaluation item, and obtain the total score corresponding to the enterprise bidding data; the total score is used to sort different enterprise bidding data.

[0124] This invention provides an automatic evaluation device for enterprise bidding data. It extracts text data from bidding data and uses keyword matching technology to accurately locate the text regions corresponding to evaluation items. This method reduces errors from manual identification and improves the accuracy of information extraction. Based on keyword matching, it further analyzes the contextual relationships of keywords to ensure the accuracy of the extracted text regions. This process effectively avoids keyword misjudgment and improves the accuracy of text region confirmation. According to a preset scoring strategy, the extracted project information is automatically scored, ensuring the consistency and objectivity of the scoring process. The preset scoring strategy is based on the requirements of specific evaluation items and can flexibly adapt to different evaluation standards. By calculating the total score of the enterprise bidding data, the bidding data of different enterprises are ranked, providing an objective and fair evaluation result. This function not only improves the efficiency of the evaluation work but also reduces the possibility of human intervention, ensuring the fairness and transparency of the evaluation results. The automated evaluation method significantly reduces the time and effort required for manual operation, improving the overall efficiency of the evaluation work. Evaluation personnel can focus more on optimizing the evaluation strategy and reviewing the results. Through a systematic scoring and ranking mechanism, this method minimizes potential subjective biases during the evaluation process, ensuring fair and impartial assessment of every evaluation item. The method can flexibly adjust keywords and scoring strategies according to the needs of different enterprises and projects, exhibiting strong adaptability and scalability, and can be widely applied to various bidding and tendering evaluation scenarios. In summary, the automatic evaluation method for enterprise bidding data of this invention, through precise information extraction, contextual analysis, and automated scoring and ranking, not only improves the accuracy and reliability of evaluation results but also significantly enhances the efficiency and fairness of the evaluation process.

[0125] Figure 3 This is a schematic diagram of a terminal device provided in an embodiment of the present invention. Figure 3 As shown, a terminal device 3 in this embodiment includes: a processor 30, a memory 31, and a computer program 32 stored in the memory 31 and executable on the processor 30, such as an automatic evaluation program for enterprise bidding data. When the processor 30 executes the computer program 32, it implements the steps in the various embodiments of the automatic evaluation method for enterprise bidding data described above, for example... Figure 1Steps 101 to 105 are shown. Alternatively, when the processor 30 executes the computer program 32, it implements the functions of each unit in the above-described device embodiments, for example... Figure 2 The function of the unit shown.

[0126] For example, the computer program 32 can be divided into one or more units, which are stored in the memory 31 and executed by the processor 30 to complete the present invention. The one or more units can be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program 32 in the terminal device 3. For example, the specific functions of each unit of the computer program 32 can be divided as follows:

[0127] The extraction unit is used to extract text data from the enterprise bidding data;

[0128] The acquisition unit is used to acquire the keywords corresponding to each evaluation item and match the text positions corresponding to the keywords in the text data.

[0129] The confirmation unit is used to confirm whether the text location is the text area corresponding to the evaluation item based on the contextual relationship of the keywords in the text location;

[0130] The judgment unit is used to extract the item information located after the keyword if it is determined that the text position is the text area corresponding to the evaluation item;

[0131] The scoring unit is used to score the project information according to the preset scoring strategy corresponding to the evaluation item, and obtain the total score corresponding to the enterprise bidding data; the total score is used to sort different enterprise bidding data.

[0132] The terminal device includes, but is not limited to, a processor 30 and a memory 31. Those skilled in the art will understand that... Figure 3 This is merely an example of a terminal device 3 and does not constitute a limitation on a terminal device 3. It may include more or fewer components than shown, or combine certain components, or different components. For example, the terminal device may also include input / output devices, network access devices, buses, etc.

[0133] The processor 30 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0134] The memory 31 can be an internal storage unit of the terminal device 3, such as a hard disk or memory of the terminal device 3. The memory 31 can also be an external storage device of the terminal device 3, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the terminal device 3. Furthermore, the memory 31 can include both internal and external storage units of the terminal device 3. The memory 31 is used to store the computer program and other programs and data required by the roaming control device. The memory 31 can also be used to temporarily store data that has been output or will be output.

[0135] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0136] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of the present invention. For details on their specific functions and technical effects, please refer to the method embodiments section, which will not be repeated here.

[0137] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this invention. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0138] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0139] This invention provides a computer program product that, when run on a mobile terminal, enables the mobile terminal to implement the steps described in the above-described method embodiments.

[0140] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0141] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0142] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0143] In the embodiments provided by this invention, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0144] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units.

[0145] It should be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0146] It should also be understood that the term “and / or” as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0147] As used in this specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if [the described condition or event] is detected" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once [the described condition or event] is detected," or "in response to detection of [the described condition or event]."

[0148] Furthermore, in the description of this invention and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0149] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of the invention include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0150] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. An automatic evaluation method for enterprise bidding data, characterized in that, The automatic evaluation method for enterprise bidding data includes: Extract the text data from the enterprise bidding data; Obtain the keywords corresponding to each evaluation item, and match the text positions corresponding to the keywords in the text data; Based on the contextual relationship of the keywords in the text location, confirm whether the text location is the text area corresponding to the evaluation item; If the text location is determined to be the text area corresponding to the evaluation item, then extract the item information located after the keyword; The project information is scored according to the preset scoring strategy corresponding to the evaluation items to obtain the total score corresponding to the enterprise bidding data; the total score is used to sort different enterprise bidding data. The step of determining whether a text location corresponds to a text region for an evaluation item based on the contextual relationship of keywords within that text location includes: If there is one and only one text location corresponding to the keyword, then the text location is confirmed as the text area corresponding to the evaluation item. If there are multiple text locations corresponding to the keyword, then the semantic scores corresponding to the multiple text locations are calculated; The text position corresponding to the maximum semantic score is taken as the text region corresponding to the evaluation item; Wherein, the step of calculating the semantic score corresponding to the multiple text positions if the keyword corresponds to multiple text positions includes: If there are multiple text positions corresponding to the keyword, then obtain the related words and standard sentence library corresponding to the keyword; Construct the first distribution feature matrix corresponding to the text position; Construct the second distribution feature matrix corresponding to each of the multiple standard statements in the standard statement library; Calculate the similarity between the first distribution feature matrix and the second distribution feature matrix; The maximum similarity among the multiple standard statements is taken as the semantic score; The step of constructing the first distribution feature matrix corresponding to the text position includes: Match the currently relevant words present at the text position, and calculate the first word code corresponding to the currently relevant words; Extract the first word spacing between keywords and multiple currently related words in the text location; wherein, the first word spacing refers to the positional difference of words in the text; Based on multiple first word codes and multiple first word intervals, a first distribution feature matrix is ​​constructed; wherein, the construction logic of the second distribution feature matrix is ​​the same as that of the first distribution feature matrix.

2. The automatic evaluation method for enterprise bidding data as described in claim 1, characterized in that, The step of constructing a first distribution feature matrix based on multiple first word codes and multiple first word intervals includes: Obtain the matrix template corresponding to the keywords; wherein, the different element positions in the matrix template correspond to the first word code and the first word spacing corresponding to different related words in turn; The first word codes and first word spacings corresponding to multiple related words are matched to the corresponding element positions in the matrix template; The positions of the blank elements in the matrix template are set to fixed values ​​to obtain the first distribution feature matrix.

3. The automatic evaluation method for enterprise bidding data as described in claim 1, characterized in that, The step of scoring the project information according to the preset scoring strategy corresponding to the evaluation item to obtain the total score corresponding to the enterprise bidding data includes: Obtain a preset mapping table and match the sub-ratings corresponding to each of the multiple project information; Each of the sub-scores is multiplied by its weight to obtain multiple numerical values; The sum of multiple values ​​is taken as the total score corresponding to the enterprise's bidding data.

4. The automatic evaluation method for enterprise bidding data as described in claim 1, characterized in that, The step of using the maximum similarity among the multiple standard statements as the semantic score includes: If the similarity is greater than the threshold, the maximum similarity among the multiple standard statements will be used as the semantic score. If the similarity is not greater than the threshold, it is confirmed that there is no text region in the text data corresponding to the evaluation item corresponding to the similarity.

5. An automatic evaluation device for enterprise bidding data, characterized in that, The automatic evaluation device for enterprise bidding data includes: The extraction unit is used to extract text data from the enterprise bidding data; The acquisition unit is used to acquire the keywords corresponding to each evaluation item and match the text positions corresponding to the keywords in the text data. The confirmation unit is used to confirm whether the text location is the text area corresponding to the evaluation item based on the contextual relationship of the keywords in the text location; The judgment unit is used to extract the item information located after the keyword if it is determined that the text position is the text area corresponding to the evaluation item; The scoring unit is used to score the project information according to the preset scoring strategy corresponding to the evaluation item, and obtain the total score corresponding to the enterprise bidding data; the total score is used to sort different enterprise bidding data. The step of determining whether a text location corresponds to a text region for an evaluation item based on the contextual relationship of keywords within that text location includes: If there is one and only one text location corresponding to the keyword, then the text location is confirmed as the text area corresponding to the evaluation item. If there are multiple text locations corresponding to the keyword, then the semantic scores corresponding to the multiple text locations are calculated; The text position corresponding to the maximum semantic score is taken as the text region corresponding to the evaluation item; Wherein, the step of calculating the semantic score corresponding to the multiple text positions if the keyword corresponds to multiple text positions includes: If there are multiple text positions corresponding to the keyword, then obtain the related words and standard sentence library corresponding to the keyword; Construct the first distribution feature matrix corresponding to the text position; Construct the second distribution feature matrix corresponding to each of the multiple standard statements in the standard statement library; Calculate the similarity between the first distribution feature matrix and the second distribution feature matrix; The maximum similarity among the multiple standard statements is taken as the semantic score; The step of constructing the first distribution feature matrix corresponding to the text position includes: Match the currently relevant words present at the text position, and calculate the first word code corresponding to the currently relevant words; Extract the first word spacing between keywords and multiple currently related words in the text location; wherein, the first word spacing refers to the positional difference of words in the text; Based on multiple first word codes and multiple first word intervals, a first distribution feature matrix is ​​constructed; wherein, the construction logic of the second distribution feature matrix is ​​the same as that of the first distribution feature matrix.

6. A terminal device, characterized in that, The terminal device includes: a memory, a processor, and an automatic evaluation program for enterprise bidding data stored in the memory and executable on the processor, wherein the automatic evaluation program for enterprise bidding data is configured to implement the steps in the automatic evaluation method for enterprise bidding data as described in any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps in the automatic evaluation method for enterprise bidding data as described in any one of claims 1 to 4.

Images