Big data analysis-based bidding and tendering comprehensive evaluation decision-making method

Abstract

The invention provides a bidding and tendering comprehensive evaluation decision-making method based on big data analysis, and belongs to the technical field of data analysis, and the method comprises the steps: obtaining a bidding document of bidding and tendering, and collecting the data of all bidders to determine first data; performing feature extraction on the bidding document to determine a feature vector, determining a performance value of each bidder based on the first data, and determining a score vector of each bidder based on the first data; and determining a bidding decision of the bidding and tendering based on the feature vector and the performance values and the score vectors of all bidders. The method can ensure the completeness and comprehensiveness of analysis data, quantitatively evaluate the comprehensive strength of each bidder, reduce subjective deviation, improve the fairness and accuracy of the bidding process, and improve the scientificity and rationality of decision making, thereby improving the success rate and quality of bidding projects.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to a bidding and tendering comprehensive evaluation decision-making method based on big data analysis. Background Art

[0002] Bidding decisions directly affect the cost-effectiveness of enterprises, the quality of projects, innovation capabilities, and the establishment of partnerships. In many technical fields, especially large-scale infrastructure projects or software development projects, reasonable bidding decisions can help enterprises or organizations complete projects at the lowest cost while ensuring quality; in technology-intensive fields such as information technology, artificial intelligence, medical technology, energy, etc., bidding can motivate bidders to provide innovative solutions, thereby promoting industry progress and technological innovation. However, bidding decisions have problems such as incomplete and inaccurate data, incomplete and incomplete evaluation of bidders, and opaque and unfair bidding processes.

[0003] Therefore, the present invention provides a bidding and tendering comprehensive evaluation decision-making method based on big data analysis. Summary of the invention

[0004] The present invention provides a bidding comprehensive evaluation decision method based on big data analysis, which extracts features from bidding documents to determine feature vectors, analyzes first data to determine the performance value and score vector of each bidder, determines the comprehensive scores of all bidders according to the feature vectors and the performance values ​​and score vectors of all bidders, and determines the bidding decision of the bidding, which can ensure the completeness and comprehensiveness of the analysis data, quantitatively evaluate the comprehensive strength of each bidder, reduce subjective bias, improve the fairness and accuracy of the bidding process, enhance the scientificity and rationality of decision-making, and thus improve the success rate and quality of bidding projects.

[0005] The present invention provides a bidding comprehensive evaluation decision method based on big data analysis, comprising:

[0006] 101: Obtain the bidding documents for the tender, and at the same time, collect data of all bidders to determine first data;

[0007] 102: extracting features from the bidding document to determine a feature vector, determining a performance value of each bidder based on the first data, and determining a scoring vector of each bidder based on the first data;

[0008] 103: Determine the bidding decision of the bidding based on the feature vector and the performance values ​​and score vectors of all bidders.

[0009] According to a bidding comprehensive evaluation decision-making method based on big data analysis provided by the present invention, the first data includes the current bidding documents and historical bidding data of each bidder, wherein the historical bidding data includes the historical bidding times of the corresponding bidder, all winning projects and the historical bidding documents, project scale, project completion and customer evaluation of each winning project.

[0010] According to a bidding comprehensive evaluation decision method based on big data analysis provided by the present invention, feature extraction is performed on the bidding document to determine a feature vector, including:

[0011] Performing text cleaning on the bidding document, performing word segmentation and part-of-speech tagging on the bidding document after the text cleaning to determine the first word, wherein the first word includes all the word segments of the bidding document and the part-of-speech tagging corresponding to each word segmentation;

[0012] Extract the segmentations related to the market and risk of the bidding in the first word, determine the market adjustment parameters based on the segmentations related to the market, and determine the risk adjustment parameters based on the segmentations related to the risk;

[0013] Extracting features from the bidding document based on the first word, determining the first feature and the bidding type, and ranking all features in the first feature in terms of importance;

[0014] Determine a feature vector based on the first feature after importance sorting;

[0015] F=[F1,…,F k ,…,F N3 ];

[0016] Among them, F represents the characteristic vector of the bidding, F1, F k 、F Ni They respectively represent the 1st feature, the kth feature, and the N3th feature in the first feature, and N3 represents the number of features in the first feature.

[0017] According to a bidding comprehensive evaluation decision method based on big data analysis provided by the present invention, the performance value of each bidder is determined based on first data, including:

[0018] Respectively extract the historical bidding documents of all the winning projects from the historical bidding data of each bidder in the first data;

[0019] Perform text cleaning on the historical bidding documents of each winning project of each bidder, and determine the bidding type of the corresponding winning project based on the historical bidding documents after text cleaning;

[0020] Based on the similarity between the bidding type of each winning project and the bidding type of the tender, the project importance value of the winning project is determined.

[0021] According to a bidding comprehensive evaluation decision method based on big data analysis provided by the present invention, the performance value of each bidder is determined based on the first data, and further includes:

[0022] Calculate the performance value of each bidder based on the historical bidding data of all bidders in the first data;

[0023]

[0024] A i =A1 i ×W1 i +A2 i ×W2 i ;

[0025] Among them, A1 i 、A2 i They represent the first performance value and the second performance value of bidder i, respectively, W1 i 、W2 i denote the first weight value and the second weight value of bidder i respectively, and A i represents the performance value of bidder i, N1 represents the number of bidders in the bidding, iN2 represents the number of successful bids of bidder i, D1 ij 、D2 ij S1 represents the first project scale and the second project scale of the j-th winning project of bidder i, ij 、S2 ij R1 represents the completion degree of the first project and the second project of the j-th winning project of bidder i. ij , R2 ij represents the first customer score and the second customer score of the j-th winning project of bidder i, σD1, σS1, σR1 represent the standard deviation of project scale, project completion and customer score of all winning projects of all bids, respectively, T ij represents the project completion time of the j-th winning project of bidder i, P i represents the number of bids of bidder i, represents the ratio of the number of successful bids of bidder i to the number of successful bids of all bidders, F ij 、V ij ,EC ij , AC ij They represent the project importance value, project revenue, estimated cost, and actual cost of the j-th winning project of bidder i, respectively. (AC ij -EC ij ) / EC ij represents the cost overrun rate of the j-th winning bid of bidder i.

[0026] According to a bidding comprehensive evaluation decision method based on big data analysis provided by the present invention, a scoring vector of each bidder is determined based on first data, including:

[0027] Performing text cleaning on the current bidding document of each bidder in the first data, and performing word segmentation and part-of-speech tagging on the current bidding document after the text cleaning to determine a corresponding second word, wherein the second word includes all the word segments of the corresponding current bidding document and the part-of-speech tagging corresponding to each word segmentation;

[0028] Match all the word segments in the second word of each bidder with each feature in the feature vector;

[0029] Calculate a score vector for each bidder based on the matching results;

[0030]

[0031] in, represents the score of bidder i based on the 1st feature, the kth feature, and the N3th feature, WN i represents the number of participles in the first word of bidder i, represents the participle a in the first word of bidder i that matches the kth feature, represents the matching degree between the first word of bidder i and the word a of the kth feature match, W b represents the first word of bidder b among all bidders other than bidder i, represents the number of first words that contain word a matching the kth feature in all first words except bidder i, ikN4 represents the number of all word matching the kth feature in the first word of bidder i, represents the frequency of all the words in the first word of bidder i that match the kth feature, represents the frequency of all the words in the first word of bidder i that match the kth feature in all other first words except bidder i, represents the correlation scores of all the word segments in the first word of bidder i that match the k-th feature and the k-th feature, Represents the anti-correlation scores of all the participles in the first word of bidder i that match the k-th feature in all other first words except bidder i and the k-th feature.

[0032] According to a bidding comprehensive evaluation decision method based on big data analysis provided by the present invention, the bidding decision of the bidding is determined based on a feature vector and the performance values ​​and scoring vectors of all bidders, including:

[0033] Calculate the corresponding comprehensive score based on the feature vector and the performance value and score vector of each bidder;

[0034]

[0035] OR i =A i ×(OR1 i -OR2 i );

[0036] Among them, OR i represents the comprehensive score of bidder i, OR1 i 、OR2 i denote the first sub-comprehensive score and the second sub-comprehensive score of bidder i, respectively, and M i represents the market adjustment parameter for bidder i, RA i represents the risk adjustment parameter of bidder i, N4 represents the adjustment of the first N4 features in the feature vector based on market data, represents the importance of the kth feature, and τ represents the importance adjustment parameter.

[0037] According to a bidding comprehensive evaluation decision method based on big data analysis provided by the present invention, the bidding decision of the bidding is determined based on the feature vector and the performance values ​​and scoring vectors of all bidders, and further includes:

[0038] Determine a comprehensive score for each bidder based on the feature vector and the performance values ​​and score vectors of all bidders;

[0039] A corresponding evaluation report is generated based on the performance value, the scoring vector and the comprehensive score of each bidder, and a bidding decision of the bidding is determined based on the evaluation report.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] By extracting features from the bidding documents to determine feature vectors, analyzing the first data to determine the performance value and score vector of each bidder, determining the comprehensive scores of all bidders based on the feature vectors and the performance values ​​and score vectors of all bidders, and determining the bidding decision of the bidding, it is possible to ensure the completeness and comprehensiveness of the analysis data, quantitatively evaluate the comprehensive strength of each bidder, reduce subjective bias, improve the fairness and accuracy of the bidding process, enhance the scientificity and rationality of decision-making, and thus improve the success rate and quality of bidding projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 It is a flow chart of a bidding and tendering comprehensive evaluation decision-making method based on big data analysis provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] Embodiment 1:

[0046] The embodiment of the present invention provides a bidding comprehensive evaluation decision method based on big data analysis, such as Figure 1 As shown, including:

[0047] 101: Obtain the bidding documents for the tender, and at the same time, collect data of all bidders to determine first data;

[0048] 102: extracting features from the bidding document to determine a feature vector, determining a performance value of each bidder based on the first data, and determining a scoring vector of each bidder based on the first data;

[0049] 103: Determine the bidding decision of the bidding based on the feature vector and the performance values ​​and score vectors of all bidders.

[0050] In this embodiment, the bidding document is an official notification document issued by the bidding party, which contains detailed information about the project bidding and provides important content such as bidding requirements, project overview, qualification conditions, timetable, technical specifications, submission method, etc.

[0051] In this embodiment, feature extraction is performed on the bidding documents to identify important evaluation indicators in the bidding project, such as technical requirements, project experience, price competitiveness, delivery time, risk management capabilities, etc., and a feature vector is constructed based on the extracted features.

[0052] In this embodiment, the performance value of each bidder is calculated based on the bidder's historical bidding data, and the scoring vector of each bidder is calculated based on the current bidding document and the feature vector. The scoring vector reflects the bidder's performance on each feature.

[0053] The beneficial effects of the above technical solution are as follows: by extracting features from the bidding documents to determine the feature vector, analyzing the first data to determine the performance value and scoring vector of each bidder, determining the comprehensive scores of all bidders based on the feature vectors and the performance values ​​and scoring vectors of all bidders, and determining the bidding decision of the bidding, the completeness and comprehensiveness of the analysis data can be ensured, the comprehensive strength of each bidder can be quantitatively evaluated, subjective bias can be reduced, the fairness and accuracy of the bidding process can be improved, the scientificity and rationality of the decision-making can be enhanced, and thus the success rate and quality of the bidding projects can be improved.

[0054] Embodiment 2:

[0055] An embodiment of the present invention provides a bidding and tendering comprehensive evaluation decision-making method based on big data analysis, wherein the first data includes each bidder's current bidding documents and historical bidding data, wherein the historical bidding data includes the corresponding bidder's historical bidding times, all winning projects, and the historical bidding documents of each winning project, project scale, project completion and customer evaluation.

[0056] In this embodiment, the historical bidding times record the number of biddings the bidder has participated in, reflecting the bidder's activity in the market and his / her experience accumulation.

[0057] In this embodiment, the historical bidding documents archive the bidder's bidding documents in the winning project, including technical solutions, quotation documents, etc.

[0058] In this embodiment, the project scale describes the scale of the project, such as project budget, engineering volume, service scope, etc. The project scale may be an amount, time, resources involved, or other quantifiable indicators.

[0059] In this embodiment, the project completion reflects whether the project is completed on time, on budget, and in accordance with quality requirements. The completion can be measured through multiple dimensions, such as completion time, budget deviation, quality compliance, etc.

[0060] In this embodiment, the customer evaluation collects customer feedback on the bidder's performance in the project, which may include customer satisfaction ratings, service evaluations, problem solving status, etc.

[0061] The beneficial effect of the above technical solution is: by determining the first data, a complete and comprehensive data basis is provided for determining the performance value and scoring vector of each bidder, which can improve the fairness and accuracy of the bidding process.

[0062] Embodiment 3:

[0063] The embodiment of the present invention provides a bidding comprehensive evaluation decision method based on big data analysis, which extracts features from bidding documents to determine feature vectors, including:

[0064] Perform text cleaning on the bidding documents, perform word segmentation and part-of-speech tagging on the bidding documents after text cleaning to determine the first word, where the first word includes all the word segments of the bidding documents and the corresponding part-of-speech tagging for each word segment;

[0065] Extract the word segments in the first word that are related to the market and risks of bidding, determine the market adjustment parameter based on the word segments related to the market, and at the same time, determine the risk adjustment parameter based on the word segments related to the risks;

[0066] Extract features from the bidding documents based on the first word, determine the first feature and the bidding type, and sort all the features in the first feature by importance;

[0067] Determine the feature vector based on the first feature sorted by importance;

[0068] F = [F1, …, F k , …, F N3 ;

[0069] where F represents the feature vector of bidding, and F1, F k , F N3 represent the 1st feature, the kth feature, and the N3rd feature in the first feature respectively, and N3 represents the number of features in the first feature.

[0070] In this embodiment, text cleaning means cleaning the bidding documents to remove irrelevant information, such as punctuation marks, stop words (such as "de", "le", etc.), format tags, etc.

[0071] In this embodiment, word segmentation means splitting a continuous Chinese text into individual words.

[0072] In this embodiment, word segmentation tools include Jieba: a Chinese word segmentation tool based on dictionaries and statistics, supporting accurate mode, full mode, and search engine mode; THULAC: a Chinese lexical analysis tool launched by Tsinghua University, with efficient word segmentation and part-of-speech tagging functions; HanLP: a deep learning natural language processing tool, supporting word segmentation, part-of-speech tagging, named entity recognition, etc.

[0073] In this embodiment, the word segmentation algorithm can be dictionary-based word segmentation: matching and segmenting the text according to a preset dictionary, suitable for processing more formal and common words; or statistical and machine learning word segmentation: using statistical models and machine learning techniques to segment words according to the word frequencies and probabilities appearing in the corpus, capable of processing more complex phrases and new words.

[0074] In this embodiment, part-of-speech tagging means tagging the corresponding part of speech for each word segment after word segmentation, such as noun (n), verb (v), adjective (a), etc.

[0075] In this embodiment, feature extraction includes keyword recognition: using keyword extraction algorithms (such as TF-IDF, TextRank) to identify high-frequency words or key phrases related to project scale, budget, technical requirements, etc.; entity recognition: applying natural language processing (NLP) technology to identify entities in the announcement (such as date, amount, location, organization name, etc.) to extract key information such as time requirements, budget range, qualification requirements, etc.; sentence parsing and dependency analysis: analyzing the structure of sentences and understanding how sentences in the bidding documents describe project requirements, qualification requirements, construction period and other information to ensure that the extracted features have correct semantic understanding.

[0076] In this embodiment, all sub-features in the first feature are ranked by importance through statistical analysis (such as correlation analysis, cluster analysis), the relationship between the extracted features is verified, and the importance is ranked according to the correlation between the features and the success of the project.

[0077] The beneficial effects of the above technical solution are as follows: by extracting features from the bidding documents to determine feature vectors, the bidding documents can be visualized, further improving the fairness and accuracy of the bidding process.

[0078] Embodiment 4:

[0079] The embodiment of the present invention provides a bidding comprehensive evaluation decision method based on big data analysis, which determines the performance value of each bidder based on first data, including:

[0080] Respectively extract the historical bidding documents of all the winning projects from the historical bidding data of each bidder in the first data;

[0081] Perform text cleaning on the historical bidding documents of each winning project of each bidder, and determine the bidding type of the corresponding winning project based on the historical bidding documents after text cleaning;

[0082] Based on the similarity between the bidding type of each winning project and the bidding type of the tender, the project importance value of the winning project is determined.

[0083] In this embodiment, the historical bidding data of each bidder includes one or more winning bid items and corresponding historical bidding documents.

[0084] In this embodiment, the corresponding bidding type is determined based on the keywords and phrases appearing in the historical bidding documents after the text cleaning.

[0085] In this embodiment, the similarity between the bidding type of each winning project and the bidding type of the current bidding project is determined, and the importance value of each winning project is determined to indicate the relevance or reference value of the winning project to the current bidding project.

[0086] The beneficial effects of the above technical solution are as follows: by determining the bidding type of each historical bidding document of each bidder and determining the project importance value of each winning project of all bidders, the data basis for determining the performance value of each bidder can be improved and subjective bias can be reduced.

[0087] Embodiment 5:

[0088] The embodiment of the present invention provides a bidding comprehensive evaluation decision method based on big data analysis, which determines the performance value of each bidder based on first data, and also includes:

[0089] Calculate the performance value of each bidder based on the historical bidding data of all bidders in the first data;

[0090]

[0091]

[0092] A i =A1 i ×W1 i +A2 i ×W2 i ;

[0093] Among them, A1 i 、A2 i They represent the first performance value and the second performance value of bidder i, respectively, W1 i 、W2 i denote the first weight value and the second weight value of bidder i respectively, and A i represents the performance value of bidder i, N1 represents the number of bidders in the bidding, iN2 represents the number of successful bids of bidder i, D1 ij 、D2 ij S1 represents the first project scale and the second project scale of the j-th winning project of bidder i, ij 、S2 ij R1 represents the completion degree of the first project and the second project of the j-th winning project of bidder i. ij , R2 ij represents the first customer score and the second customer score of the j-th winning project of bidder i, σD1, σS1, σR1 represent the standard deviation of project scale, project completion and customer score of all winning projects of all bids, respectively, T ij represents the project completion time of the j-th winning project of bidder i, P i represents the number of bids of bidder i, represents the ratio of the number of successful bids of bidder i to the number of successful bids of all bidders, F ij 、V ij ,ECij , AC ij They represent the project importance value, project revenue, estimated cost, and actual cost of the j-th winning project of bidder i, respectively. (AC ij -EC ij ) / EC ij represents the cost overrun rate of the j-th winning bid of bidder i.

[0094] In this embodiment, iN2 represents the number of successful bids of bidder i. For example, the number of successful bids of bidder 1 is 1N1, and the number of successful bids of bidder 4 is 4N1. The value range of i is 1 to N1. The number of successful bids of every two bidders may be the same or different.

[0095] In this embodiment, the second project scale indicates a value normalized according to the first project scales of all successful bid projects of all bidders.

[0096] In this embodiment, the second project completion degree represents a value normalized according to the first project completion degrees of all successful bids of all bidders.

[0097] In this embodiment, the second customer score represents a value normalized according to the first customer scores of all successful bids of all bidders.

[0098] In this embodiment, the project size standard deviation is calculated based on the first project size of all winning projects of all bidders and Calculated.

[0099] In this embodiment, the project completion standard deviation is calculated based on the first project completion of all winning projects of all bidders and Calculated.

[0100] In this embodiment, the customer score standard deviation is based on the first customer scores of all winning items of all bidders and Calculated.

[0101] In this embodiment, the first performance value A1 i It reflects the comprehensive impact of project completion time and cost overrun rate on project completion and customer ratings. The shorter the project completion time, the smaller the impact on project completion and customer ratings; conversely, if the completion time is longer, the impact is greater; the greater the cost overrun rate, the smaller the impact on project completion and customer ratings; conversely, if the completion time is longer, the impact is greater.

[0102] In this embodiment, the second performance value A2 i Reflects the negative impact of cost overrun rate on project importance and project benefits.

[0103] The beneficial effects of the above technical solution are as follows: by determining the performance values ​​of all bidders, the performance of bidders in all past bidding projects can be quantified, providing key data basis for determining the comprehensive scores of bidders, and quantitatively evaluating the comprehensive strength of each bidder, thereby improving the fairness and accuracy of the bidding process.

[0104] Embodiment 6:

[0105] The embodiment of the present invention provides a bidding comprehensive evaluation decision method based on big data analysis, which determines a scoring vector for each bidder based on first data, including:

[0106] Performing text cleaning on the current bidding document of each bidder in the first data, and performing word segmentation and part-of-speech tagging on the current bidding document after the text cleaning to determine a corresponding second word, wherein the second word includes all the word segments of the corresponding current bidding document and the part-of-speech tagging corresponding to each word segmentation;

[0107] Match all the word segments in the second word of each bidder with each feature in the feature vector;

[0108] Calculate a score vector for each bidder based on the matching results;

[0109]

[0110] in, represents the score of bidder i based on the 1st feature, the kth feature, and the N3th feature, WN i represents the number of participles in the first word of bidder i, represents the participle a in the first word of bidder i that matches the kth feature, represents the matching degree between the first word of bidder i and the word a of the kth feature match, W b represents the first word of bidder b among all bidders other than bidder i, represents the number of first words that contain word a matching the kth feature in all first words except bidder i, ikN4 represents the number of all word matching the kth feature in the first word of bidder i, represents the frequency of all the words in the first word of bidder i that match the kth feature, represents the frequency of all the words in the first word of bidder i that match the kth feature in all other first words except bidder i, represents the correlation scores of all the words matching the k-th feature in the first word of bidder i and the k-th feature, Represents the anti-correlation scores of all the participles in the first word of bidder i that match the k-th feature in all other first words except bidder i and the k-th feature.

[0111] In this embodiment, the calculation method of the correlation scores of all the participles matching the kth feature and the kth feature in the first word of bidder i is as follows:

[0112] In this embodiment, the current bidding document of each bidder corresponds to a second word, and the second word includes multiple word segments of the current bidding documents and corresponding part-of-speech tags.

[0113] In this embodiment, the current bidding document after each bidder's text cleaning is parsed, and its content is segmented using natural language processing (NLP) technology to determine the second word and the keywords or key phrases in the second word, and the segmented words of each bidder (extracted from the second word) are matched one by one with each feature in the feature vector.

[0114] In this embodiment, the score vector of the bidder includes the score value of each feature in the feature vector of the corresponding bidder.

[0115] In this embodiment, each bidder pair uses a score vector, and the dimension of the score vector is the same as the dimension of the feature vector.

[0116] The beneficial effects of the above technical solution are as follows: by determining the scoring vector of each bidder based on the first data, the current bidding document of the bidder can be quantified, providing key data basis for determining the comprehensive score of the bidder, which can comprehensively reflect the comprehensive strength of each bidder and further improve the fairness and accuracy of the bidding process.

[0117] Embodiment 7:

[0118] The embodiment of the present invention provides a bidding comprehensive evaluation decision method based on big data analysis, which determines the bidding decision of the bidding based on a feature vector and the performance values ​​and scoring vectors of all bidders, including:

[0119] Calculate the corresponding comprehensive score based on the feature vector and the performance value and score vector of each bidder;

[0120]

[0121] OR i =A i ×(OR1 i -OR2 i );

[0122] Among them, OR i represents the comprehensive score of bidder i, OR1 i 、OR2 i denote the first sub-comprehensive score and the second sub-comprehensive score of bidder i, respectively, and M i represents the market adjustment parameter for bidder i, RAi represents the risk adjustment parameter of bidder i, N4 represents the adjustment of the first N4 features in the feature vector based on market data, represents the importance of the kth feature, and τ represents the importance adjustment parameter.

[0123] In this embodiment, the first sub-comprehensive score represents the sub-score value after adjusting the first N4 features in the feature vector based on market data, and the second sub-comprehensive score represents the sub-score value after adjusting the last N4+1 features in the feature vector based on risk data.

[0124] In this embodiment, the features are divided into two sections. The first N4 features are scored through market adjustment, and the last N3-N4 features are scored through risk adjustment. The ratio of key features to risk features can be flexibly adjusted, and the size of N4 can be adjusted according to specific business scenarios.

[0125] In this embodiment, It means that in the feature vector, the smaller the k value of the kth feature, the greater the importance of the corresponding feature. For example, when N3=7, the importance of the 7th feature is smaller, and the importance of the 3rd feature is greater than that of the 5th feature.

[0126] In this embodiment, the importance adjustment parameter controls the speed at which the importance of features in the feature vector decreases, and the importance of the kth feature decreases as k increases.

[0127] In this embodiment, the comprehensive score of each bidder is used to evaluate the overall strength of the corresponding bidder.

[0128] The beneficial effects of the above technical solution are as follows: the corresponding comprehensive score is determined according to the feature vector and the performance value and scoring vector of each bidder, which can effectively integrate the feature vector and the performance value and scoring vector of each bidder, making the bidding decision more transparent and data-driven, and improving the scientificity and rationality of the decision.

[0129] Embodiment 8:

[0130] The embodiment of the present invention provides a bidding comprehensive evaluation decision method based on big data analysis, which determines the bidding decision of the bidding based on the feature vector and the performance values ​​and scoring vectors of all bidders, and also includes:

[0131] Determine a comprehensive score for each bidder based on the feature vector and the performance values ​​and score vectors of all bidders;

[0132] A corresponding evaluation report is generated based on the performance value, the scoring vector and the comprehensive score of each bidder, and a bidding decision of the bidding is determined based on the evaluation report.

[0133] In this embodiment, the bidder's evaluation report includes the bidder's basic information, performance value, scoring vector, comprehensive score, comprehensive score ranking, and recommendations for bidding decisions based on the comprehensive score, clarifying the bidder's strengths and weaknesses.

[0134] In this embodiment, based on the comprehensive score in the evaluation report, the tenderer can make more transparent and data-driven decisions, bidders with higher comprehensive scores will be given priority, and the tender committee can discuss based on the comprehensive scores of all bidders and ultimately select the successful bidder.

[0135] The beneficial effects of the above technical solution are as follows: determining the bidding decision of the bidding according to the feature vector and the performance values, scoring vectors and comprehensive scores of all bidders can reduce subjective bias, improve the fairness and accuracy of the bidding process, enhance the scientificity and rationality of the decision, and thus improve the success rate and quality of the bidding projects.

[0136] The method embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative work.

[0137] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0138] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A bidding comprehensive evaluation decision method based on big data analysis, characterized in that: include: 101: Obtain the bidding documents for the tender, and at the same time, collect data of all bidders to determine first data; 102: extracting features from the bidding document to determine a feature vector, determining a performance value of each bidder based on the first data, and determining a scoring vector of each bidder based on the first data; 103: Determine the bidding decision of the bidding based on the feature vector and the performance values ​​and score vectors of all bidders.

2. According to claim 1, a bidding comprehensive evaluation decision-making method based on big data analysis is characterized in that: The first data includes each bidder's current bidding documents and historical bidding data, wherein the historical bidding data includes the corresponding bidder's historical bidding times, all winning projects, and historical bidding documents, project scale, project completion and customer evaluation of each winning project.

3. The bidding and tendering comprehensive evaluation decision-making method based on big data analysis according to claim 2 is characterized in that: Extract features from the bidding documents to determine feature vectors, including: Performing text cleaning on the bidding document, performing word segmentation and part-of-speech tagging on the bidding document after the text cleaning to determine the first word, wherein the first word includes all the word segments of the bidding document and the part-of-speech tagging corresponding to each word segmentation; Extract the segmentations related to the market and risk of the bidding in the first word, determine the market adjustment parameters based on the segmentations related to the market, and determine the risk adjustment parameters based on the segmentations related to the risk; Extracting features from the bidding document based on the first word, determining the first feature and the bidding type, and ranking all features in the first feature in terms of importance; Determine a feature vector based on the first feature after importance sorting; F=[F1,…,F k ,…,F N3 ]; Among them, F represents the characteristic vector of the bidding, F1, F k 、F Ni They respectively represent the 1st feature, the kth feature, and the N3th feature in the first feature, and N3 represents the number of features in the first feature.

4. According to claim 2, a bidding and tendering comprehensive evaluation decision-making method based on big data analysis is characterized in that: Determining a performance value of each bidder based on the first data includes: Respectively extract the historical bidding documents of all the winning projects from the historical bidding data of each bidder in the first data; Perform text cleaning on the historical bidding documents of each winning project of each bidder, and determine the bidding type of the corresponding winning project based on the historical bidding documents after text cleaning; Based on the similarity between the bidding type of each winning project and the bidding type of the tender, the project importance value of the winning project is determined.

5. According to claim 4, a bidding comprehensive evaluation decision-making method based on big data analysis is characterized in that: Determining the performance value of each bidder based on the first data further includes: Calculate the performance value of each bidder based on the historical bidding data of all bidders in the first data; <h2 style=";text-align:left;direction:ltr">A<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> =A1<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> ×W1<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> +A2<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> ×W2<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> ; Among them, A1 i 、A2 i They represent the first performance value and the second performance value of bidder i, respectively, W1 i 、W2 i denote the first weight value and the second weight value of bidder i respectively, and A i represents the performance value of bidder i, N1 represents the number of bidders in the bidding, iN2 represents the number of successful bids of bidder i, D1 ij 、D2 ij S1 represents the first project scale and the second project scale of the j-th winning project of bidder i, ij 、S2 ij R1 represents the completion degree of the first project and the second project of the j-th winning project of bidder i. ij , R2 ij represents the first customer score and the second customer score of the j-th winning project of bidder i, σD1, σS1, σR1 represent the standard deviation of project scale, project completion and customer score of all winning projects of all bids, respectively, T ij represents the project completion time of the j-th winning project of bidder i, P i represents the number of bids of bidder i, represents the ratio of the number of successful bids of bidder i to the number of successful bids of all bidders, F ij 、V ij ,EC ij , AC ij They represent the project importance value, project revenue, estimated cost, and actual cost of the j-th winning project of bidder i, respectively. (AC ij -EC ij ) / EC ij represents the cost overrun rate of the j-th winning bid of bidder i.

6. The bidding and tendering comprehensive evaluation decision-making method based on big data analysis according to claim 3 is characterized in that: Determining a score vector for each bidder based on the first data includes: Performing text cleaning on the current bidding document of each bidder in the first data, and performing word segmentation and part-of-speech tagging on the current bidding document after the text cleaning to determine a corresponding second word, wherein the second word includes all the word segments of the corresponding current bidding document and the part-of-speech tagging corresponding to each word segmentation; Match all the word segments in the second word of each bidder with each feature in the feature vector; Calculate a score vector for each bidder based on the matching results; in, represents the score of bidder i based on the 1st feature, the kth feature, and the N3th feature, WN i represents the number of participles in the first word of bidder i, represents the participle a in the first word of bidder i that matches the kth feature, represents the matching degree between the first word of bidder i and the word a of the kth feature match, W b represents the first word of bidder b among all bidders other than bidder i, represents the number of first words that contain word a matching the kth feature in all first words except bidder i, ikN4 represents the number of all word matching the kth feature in the first word of bidder i, represents the frequency of all the words in the first word of bidder i that match the kth feature, represents the frequency of all the words in the first word of bidder i that match the kth feature in all other first words except bidder i, represents the correlation scores of all the word segments in the first word of bidder i that match the k-th feature and the k-th feature, Represents the anti-correlation scores of all the participles in the first word of bidder i that match the k-th feature in all other first words except bidder i and the k-th feature.

7. The bidding and tendering comprehensive evaluation decision-making method based on big data analysis according to claim 6 is characterized in that: The bidding decision of the bidding is determined based on the feature vector and the performance values ​​and scoring vectors of all bidders, including: Calculate the corresponding comprehensive score based on the feature vector and the performance value and score vector of each bidder; OR i =A i ×(OR1 i -OR2 i ); Among them, OR i represents the comprehensive score of bidder i, OR1 i 、OR2 i denote the first sub-comprehensive score and the second sub-comprehensive score of bidder i, respectively, and M i represents the market adjustment parameter for bidder i, RA i represents the risk adjustment parameter of bidder i, N4 represents the adjustment of the first N4 features in the feature vector based on market data, represents the importance of the kth feature, and τ represents the importance adjustment parameter.

8. The bidding and tendering comprehensive evaluation decision-making method based on big data analysis according to claim 1 is characterized in that: Determining the bidding decision based on the feature vector and the performance values ​​and scoring vectors of all bidders also includes: Determine a comprehensive score for each bidder based on the feature vector and the performance values ​​and score vectors of all bidders; A corresponding evaluation report is generated based on the performance value, the scoring vector and the comprehensive score of each bidder, and a bidding decision of the bidding is determined based on the evaluation report.

Images