Intelligent generation system and method for personalized bidding and tendering files

Abstract

The invention discloses a personalized tendering and bidding file intelligent generation system and method, and the system comprises a model generation module which is used for generating a tendering and bidding model according to target text data and preset tendering and bidding contents; the construction module is used for constructing an association structure chart which comprises a plurality of concerned nodes, and the concerned nodes contain concerned text data; the bid evaluation module is used for analyzing the attention text data according to a plurality of bid evaluation indexes and generating bid evaluation values corresponding to the attention nodes according to analysis results; the adjustment module is used for judging whether adjustment is needed or not according to the bid evaluation value, and if yes, an adjustment instruction is generated according to the bid evaluation value and a user-defined function; and the file generation module is used for carrying out analogue simulation on the plurality of adjustment instructions, judging whether the adjustment instructions are adjusted again or not, if not, adjusting the corresponding attention nodes according to the adjustment instructions, and generating bidding and tendering files, so that the completeness and normalization of the bidding and tendering files are ensured, and the file generation efficiency is improved.

Description

Technical Field

[0001] The present application relates to the technical field of bidding documents, and in particular to a system and method for intelligently generating personalized bidding documents. Background Art

[0002] The bidding documents are the outline of the bidding project construction and the basis for the construction unit to implement the project construction. Therefore, the quality and depth of the bidding documents are related to the success or failure of the entire bidding work.

[0003] In the prior art, bidders usually review and modify the relevant information in the bidding templates, which is time-consuming and labor-intensive. In addition, the bidding templates are rich in content and prone to errors or omissions, resulting in incomplete or non-standard final bidding documents, thereby reducing the quality of the bidding documents and the efficiency of document generation. Summary of the invention

[0004] To solve the above technical problems, the present application provides a personalized bidding document intelligent generation system and method, which extracts target text data according to user needs, generates bidding templates, and constructs an association structure diagram of the bidding template. The association structure diagram includes multiple focus nodes, and each focus node has multiple focus text data. The focus text data of the focus node is analyzed according to multiple evaluation indicators, and an evaluation value of each focus node is generated. Adjustment instructions are generated according to the evaluation value and user-defined functions, and each detail of the bidding template is accurately evaluated, so as to formulate reasonable adjustment instructions, ensure the integrity and standardization of the bidding documents, and improve the efficiency of document generation.

[0005] In some embodiments of the present application, a personalized bidding document intelligent generation system is provided, the system comprising:

[0006] A template generation module is used to extract target text data from historical bidding documents according to user needs, and generate a bidding template according to the target text data and preset bidding content;

[0007] A construction module is used to construct a correlation structure diagram of the bidding template, wherein the correlation structure diagram includes a plurality of attention nodes and each of the association nodes contains a plurality of attention text data;

[0008] The bid evaluation module is used to pre-set multiple bid evaluation indicators, analyze the concerned text data of the concerned node according to the multiple bid evaluation indicators, and generate the bid evaluation value of the corresponding concerned node according to the analysis result;

[0009] The adjustment module is used to determine whether adjustment is needed according to the bid evaluation value. If necessary, multiple adjustment instructions corresponding to the concerned nodes are generated according to the bid evaluation value and user-defined functions;

[0010] The file generation module is used to simulate multiple adjustment instructions, determine whether to adjust the adjustment instructions again based on the simulation results, and if not, adjust the corresponding focus nodes according to the adjustment instructions and generate bidding documents.

[0011] In some embodiments of the present application, a bidding template is generated according to target text data and preset bidding content, including:

[0012] Determine bidding-related information and historical bidding document information of the corresponding user according to user needs;

[0013] Screening historical bidding documents in the bidding document library according to bidding related information and historical bidding document information, and obtaining historical text data in the historical bidding documents whose correlation with user needs is greater than a preset correlation threshold;

[0014] Sort the correlation degree of the selected historical text data with the same business demand attribute, and set the historical text data with the highest ranking as the target text data corresponding to the demand attribute;

[0015] The corresponding positions in the preset bidding content are dynamically filled according to multiple target text data to generate a bidding template.

[0016] In some embodiments of the present application, a correlation structure diagram of a bidding template is constructed, including:

[0017] Divide the bidding template into a plurality of focus modules according to the functional type of the bidding template, construct an initial structure diagram according to the logical relationship of the corresponding functional type of the focus module, set the maximum importance of the functional type of the focus module as the structure center of the initial structure diagram, and set the connection distance according to the logical relationship between the structure center and other focus modules;

[0018] Analyze the text data contained in each module of interest in the initial structure diagram, obtain the utilization coefficient and influence coefficient of the main functional type of the text data in the corresponding module of interest, generate the importance coefficient based on the utilization coefficient and the influence coefficient, and determine the characteristic text data of each module of interest according to the importance coefficient;

[0019] Based on the comparison relationship between the important coefficient of the feature text data and the preset important coefficient threshold, the text data of the corresponding focus module is grouped to obtain multiple text data groups of each focus module, and each text data group is set as a focus node;

[0020] Analyze the dependency and association relationship between the feature text data of multiple focus nodes of the same focus module, determine the node center of the corresponding focus module according to the dependency and association relationship, set the connection distance between the node center and the remaining focus nodes of the same focus module according to the dependency degree and association degree, and generate a connection structure diagram of the corresponding focus module;

[0021] Import the connection structure diagram of each module of interest into the corresponding position of the initial structure diagram to form the associated structure diagram of the bidding template.

[0022] In some embodiments of the present application, each focus node contains multiple focus text data, including:

[0023] Performing a first correlation analysis on the remaining text data of each focus node and the characteristic text data of the node center of the focus module of other functional types, and performing a second correlation analysis on the remaining text data of each focus node and the characteristic text data of the remaining focus nodes excluding the node center of the focus module of other functional types, to obtain a plurality of first correlation values ​​and a plurality of second correlation values;

[0024] Analyze the multiple first correlation values ​​to obtain the number of first correlation values ​​greater than the first preset correlation threshold and the first correlation difference value of the first correlation value greater than the first preset correlation threshold to generate a first attention coefficient corresponding to the remaining text data;

[0025] Analyze the plurality of second correlation values ​​to obtain the number of second correlation values ​​greater than a second preset correlation threshold and a second correlation difference value greater than the second preset correlation threshold to generate a second attention coefficient corresponding to the remaining text data;

[0026] Generate a compensation coefficient according to the connection distance length between the remaining text data and the feature text data of the node center of the corresponding attention module;

[0027] A comprehensive attention coefficient of the remaining text data is generated according to the compensation coefficient, the first attention coefficient, and the second attention coefficient.

[0028] In some embodiments of the present application, each focus node contains a plurality of focus text data, and further includes:

[0029] The calculation formula of the comprehensive attention coefficient is:

[0030]

[0031] Among them, Z is the comprehensive attention coefficient, b is the compensation coefficient, g is the conversion coefficient, is the first attention coefficient, n is the number of first correlation values ​​greater than the first preset correlation threshold, △p1i is the ith first correlation difference, yi is the weight coefficient of the ith first correlation difference, a1 is the weight coefficient of the first attention coefficient, is the second attention coefficient, m is the number of second association values ​​greater than the second preset association threshold, ys is the weight coefficient of the sth second association difference value, △p2s is the sth second association difference value;

[0032] Pre-set the attention factor threshold;

[0033] The remaining text data whose comprehensive attention coefficient is greater than the attention coefficient threshold and the characteristic text data of the same attention node are set as the attention text data of the corresponding attention node.

[0034] In some embodiments of the present application, generating a bid evaluation value corresponding to a node of interest according to the analysis result includes:

[0035] Obtain a historical database for each bid evaluation indicator, analyze historical text data in the historical database, and determine multiple historical data types, historical data functions, and historical characteristic characters corresponding to the bid evaluation indicator;

[0036] Obtain the real-time data type, real-time data function and real-time characteristic character of the concerned text data of each concerned node, and compare them with the historical data type, historical data function and historical characteristic character of each bid evaluation indicator to obtain the first metric threshold, the second metric threshold and the third metric threshold respectively;

[0037] Generate a metric value mean of the concerned text data and the corresponding bid evaluation index according to the first metric value threshold, the second metric value threshold and the third metric value threshold, and set the concerned text data whose metric value mean is greater than the preset metric value threshold as the bid evaluation text data of the corresponding bid evaluation index;

[0038] Inputting the bid evaluation text data of each bid evaluation indicator of the same focus node into the preset bid evaluation model of the corresponding bid evaluation indicator to obtain the indicator evaluation values ​​of multiple bid evaluation indicators of the same focus node;

[0039] The bid evaluation value of the corresponding focus node is generated according to the indicator evaluation values ​​and weight coefficients of multiple bid evaluation indicators of the same focus node.

[0040] In some embodiments of the present application, generating a plurality of adjustment instructions corresponding to the concerned nodes includes:

[0041] Pre-set bid evaluation value thresholds;

[0042] If the bid evaluation value of the concerned node is less than the bid evaluation value threshold, it needs to be adjusted. The concerned text data that needs to be adjusted for the corresponding concerned node is screened out according to the indicator evaluation value of each bid evaluation indicator, and multiple adjustment strategies for the concerned text data are automatically generated;

[0043] The attention text data that needs to be adjusted and the corresponding adjustment strategies are sent to the user end, and the attention text data that needs to be adjusted and the corresponding multiple adjustment strategies are optimized according to the user-defined function, and multiple adjustment instructions corresponding to the attention nodes are generated according to the optimized multiple adjustment strategies.

[0044] In some embodiments of the present application, multiple adjustment instructions are simulated, including:

[0045] Obtain the location information and connection structure information of each node of interest in the association structure diagram of the bidding template, and generate a static simulation structure;

[0046] Acquire the dynamic simulation information to be set according to the adjustment requirements of the adjustment instructions;

[0047] Acquire the structural relationship of each concerned text data to be adjusted in the dynamic simulation information to be set, and establish the dynamic relationship of each concerned text node under the same adjustment instruction;

[0048] The dynamic relationship is linked to the static simulation structure in position, and the simulation of the corresponding adjustment instruction is realized to obtain the simulation result of the corresponding adjustment instruction.

[0049] In some embodiments of the present application, judging whether to adjust the adjustment instruction again according to the simulation result includes:

[0050] The simulation result includes the adjustment text data of the concerned text data to be adjusted in each adjustment instruction and the influence change amount on the concerned text data not to be adjusted;

[0051] Re-evaluate the bid evaluation index corresponding to the adjustment text data to obtain the simulation index evaluation value. If the simulation index evaluation value is greater than the preset index evaluation value threshold, and the impact change amount is less than the preset change amount threshold, there is no need to adjust the adjustment instruction again. Adjust the corresponding focus nodes according to the adjustment instruction and generate the bidding documents.

[0052] If the simulation index evaluation value is less than the preset index evaluation value threshold, and there is an impact change greater than the preset change threshold, the corresponding adjustment instruction is fine-tuned.

[0053] In some embodiments of the present application, a personalized bidding document intelligent generation method is also included:

[0054] Extract target text data from historical bidding documents according to user needs, and generate bidding templates based on the target text data and preset bidding content;

[0055] Constructing a correlation structure diagram of a bidding template, wherein the correlation structure diagram includes a plurality of focus nodes and each correlation node contains a plurality of focus text data;

[0056] Preset multiple bid evaluation indicators, analyze the concerned text data of the concerned node according to the multiple bid evaluation indicators, and generate the bid evaluation value of the corresponding concerned node according to the analysis result;

[0057] Determine whether adjustment is needed based on the bid evaluation value. If necessary, generate multiple adjustment instructions for corresponding nodes of interest based on the bid evaluation value and user-defined functions;

[0058] Simulate multiple adjustment instructions, and determine whether to adjust the adjustment instructions again based on the simulation results. If not, adjust the corresponding focus nodes according to the adjustment instructions and generate bidding documents.

[0059] Compared with the prior art, the personalized bidding document intelligent generation system and method of the embodiment of the present application has the following beneficial effects:

[0060] The target text data is extracted according to user needs, and a bidding template is generated, and an association structure diagram of the bidding template is constructed. The association structure diagram includes multiple focus nodes, and each focus node has multiple focus text data. The focus text data of the focus node is analyzed according to multiple evaluation indicators, and an evaluation value of each focus node is generated. Adjustment instructions are generated according to the evaluation value and user-defined functions, and every detail of the bidding template is accurately evaluated, so as to formulate reasonable adjustment instructions, ensure the integrity and standardization of the bidding documents, and improve the efficiency of document generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a schematic diagram of a personalized bidding document intelligent generation system in a preferred embodiment of the present application;

[0062] Figure 2 It is a flow chart of a method for intelligently generating personalized bidding documents in a preferred embodiment of the present application. DETAILED DESCRIPTION

[0063] The specific implementation methods of the present application are further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application but are not intended to limit the scope of the present application.

[0064] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0065] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0066] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0067] like Figure 1 As shown, a personalized bidding document intelligent generation system according to a preferred embodiment of the present application includes:

[0068] A template generation module is used to extract target text data from historical bidding documents according to user needs, and generate a bidding template according to the target text data and preset bidding content;

[0069] A construction module is used to construct a correlation structure diagram of a bidding template, wherein the correlation structure diagram of the attention module includes a plurality of attention nodes and each of the association nodes contains a plurality of attention text data;

[0070] The bid evaluation module is used to pre-set multiple bid evaluation indicators, analyze the concerned text data of the concerned node according to the multiple bid evaluation indicators, and generate the bid evaluation value of the corresponding concerned node according to the analysis result;

[0071] The adjustment module is used to determine whether adjustment is needed according to the bid evaluation value. If necessary, multiple adjustment instructions corresponding to the concerned nodes are generated according to the bid evaluation value and user-defined functions;

[0072] The file generation module is used to simulate multiple adjustment instructions, determine whether to adjust the adjustment instructions again based on the simulation results, and if not, adjust the corresponding focus nodes according to the adjustment instructions and generate bidding documents.

[0073] In this embodiment, the preset bidding content includes common formats such as bidding announcements, bidding sample templates, and bidding instructions.

[0074] In this embodiment, target text data is extracted according to user needs, and a bidding template is generated, and an association structure diagram of the bidding template is constructed. The association structure diagram includes multiple focus nodes, and each focus node has multiple focus text data. The focus text data of the focus node is analyzed according to multiple bid evaluation indicators, and an evaluation value of each focus node is generated. Adjustment instructions are generated according to the bid evaluation value and user-defined functions, and every detail of the bidding template is accurately evaluated, so as to formulate reasonable adjustment instructions, ensure the integrity and standardization of the bidding documents, and improve the efficiency of document generation.

[0075] In some embodiments of the present application, a bidding template is generated according to target text data and preset bidding content, including:

[0076] Determine bidding-related information and historical bidding document information of the corresponding user according to user needs;

[0077] Screening historical bidding documents in the bidding document library according to bidding related information and historical bidding document information, and obtaining historical text data in the historical bidding documents whose correlation with user needs is greater than a preset correlation threshold;

[0078] Sort the correlation degree of the selected historical text data with the same business demand attribute, and set the historical text data with the highest ranking as the target text data corresponding to the demand attribute;

[0079] The corresponding positions in the preset bidding content are dynamically filled according to multiple target text data to generate a bidding template.

[0080] In this embodiment, the bidding related information includes the user's bidding business type, business project, business components, etc. The historical bidding document information refers to the user's historical bidding documents, the evaluation values ​​of the historical bidding documents, historical bidding requirements, etc. The business requirement attributes refer to the set of specific requirements and characteristics that the business needs to meet.

[0081] In this embodiment, multiple target text data are screened out according to user needs, and the target text data are filled into the corresponding position of the preset bidding content, and combined with the preset bidding content, a bidding template that meets the user's needs is generated, which provides a basis for the subsequent generation of bidding documents and improves the efficiency of document generation.

[0082] In some embodiments of the present application, a correlation structure diagram of a bidding template is constructed, including:

[0083] Divide the bidding template into a plurality of focus modules according to the functional type of the bidding template, construct an initial structure diagram according to the logical relationship of the corresponding functional type of the focus module, set the maximum importance of the functional type of the focus module as the structure center of the initial structure diagram, and set the connection distance according to the logical relationship between the structure center and other focus modules;

[0084] Analyze the text data contained in each module of interest in the initial structure diagram, obtain the utilization coefficient and influence coefficient of the main functional type of the text data in the corresponding module of interest, generate the importance coefficient based on the utilization coefficient and the influence coefficient, and determine the characteristic text data of each module of interest according to the importance coefficient;

[0085] Based on the comparison relationship between the important coefficient of the feature text data and the preset important coefficient threshold, the text data of the corresponding focus module is grouped to obtain multiple text data groups of each focus module, and each text data group is set as a focus node;

[0086] Analyze the dependency and association relationship between the feature text data of multiple focus nodes of the same focus module, determine the node center of the corresponding focus module according to the dependency and association relationship, set the connection distance between the node center and the remaining focus nodes of the same focus module according to the dependency degree and association degree, and generate a connection structure diagram of the corresponding focus module;

[0087] Import the connection structure diagram of each module of interest into the corresponding position of the initial structure diagram to form the associated structure diagram of the bidding template.

[0088] In this embodiment, the functional types include cover and catalog, qualification type, business type, technical type, and attachment and certification type. The qualification type refers to the display of the legality and compliance of the bidder. The business type is the commercial terms such as indicator quotation, payment method, after-sales service, etc. The technical type refers to the technical solution, process flow, equipment selection and other contents of the project. The attachment and certification type is used to prove the authenticity and reliability of the content stated by the bidder.

[0089] In this embodiment, the utilization coefficient refers to the frequency of use of text data in historical bidding documents when evaluating corresponding function types, and the impact coefficient refers to the degree of influence of text data in historical bidding documents when evaluating corresponding function types. The importance coefficient = utilization coefficient * 0.4 + impact coefficient * 0.6.

[0090] In this embodiment, the dependency relationship refers to the change of feature text data causing the change of other feature text data, the association relationship refers to the mutual influence between feature text data, the degree of dependence refers to the degree of change of other feature text data caused by the change of feature text data, and the degree of association refers to the degree of mutual influence between feature text data.

[0091] In this embodiment, the preset importance coefficient threshold is set in advance, the sum of the importance coefficients of adjacent feature text data is calculated, and the feature text data whose sum of the importance coefficients is greater than the preset importance coefficient threshold is divided, and the text data included in the feature text data whose sum of the importance coefficients is less than the preset importance coefficient threshold is divided into the same text data group, thereby reducing the amount of data analysis for each focus node, improving the accuracy of data analysis, laying the foundation for the subsequent calculation of the bid evaluation value, and improving the accuracy of the bid evaluation value.

[0092] In this embodiment, the general framework of the initial structure diagram of the bidding template is constructed according to the functional type, and the important coefficients of the text data in each focus module are calculated to obtain the characteristic text data of each focus module, and a connection structure diagram of the focus nodes of each focus module is generated and imported into the position of the focus module in the initial structure diagram, and then refined to the connection structure of the focus nodes of different focus modules to obtain the associated structure diagram of the bidding template. The bidding template is refined in multiple layers to obtain a precise structure diagram, and each detail is evaluated to quickly determine the evaluation value of each focus node, so as to formulate precise adjustment instructions, ensure the integrity and standardization of the bidding documents, and improve the efficiency of document generation.

[0093] In some embodiments of the present application, each focus node contains multiple focus text data, including:

[0094] Performing a first correlation analysis on the remaining text data of each focus node and the characteristic text data of the node center of the focus module of other functional types, and performing a second correlation analysis on the remaining text data of each focus node and the characteristic text data of the remaining focus nodes excluding the node center of the focus module of other functional types, to obtain a plurality of first correlation values ​​and a plurality of second correlation values;

[0095] Analyze the multiple first correlation values ​​to obtain the number of first correlation values ​​greater than the first preset correlation threshold and the first correlation difference value of the first correlation value greater than the first preset correlation threshold to generate a first attention coefficient corresponding to the remaining text data;

[0096] Analyze the plurality of second correlation values ​​to obtain the number of second correlation values ​​greater than a second preset correlation threshold and a second correlation difference value greater than the second preset correlation threshold to generate a second attention coefficient corresponding to the remaining text data;

[0097] Generate a compensation coefficient according to the connection distance length between the remaining text data and the feature text data of the node center of the corresponding attention module;

[0098] A comprehensive attention coefficient of the remaining text data is generated according to the compensation coefficient, the first attention coefficient, and the second attention coefficient.

[0099] In this embodiment, the remaining text data refers to the data of the text data group in each focus node excluding the characteristic text data. The comprehensive attention coefficient of the corresponding text data is calculated based on the first association value of the remaining text data to the characteristic text data of the node center of other functional types and the second association value of the characteristic text data to the remaining focus nodes of other functional types, so as to judge the degree of association and influence of the remaining text data on other functional types, accurately screen out the focus text data of each focus node, lay a data foundation for the subsequent calculation of the evaluation value of each focus node, and improve the integrity and standardization of the bidding documents, as well as the efficiency of document generation.

[0100] In some embodiments of the present application, each focus node contains a plurality of focus text data, and further includes:

[0101] The calculation formula of the comprehensive attention coefficient is:

[0102]

[0103] Among them, Z is the comprehensive attention coefficient, b is the compensation coefficient, g is the conversion coefficient, is the first attention coefficient, n is the number of first correlation values ​​greater than the first preset correlation threshold, △p1i is the ith first correlation difference, yi is the weight coefficient of the ith first correlation difference, a1 is the weight coefficient of the first attention coefficient, is the second attention coefficient, m is the number of second association values ​​greater than the second preset association threshold, ys is the weight coefficient of the sth second association difference value, △p2s is the sth second association difference value;

[0104] Pre-set the attention factor threshold;

[0105] The remaining text data whose comprehensive attention coefficient is greater than the attention coefficient threshold and the characteristic text data of the same attention node are set as the attention text data of the corresponding attention node.

[0106] In this embodiment, the weight coefficient of the first association difference is set according to the influence coefficient of the characteristic text data corresponding to the first association value, and the compensation coefficient is set according to the comparison relationship between the connection distance length and the minimum preset length. The closer the connection distance length is to the minimum preset length, the larger the compensation coefficient is, and vice versa. The value range of the compensation coefficient is (0.75, 1.25).

[0107] In this embodiment, the focus text data of each focus node is determined by calculating the comprehensive focus coefficient of the remaining text data of each focus node, so as to accurately and completely evaluate the evaluation indicators of each focus node and obtain the evaluation value. The bidding template is adjusted according to the evaluation value to improve the integrity and standardization of the bidding documents and the efficiency of document generation.

[0108] In some embodiments of the present application, generating a bid evaluation value corresponding to a node of interest according to the analysis result includes:

[0109] Obtain a historical database for each bid evaluation indicator, analyze historical text data in the historical database, and determine multiple historical data types, historical data functions, and historical characteristic characters corresponding to the bid evaluation indicator;

[0110] Obtain the real-time data type, real-time data function and real-time characteristic character of the concerned text data of each concerned node, and compare them with the historical data type, historical data function and historical characteristic character of each bid evaluation indicator to obtain the first metric threshold, the second metric threshold and the third metric threshold respectively;

[0111] Generate a metric value mean of the concerned text data and the corresponding bid evaluation index according to the first metric value threshold, the second metric value threshold and the third metric value threshold, and set the concerned text data whose metric value mean is greater than the preset metric value threshold as the bid evaluation text data of the corresponding bid evaluation index;

[0112] Inputting the bid evaluation text data of each bid evaluation indicator of the same focus node into the preset bid evaluation model of the corresponding bid evaluation indicator to obtain the indicator evaluation values ​​of multiple bid evaluation indicators of the same focus node;

[0113] The bid evaluation value of the corresponding focus node is generated according to the indicator evaluation values ​​and weight coefficients of multiple bid evaluation indicators of the same focus node.

[0114] In this embodiment, the first metric threshold, the second metric threshold and the third metric threshold refer to the maximum metric value of the historical data type of each bid evaluation indicator, the maximum metric value of the historical data function and the maximum metric value of the historical characteristic character, that is, the similarity between the text data of interest and the historical data type, historical data function and historical characteristic character of each bid evaluation indicator is obtained. The construction of the preset bid evaluation model involves multiple core components, including scoring items, scoring factors, scoring indicators and scoring indicator items. These elements present a one-to-many-to-many-to-many-to-many hierarchical relationship from left to right, forming the basic framework of the model, and are continuously revised and optimized according to the historical database of each bid evaluation indicator to obtain an efficient and accurate preset bid evaluation model.

[0115] In some embodiments of the present application, generating a plurality of adjustment instructions corresponding to the concerned nodes includes:

[0116] Pre-set bid evaluation value thresholds;

[0117] If the bid evaluation value of the concerned node is less than the bid evaluation value threshold, it needs to be adjusted. The concerned text data that needs to be adjusted for the corresponding concerned node is screened out according to the indicator evaluation value of each bid evaluation indicator, and multiple adjustment strategies for the concerned text data are automatically generated;

[0118] The attention text data that needs to be adjusted and the corresponding adjustment strategies are sent to the user end, and the attention text data that needs to be adjusted and the corresponding multiple adjustment strategies are optimized according to the user-defined function, and multiple adjustment instructions corresponding to the attention nodes are generated according to the optimized multiple adjustment strategies.

[0119] In this embodiment, the focus text data that needs to be adjusted is determined based on the evaluation text data corresponding to the evaluation indicators with lower indicator evaluation values, and the adjustment strategy for the focus text data that needs to be adjusted is determined based on the text data corresponding to the larger indicator evaluation value in the preset evaluation model of the corresponding evaluation indicator.

[0120] In this embodiment, the text data of each focus node that needs to be adjusted is adjusted according to the adjustment instructions generated for each focus node, so as to ensure the accuracy of the text data of each focus node and the accuracy of the bid evaluation indicators, and make efficient and accurate judgments on every detail of the bidding template, so as to achieve the integrity and standardization of the bidding documents and improve the efficiency of document generation.

[0121] In some embodiments of the present application, multiple adjustment instructions are simulated, including:

[0122] Obtain the location information and connection structure information of each node of interest in the association structure diagram of the bidding template, and generate a static simulation structure;

[0123] Acquire the dynamic simulation information to be set according to the adjustment requirements of the adjustment instructions;

[0124] Acquire the structural relationship of each concerned text data to be adjusted in the dynamic simulation information to be set, and establish the dynamic relationship of each concerned text node under the same adjustment instruction;

[0125] The dynamic relationship is linked to the static simulation structure in position, and the simulation of the corresponding adjustment instruction is realized to obtain the simulation result of the corresponding adjustment instruction.

[0126] In this embodiment, the simulation process of adjusting the bidding template is realized through the dynamic relationship between the static simulation structure and the adjustment instruction, so that the adjustment instruction is simulated to adjust the corresponding text data of concern, which is conducive to discovering and improving the deficiencies of the adjustment instruction, thereby improving the efficiency of generating bidding documents.

[0127] In some embodiments of the present application, judging whether to adjust the adjustment instruction again according to the simulation result includes:

[0128] The simulation result includes the adjustment text data of the concerned text data to be adjusted in each adjustment instruction and the influence change amount on the concerned text data not to be adjusted;

[0129] Re-evaluate the bid evaluation index corresponding to the adjustment text data to obtain the simulation index evaluation value. If the simulation index evaluation value is greater than the preset index evaluation value threshold, and the impact change amount is less than the preset change amount threshold, there is no need to adjust the adjustment instruction again. Adjust the corresponding focus nodes according to the adjustment instruction and generate the bidding documents.

[0130] If the simulation index evaluation value is less than the preset index evaluation value threshold, and there is an impact change greater than the preset change threshold, the corresponding adjustment instruction is fine-tuned.

[0131] In this embodiment, the application effect of the adjustment instruction is judged according to the simulation results, so as to avoid the quality of the bidding template and the document generation efficiency being reduced due to the error in the formulation of the adjustment instruction, and the shortcomings of the adjustment instruction are discovered in time and fine-tuned to ensure the adjustment efficiency and accuracy of the adjustment instruction, so as to achieve the integrity and standardization of the bidding documents and improve the document generation efficiency.

[0132] In some embodiments of the present application, Figure 2 As shown, it also includes a personalized bidding document intelligent generation method:

[0133] Step S201: extracting target text data from historical bidding documents according to user needs, and generating a bidding template according to the target text data and preset bidding content;

[0134] Step S202: constructing a correlation structure diagram of the bidding template, wherein the correlation structure diagram includes a plurality of focus nodes and each of the correlation nodes contains a plurality of focus text data;

[0135] Step S203: presetting a plurality of bid evaluation indicators, analyzing the concerned text data of the concerned node according to the plurality of bid evaluation indicators, and generating a bid evaluation value corresponding to the concerned node according to the analysis result;

[0136] Step S204: judging whether adjustment is needed according to the bid evaluation value, and if necessary, generating a plurality of adjustment instructions corresponding to the concerned nodes according to the bid evaluation value and the user-defined function;

[0137] Step S205: Simulate multiple adjustment instructions, and determine whether to adjust the adjustment instructions again based on the simulation results. If not, adjust the corresponding focus nodes according to the adjustment instructions and generate bidding documents.

[0138] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principles of the present application. These improvements and substitutions should also be regarded as the scope of protection of the present application.

Claims

1. A personalized bidding document intelligent generation system, characterized by: include: A template generation module is used to extract target text data from historical bidding documents according to user needs, and generate a bidding template according to the target text data and preset bidding content; A construction module, used to construct a correlation structure diagram of a bidding template, wherein the correlation structure diagram includes a plurality of attention nodes and each of the association nodes contains a plurality of attention text data; The bid evaluation module is used to pre-set multiple bid evaluation indicators, analyze the concerned text data of the concerned node according to the multiple bid evaluation indicators, and generate the bid evaluation value of the corresponding concerned node according to the analysis result; The adjustment module is used to determine whether adjustment is needed according to the bid evaluation value. If necessary, multiple adjustment instructions corresponding to the concerned nodes are generated according to the bid evaluation value and user-defined functions; The file generation module is used to simulate multiple adjustment instructions, determine whether to adjust the adjustment instructions again based on the simulation results, and if not, adjust the corresponding focus nodes according to the adjustment instructions and generate bidding documents.

2. The personalized bidding document intelligent generation system according to claim 1, characterized in that: Generate bidding templates based on target text data and preset bidding content, including: Determine bidding-related information and historical bidding document information of the corresponding user according to user needs; Screening historical bidding documents in the bidding document library according to bidding related information and historical bidding document information, and obtaining historical text data in the historical bidding documents whose correlation with user needs is greater than a preset correlation threshold; Sort the correlation degree of the selected historical text data with the same business demand attribute, and set the historical text data with the highest ranking as the target text data corresponding to the demand attribute; The corresponding positions in the preset bidding content are dynamically filled according to multiple target text data to generate a bidding template.

3. The personalized bidding document intelligent generation system according to claim 2, characterized in that: Construct the associated structure diagram of the bidding template, including: Divide the bidding template into a plurality of focus modules according to the functional type of the bidding template, construct an initial structure diagram according to the logical relationship of the corresponding functional type of the focus module, set the maximum importance of the functional type of the focus module as the structure center of the initial structure diagram, and set the connection distance according to the logical relationship between the structure center and other focus modules; Analyze the text data contained in each module of interest in the initial structure diagram, obtain the utilization coefficient and influence coefficient of the main functional type of the text data in the corresponding module of interest, generate the importance coefficient based on the utilization coefficient and the influence coefficient, and determine the characteristic text data of each module of interest according to the importance coefficient; Based on the comparison relationship between the important coefficient of the feature text data and the preset important coefficient threshold, the text data of the corresponding focus module is grouped to obtain multiple text data groups of each focus module, and each text data group is set as a focus node; Analyze the dependency and association relationship between the feature text data of multiple focus nodes of the same focus module, determine the node center of the corresponding focus module according to the dependency and association relationship, set the connection distance between the node center and the remaining focus nodes of the same focus module according to the dependency degree and association degree, and generate a connection structure diagram of the corresponding focus module; Import the connection structure diagram of each module of interest into the corresponding position of the initial structure diagram to form the associated structure diagram of the bidding template.

4. The personalized bidding document intelligent generation system according to claim 3, characterized in that: Each attention node contains multiple attention text data, including: Performing a first correlation analysis on the remaining text data of each focus node and the characteristic text data of the node center of the focus module of other functional types, and performing a second correlation analysis on the remaining text data of each focus node and the characteristic text data of the remaining focus nodes excluding the node center of the focus module of other functional types, to obtain a plurality of first correlation values ​​and a plurality of second correlation values; Analyze the multiple first correlation values ​​to obtain the number of first correlation values ​​greater than the first preset correlation threshold and the first correlation difference value of the first correlation value greater than the first preset correlation threshold to generate a first attention coefficient corresponding to the remaining text data; Analyze the plurality of second correlation values ​​to obtain the number of second correlation values ​​greater than a second preset correlation threshold and a second correlation difference value greater than the second preset correlation threshold to generate a second attention coefficient corresponding to the remaining text data; Generate a compensation coefficient according to the connection distance length between the remaining text data and the feature text data of the node center of the corresponding attention module; A comprehensive attention coefficient of the remaining text data is generated according to the compensation coefficient, the first attention coefficient, and the second attention coefficient.

5. The personalized bidding document intelligent generation system according to claim 4, characterized in that: Each attention node contains multiple attention text data, including: The calculation formula of the comprehensive attention coefficient is: Among them, Z is the comprehensive attention coefficient, b is the compensation coefficient, g is the conversion coefficient, is the first attention coefficient, n is the number of first correlation values ​​greater than the first preset correlation threshold, △p1i is the ith first correlation difference, yi is the weight coefficient of the ith first correlation difference, a1 is the weight coefficient of the first attention coefficient, is the second attention coefficient, m is the number of second association values ​​greater than the second preset association threshold, ys is the weight coefficient of the sth second association difference value, △p2s is the sth second association difference value; Pre-set the attention factor threshold; The remaining text data whose comprehensive attention coefficient is greater than the attention coefficient threshold and the characteristic text data of the same attention node are set as the attention text data of the corresponding attention node.

6. The personalized bidding document intelligent generation system according to claim 5, characterized in that: Generate bid evaluation values ​​for corresponding focus nodes based on analysis results, including: Obtain a historical database for each bid evaluation indicator, analyze historical text data in the historical database, and determine multiple historical data types, historical data functions, and historical characteristic characters corresponding to the bid evaluation indicator; Obtain the real-time data type, real-time data function and real-time characteristic character of the concerned text data of each concerned node, and compare them with the historical data type, historical data function and historical characteristic character of each bid evaluation indicator to obtain the first metric threshold, the second metric threshold and the third metric threshold respectively; Generate a metric value mean of the concerned text data and the corresponding bid evaluation index according to the first metric value threshold, the second metric value threshold and the third metric value threshold, and set the concerned text data whose metric value mean is greater than the preset metric value threshold as the bid evaluation text data of the corresponding bid evaluation index; Inputting the bid evaluation text data of each bid evaluation indicator of the same focus node into the preset bid evaluation model of the corresponding bid evaluation indicator to obtain the indicator evaluation values ​​of multiple bid evaluation indicators of the same focus node; The bid evaluation value of the corresponding focus node is generated according to the indicator evaluation values ​​and weight coefficients of multiple bid evaluation indicators of the same focus node.

7. The personalized bidding document intelligent generation system according to claim 6, characterized in that: Generate multiple adjustment instructions corresponding to the concerned nodes, including: Pre-set bid evaluation value thresholds; If the bid evaluation value of the concerned node is less than the bid evaluation value threshold, it needs to be adjusted. The concerned text data that needs to be adjusted for the corresponding concerned node is screened out according to the indicator evaluation value of each bid evaluation indicator, and multiple adjustment strategies for the concerned text data are automatically generated; The attention text data that needs to be adjusted and the corresponding adjustment strategies are sent to the user end, and the attention text data that needs to be adjusted and the corresponding multiple adjustment strategies are optimized according to the user-defined function, and multiple adjustment instructions corresponding to the attention nodes are generated according to the optimized multiple adjustment strategies.

8. The personalized bidding document intelligent generation system according to claim 7, characterized in that: Simulate multiple tuning instructions, including: Obtain the location information and connection structure information of each node of interest in the associated structure diagram of the bidding template, and generate a static simulation structure; Acquire the dynamic simulation information to be set according to the adjustment requirements of the adjustment instructions; Acquire the structural relationship of each concerned text data to be adjusted in the dynamic simulation information to be set, and establish the dynamic relationship of each concerned text node under the same adjustment instruction; The dynamic relationship is linked to the static simulation structure in position, and the simulation of the corresponding adjustment instruction is realized to obtain the simulation result of the corresponding adjustment instruction.

9. The personalized bidding document intelligent generation system according to claim 8, characterized in that: Determine whether to adjust the adjustment instruction again according to the simulation results, including: The simulation result includes the adjustment text data of the concerned text data to be adjusted in each adjustment instruction and the influence change amount on the concerned text data not to be adjusted; Re-evaluate the bid evaluation index corresponding to the adjustment text data to obtain the simulation index evaluation value. If the simulation index evaluation value is greater than the preset index evaluation value threshold, and the impact change amount is less than the preset change amount threshold, there is no need to adjust the adjustment instruction again. Adjust the corresponding focus nodes according to the adjustment instruction and generate the bidding documents. If the simulation index evaluation value is less than the preset index evaluation value threshold, and there is an impact change greater than the preset change threshold, the corresponding adjustment instruction is fine-tuned.

10. A personalized bidding document intelligent generation method, characterized in that: include: Extract target text data from historical bidding documents according to user needs, and generate bidding templates based on the target text data and preset bidding content; Constructing a correlation structure diagram of a bidding template, wherein the correlation structure diagram includes a plurality of focus nodes and each correlation node contains a plurality of focus text data; Preset multiple bid evaluation indicators, analyze the concerned text data of the concerned node according to the multiple bid evaluation indicators, and generate the bid evaluation value of the corresponding concerned node according to the analysis results; Determine whether adjustment is needed based on the bid evaluation value. If necessary, generate multiple adjustment instructions for corresponding nodes of interest based on the bid evaluation value and user-defined functions; Simulate multiple adjustment instructions, and determine whether to adjust the adjustment instructions again based on the simulation results. If not, adjust the corresponding focus nodes according to the adjustment instructions and generate bidding documents.

Images