Method, system and equipment for correcting and combining bidding documents of road and bridge engineering and medium

Abstract

The embodiment of the invention provides a road and bridge project bidding document correcting and combining method, system and device and a medium, and belongs to the field of computers. The method comprises the following steps: receiving an original bidding file of a road and bridge project and bid invitation party format requirement data, and synchronizing the original bidding file and the bid invitation party format requirement data to a road and bridge project sub-scene format sub-library; performing basic element extraction and road and bridge industry exclusive element identification on the received road and bridge project original bidding file, and outputting an analysis result containing industry element information; calling a rule, performing road and bridge industry special format correction on the analysis result, and outputting a first preliminary correction result; correcting the analysis result by adopting a machine learning model to obtain a second preliminary correction result; and fusing the first preliminary correction result and the second preliminary correction result to obtain conflict-free data, and merging the conflict-free data according to a standard logic sequence of the road and bridge project bidding file to generate a road and bridge project target bidding file. A dual correction mode is adopted for industry elements, the bidding compliance rate is increased, and the correction period is shortened.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular to a road and bridge engineering bidding document revision and compilation method, system, device and medium. BACKGROUND

[0002] In the procurement behavior of goods, engineering and services, the tenderer attracts a large number of bidders to compete on equal terms according to the same conditions, organizes technical, economic and legal experts to conduct comprehensive evaluation on a large number of bidders according to the prescribed procedures, and selects the winning bidder of the project from among them.

[0003] The special element format requirement of the road and bridge industry is special, and the manual adjustment is much more difficult than the general document: the column sequence of the bid document engineering quantity list needs to be fixed "list number-project name-unit-quantity-unit price", the construction process flowchart needs to use "bored pile" "box girder pouring" and other industry special symbols, the safety production special scheme table needs to mark "risk level-prevention and control measures" special column, and manual adjustment needs to consider format specification and industry terminology at the same time, which is prone to the problem of "format is correct but industry identification is wrong".

[0004] The dynamic change of the format requirement of the tenderer is lagging behind, and the bidding opportunity is easily missed: the tenderer of the road and bridge engineering often temporarily supplements the format requirement (such as adding "smart construction chapter" font specification) 3-5 days before the bidding deadline, and the traditional manual adjustment needs to be re-adjusted page by page, especially the engineering quantity list table needs to be re-calibrated column sequence, which is prone to format omission due to insufficient time.

[0005] Moreover, the existing tool lacks adaptability to the road and bridge industry, and the special revision capability is insufficient: the general document processing tool (such as Word format brush) cannot identify industry-specific problems such as "engineering quantity list column sequence error" and "construction process flowchart symbol not standardized", which needs to be checked manually one by one, resulting in the dilemma of "format compliance difficulty and low compilation efficiency" for road and bridge companies. SUMMARY

[0006] The purpose of the embodiment of the present application is to provide a road and bridge engineering bidding document revision and compilation method, system, device and medium, which is customized for the revision of engineering quantity list, construction process flowchart and other industry elements, improves the bidding compliance rate, and completes the rule update and document revision in a short time when the tenderer temporarily supplements the format requirement, which greatly shortens the response cycle compared with manual adjustment.

[0007] In order to achieve the above purpose, the embodiment of the present application provides a road and bridge engineering bidding document revision and compilation method, which comprises: receiving a road and bridge engineering original bidding document and tenderer format requirement data, and synchronizing the tenderer format requirement data to a road and bridge engineering scene-specific format sub-library; The received original road and bridge engineering bid document is subjected to basic element extraction and road and bridge industry exclusive element identification, and the analysis result containing industry element information is output; The rules of the road and bridge engineering sub-scene format library are called to correct the analysis result in the road and bridge industry, and the first preliminary correction result is output; The analysis result is corrected by using a machine learning model to obtain a second preliminary correction result; The first preliminary correction result and the second preliminary correction result are fused to obtain conflict-free data, and the conflict-free data is combined according to the logical order of the road and bridge engineering bid document standard to generate the road and bridge engineering target bid document.

[0008] Optionally, the rules of the road and bridge engineering sub-scene format library are called to correct the analysis result in the road and bridge industry, and the first preliminary correction result is output, including: Based on the analysis result, the position information of the basic elements and the road and bridge industry exclusive elements in the original road and bridge engineering bid document is determined; According to the types of the basic elements and the road and bridge industry exclusive elements and the engineering scene, the corresponding rules are called from the road and bridge engineering sub-scene format library; According to the judgment logic in the rules, the actual format of the basic elements and the road and bridge industry exclusive elements is compared with the standard format, and the compliance or violation result and the violation type are output; Select a correction scheme matching the violation type to correct the violation result and obtain the first preliminary correction result.

[0009] Optionally, the analysis result is corrected by using a machine learning model to obtain a second preliminary correction result, including: For the engineering quantity table element type in the original road and bridge engineering bid document, a Transformer model is selected to perform table column name semantic identification to avoid missing judgment due to different column name expressions in the road and bridge engineering sub-scene format library; For the construction picture element type in the original road and bridge engineering bid document, a target detection model is selected to identify whether the picture contains a numbered watermark and a resolution insufficient area; For the process flowchart element type in the original road and bridge engineering bid document, a graph neural network is used to learn the road and bridge construction process logic to assist in identifying flow logic errors.

[0010] Optionally, for the engineering quantity table element type in the original road and bridge engineering bid document, a Transformer model is selected to perform table column name semantic identification to avoid missing judgment due to different column name expressions in the road and bridge engineering sub-scene format library, including: For the input engineering quantity table, the header row of the engineering quantity table is read by pandas to obtain the column name text; The obtained column name text is input into a Transformer model, and the confidence of multiple categories is output, and the category with the highest confidence is selected as the predicted category, wherein the structure of the Transformer model is an input layer, a BERT pre-training layer, the first 6 layers are frozen to retain the bottom semantic features, a pooling layer, an output of 768-dimensional vectors representing the overall semantics of the column name, a dropout layer, a first fully connected layer to compress the feature dimension, a second fully connected layer corresponding to 12 standard column name categories, and an output layer using a Softmax function to output the confidence of each category. After the Transformer model outputs the predicted category, a road and bridge engineering sub-scene format library is called to perform column name correction. If the Transformer model identifies the project code as a list number category, the standard format of the list number category in the road and bridge engineering sub-scene format library requires 12-bit coding. Then, it is checked whether the format of the current project code meets the 12-bit coding requirement. If not, a correction prompt is automatically popped up that the column name project code belongs to the list number category and needs to be supplemented to 12-bit coding. If it meets the requirement, it is marked as column name semantic compliance and enters the subsequent table column sequence correction link.

[0011] Optionally, for the construction picture element type in the original road and bridge engineering bidding document, a target detection model is selected to identify whether the picture contains a numbered watermark and a resolution insufficient area, including: The input construction picture is subjected to size normalization, pixel standardization, and noise suppression processing to obtain a target construction picture; ResNet-50 network is used to extract deep visual features of the target construction picture, and a sliding window is generated on the feature map to generate a candidate region containing the construction target; For the generated candidate region containing the construction target, the features are first aligned, and then classification and boundary box fine-tuning are performed to determine whether it is a numbered watermark or a resolution insufficient area to adapt to the precise positioning requirements of the construction target; For the classification result, the target box with a class probability greater than a preset probability threshold is retained, and non-maximum suppression is used to remove overlapping boxes to output the boundary box and category of the numbered watermark and resolution insufficient area.

[0012] Optionally, for the process flowchart element type in the original road and bridge engineering bidding document: a graph neural network is used to learn the road and bridge construction process logic to assist in identifying process logic errors, including: The process text box in the process flowchart is identified using OCR, and is used as the node set of the target graph. The arrow connection between the process text boxes is identified and used as the edge set of the target graph; The node set and edge set are integrated to obtain structured data that converts the process flowchart into node-edges; The structured data of nodes-edges is input into the input layer of the GAT graph neural network, and linear transformation is performed on the node features to enhance the expression of the node features. The attention weight of the node on its neighbor nodes is calculated, and the features of the neighbor nodes are aggregated according to the attention weight to update the features of the node. The updated final features of all nodes are globally mean-pooled to obtain the graph-level features of the entire process flowchart, and the graph-level features are input into a fully connected layer to output the probability distribution of the logical error type.

[0013] Optionally, the attention weight of the node on its neighbor nodes is calculated, and the features of the neighbor nodes are aggregated according to the attention weight to update the features of the node, including: The correlation of the node pair is calculated according to the following formula: ; In the formula, a represents the attention weight vector, Z i and Z j respectively represent the features of the node and the neighbor node after linear transformation, Z i and Z j are spliced; The attention weight normalization formula is as follows: ; In the formula, N( i ) is the set of all neighbor nodes of the node v i . The feature update formula of the node is as follows: ; In the formula, is an activation function.

[0014] In a second aspect, the present application also provides a system for revising and merging a road and bridge engineering bidding document, comprising: An input module configured to receive a road and bridge engineering original bidding document and tender party format requirement data, and synchronize the tender party format requirement data to a road and bridge engineering sub-scene format sub-library; An element analysis module configured to extract basic elements and identify road and bridge industry exclusive elements from the received road and bridge engineering original bidding document, and output an analysis result containing industry element information; A revision module configured to call rules of the road and bridge engineering sub-scene format sub-library, revise the analysis result according to road and bridge industry special formats, and output a first preliminary revision result; and revise the analysis result by using a machine learning model to obtain a second preliminary revision result. The first preliminary revision result and the second preliminary revision result are fused to obtain conflict-free data, and the conflict-free data is combined according to a logical sequence of a road and bridge engineering bidding document standard to generate a road and bridge engineering target bidding document.

[0015] In a third aspect, the present application also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the road and bridge engineering bidding document revision and compilation method when executing the program.

[0016] In a fourth aspect, the present application also provides a storage medium, which stores a computer program, wherein the computer program implements the steps of the road and bridge engineering bidding document revision and compilation method when executed by a processor.

[0017] Through the above technical solution, the road and bridge engineering bidding document is processed in a three-in-one manner of “dynamic compliance + collaborative pre-digestion + special revision”, the document is ensured to meet the industry standards and the dynamic requirements of the tendering party, the preparation period is shortened by more than 60%, and the system is integrated with the road and bridge enterprise bidding management system, and has strong industry adaptability and practicality.

[0018] Other features and advantages of the embodiments of the present application will be described in detail in the following specific implementation part. BRIEF DESCRIPTION OF DRAWINGS

[0019] The accompanying drawings are used to provide a further understanding of the embodiments of the present application, and constitute a part of the specification, and are used to explain the embodiments of the present application together with the following specific implementation, but do not constitute a limitation to the embodiments of the present application. In the drawings: Figure 1 is an implementation flowchart of a road and bridge engineering bidding document revision and compilation method provided by the embodiments of the present application; Figure 2 is a structural schematic diagram of a road and bridge engineering bidding document revision and compilation system provided by the embodiments of the present application; Figure 3 is an example diagram of a road and bridge engineering bidding document revision and compilation system provided by the embodiments of the present application; Figure 4 is a hardware structural schematic diagram of an electronic device provided by the embodiments of the present application. DETAILED DESCRIPTION

[0020] Various embodiments of this disclosure will be described more fully in the following detailed description. This disclosure may have various embodiments, and adjustments and changes may be made therein. However, it should be understood that there is no intention to limit the various embodiments of this disclosure to the specific embodiments disclosed herein, but rather this disclosure should be understood to cover all adjustments, equivalents, and / or alternatives falling within the spirit and scope of the various embodiments of this disclosure.

[0021] In the following, the terms “comprising” or “may include”, which may be used in various embodiments of this disclosure, indicate the presence of the disclosed functions or operations and do not limit the addition of one or more functions or operations. Furthermore, as used in various embodiments of this disclosure, the terms “comprising,” “having,” and their cognates are intended only to indicate a specific feature, number, step, operation, or combination of the foregoing and should not be construed as primarily excluding the presence of one or more other features, numbers, steps, operations, or combinations of the foregoing, or the possibility of adding one or more features, numbers, steps, operations, or combinations of the foregoing.

[0022] In various embodiments of this disclosure, the expression "or" or "at least one of A and / or B" includes any combination or all combinations of the words listed simultaneously. For example, the expression "A or B" or "at least one of A and / or B" may include A, may include B, or may include both A and B.

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] See Figure 1 The diagram shows a flowchart of a method for revising and finalizing tender documents for road and bridge engineering projects in a specific embodiment, including the following execution steps: Step 100: Receive the original tender documents and the tenderer's format requirements data for the road and bridge project, and synchronize the tenderer's format requirements data to the road and bridge project scenario-specific format sub-library.

[0025] Specifically, in addition to supporting Word / PDF, it adds import of road and bridge industry-specific formats (Excel, cost estimation software exported files) and automatically converts them into parsable formats; it also provides an "upload portal for bidding party format requirements" to support users to upload temporary supplementary format documents (such as a Word version of "Green Construction Chapter Requirements").

[0026] The road and bridge engineering bidding document includes the official document submitted by the road and bridge company when participating in the bidding of highway, bridge, tunnel and other engineering, in addition to the basic format, needs to meet the road and bridge industry special requirements (such as the column sequence of the bill of quantities table, the symbol of the construction process flow chart), covers the core modules such as construction organization design, bill of quantities, safety production special scheme and the like.

[0027] For example, the road and bridge engineering scene format sub-library is shown in Table 1: Table 1 Road and bridge engineering scene format sub-library

[0028] Step 101: Extracting basic elements and identifying road and bridge industry special elements from the received road and bridge engineering original bidding document, and outputting the analysis result containing industry element information.

[0029] Step 102: Calling the rules of the road and bridge engineering scene format sub-library to correct the road and bridge industry special format of the analysis result, and outputting the first preliminary correction result.

[0030] Specifically, when step 102 is executed, the following steps can be executed: S1020: Based on the analysis result, the position information of the basic elements and the road and bridge industry special elements in the road and bridge engineering original bidding document is determined.

[0031] S1021: According to the types of basic elements and road and bridge industry special elements and engineering scene, the corresponding rules are called from the road and bridge engineering scene format sub-library.

[0032] For example, the basic rules are: page margin, header and footer, etc.; the industry rules are: table: fixed column sequence of bill of quantities, "safety production table" needs to contain "risk level" column; picture: construction picture number is "chapter number-list number-picture name"; flow chart: construction process uses road and bridge special symbol, arrow is "red dotted line" (representing construction process); dynamic update: support user to upload temporary requirements of the bidder, complete rule input and sub-library update within 10 minutes.

[0033] S1022: According to the judgment logic in the rules, the actual format of the basic elements and the road and bridge industry special elements is compared with the standard format, and the compliance or violation result and the violation type are output.

[0034] S1023: Selecting the correction scheme matched with the violation type, correcting the violation result, and obtaining the first preliminary correction result.

[0035] Specifically, basic correction: title number, margin, etc.; industry-specific correction: table: automatically align the column sequence of the bill of quantities, supplement the missing "list number" column; picture: generate industry standard numbers for construction pictures, and automatically prompt "high-definition drawing needed" when the resolution is insufficient; flowchart: replace non-industry symbols, such as replacing "general arrow" with "construction process arrow", and adjusting the color scheme to industry standards (light gray background + black text); machine learning optimization: through historical road and bridge bidding file cases, optimize the accuracy of "bill column sequence alignment" and "flowchart symbol replacement".

[0036] For example, the parsing result is called first to determine the specific location of the "quantity table, construction picture, flowchart" in the bidding file; according to the element type and engineering scene, the corresponding rule set is called from the rule library; according to the "judgment logic" in the rules, the actual format of the element is compared with the standard format, and the "compliance / violation" result and the violation type are output; the violation items are operated according to the "correction scheme", for example, "column sequence error" calls the table processing interface to adjust the column order, and "number missing" generates standard numbers based on chapter numbers and list numbers.

[0037] Step 103: correcting the parsing result by using a machine learning model to obtain a second preliminary correction result.

[0038] Specifically, when step 103 is performed, the following steps can be specifically performed: S1030: For the element type of the quantity table in the original bidding file of the road and bridge project, a Transformer model is selected to perform semantic recognition of the table column name, so as to avoid missing judgment due to different column name expressions in the format sub-library of road and bridge engineering scenes.

[0039] Specifically, when step S1030 is performed, the following sub-steps can be specifically performed: SA: For the input quantity table, use pandas to read the header row of the quantity table to obtain the column name text.

[0040] SB: input the obtained column name text into the Transformer model, output the confidence of multiple categories, and select the category with the highest confidence as the predicted category, wherein the structure of the Transformer model is an input layer, a BERT pre-training layer, the first 6 layers are frozen to retain the bottom semantic features, a pooling layer, an output of 768-dimensional vector representing the overall semantic of the column name, a dropout layer, a first fully connected layer compressing feature dimension, a second fully connected layer corresponding to 12 standard column name categories, and an output layer using a Softmax function to output the confidence of each category.

[0041] SC: After the Transformer model outputs the predicted category, it calls the road and bridge engineering scenario-specific format sub-library to perform column name correction. If the Transformer model identifies the project code as a list number, and the standard format requirement for the list number in the road and bridge engineering scenario-specific format sub-library is 12-digit encoding, it checks whether the current project code format meets the 12-digit encoding requirement. If it does not meet the requirement, a correction prompt will automatically pop up, indicating that the column name project code belongs to the list number category and needs to be supplemented with a 12-digit encoding. If it meets the requirement, it is marked as semantically compliant with column name and proceeds to the subsequent table column order correction stage.

[0042] In one specific implementation, let the text of the column name in the quantity table to be identified be x, and after word segmentation, a token sequence is obtained: x =[ t 1, t 2,..., t T ], where T is the sequence length. Initial features are obtained through BERT's embedding layers: The sequence is The L-layer Transformer encoder in BERT outputs context-aware features: ,in h t For the first t The contextual features of each token; the features used to aggregate the semantics of the entire sentence are taken as the overall semantic representation of the column name: Through the weighted connection layer Mapping to the standard class space, the output predicted probability is calculated by the fully connected layer: In the formula, Let c be the weight vector of class c. This is a bias. The class probabilities are obtained by normalizing using the Softmax function: in, Let x be the predicted probability that column name x belongs to category c. The quantification formula for misidentification is as follows:

[0043] In the formula, p y For the real category y The predicted probability, Represents the true category y The one-hot encoding representation. The total loss function for the batch of samples is:

[0044] In the formula, Let be the predicted probability of the true class of the i-th sample. Let the weight of the true class of the i-th sample be . To focus on the factor.

[0045] S1031: For the construction picture element type in the original bidding file of the road and bridge project, a target detection model is selected to identify whether the picture contains a numbered watermark and a resolution insufficient area.

[0046] Specifically, when step S1031 is executed, the following sub-steps can be specifically executed: Sa: Perform size normalization, pixel standardization and noise suppression processing on the input construction picture to obtain a target construction picture.

[0047] Specifically, the input picture is resized to a fixed size, and the scaling ratio is recorded for restoring the bounding box representation; the RGB three channels are standardized respectively, and the formula is: Wherein, is the pixel value of the original picture in the c channel, h row and ω column, is the mean value of the channel in the construction picture training set, is the standard deviation.

[0048] Sb: A ResNet-50 network is used to extract deep visual features of the target construction picture, and a sliding window is generated on the feature map to generate a candidate region containing the construction target.

[0049] Specifically, ResNet-50 is composed of 4 residual groups, and the formula of each residual block is: Wherein, x is the input feature map of the residual block, W 1, W 2 is the convolution kernel weight, BN is batch normalization, ReLU is the activation function, shortcut( x ) is the shortcut connection, and after ResNet-50, the original picture is down-sampled to a feature map F of 16×16×2048.

[0050] Sc: For the generated candidate region containing the construction target, first align the features, and then classify whether it is a numbered watermark or a resolution insufficient area and fine-tune the bounding box to adapt to the precise positioning requirements of the construction target.

[0051] Specifically, at each pixel position of the feature map F, 5 scales and 3 aspect ratios of anchor boxes are generated, covering small size watermarks and large size low resolution areas, and the total number of anchor boxes is N= HF × WF × K , K is the number of anchor boxes for each pixel. For each anchor box, RPN outputs two results, the classification score: use 1×1 convolution to map the feature map F to 2× NanchorThe vector is input into the Sigmoid function to obtain the foreground confidence of each anchor box p rpn : wherein, is a classification weight, is a bias. The 1x1 convolution is mapped to 4x N anchor The vector is input into the Sigmoid function to obtain the foreground confidence of each anchor box wherein, is a regression weight, is a bias. The Softmax function is used to obtain the class probability: wherein, , is a full connection layer weight, , is a bias. The boundary fine-tuning correction formula is as follows:

[0052] In the formula, is the center coordinates, width and height of the RPN candidate box, is the final target box coordinates.

[0053] Sd: For the classification result, the target box with a class probability greater than a preset probability threshold is retained, and non-maximum suppression is used to remove overlapping boxes, and the boundary box and the class of the numbered watermark and the resolution insufficient area are output.

[0054] S1032: For the process flowchart element type in the original bidding file of the road and bridge engineering: use the graph neural network to learn the road and bridge construction process logic, and assist in identifying the process logic error.

[0055] Specifically, when step S1032 is performed, the following sub-steps can be specifically performed: S1: Use OCR to identify the process text box in the process flowchart, and take it as the node set of the target graph, and identify the arrow connection between the process text box, and take it as the edge set of the target graph.

[0056] S2: Integrate the node set and the edge set to obtain the structured data of converting the process flowchart into node-edge.

[0057] S3: Input the node-edge structured data into the input layer of the GAT graph neural network, and perform linear transformation on the node features to enhance the node feature expression.

[0058] Specifically, the linear transformation formula is as follows: wherein, is a learnable weight matrix, is a node feature.

[0059] S4: The computing node calculates the attention weight of its neighbor node, and aggregates the features of the neighbor node according to the attention weight, and updates the features of the node.

[0060] Specifically, the correlation calculation formula of the node pair is as follows: In the formula, a represents an attention weight vector, Z i and Z j respectively represent the features of the node and the neighbor node after linear transformation, and || represents the concatenation of Z i and Z j . The attention weight normalization formula is as follows: In the formula, N( i ) is the set of all neighbor nodes of the node v i . The feature update formula of the node is as follows: In the formula, f is an activation function.

[0061] S5: The updated final features of all nodes are globally mean-pooled to obtain the graph-level features of the entire process flowchart, and the graph-level features are input into a fully connected layer to output the probability distribution of the logical error type.

[0062] In actual application, the final node features are the concatenation or mean value of multi-head features:

[0063] In the formula, h K is the number of attention heads, is the attention weight of the kth head, and || represents feature concatenation.

[0064] In some embodiments, the graph-level error classification loss formula is as follows:

[0065] In the formula, y is the true label, is the class weight, and p is the predicted probability.

[0066] The node-level error positioning loss function is as follows: ​​

[0067] In the formula, N abn is the number of abnormal association nodes, y i is the node v i is the true label of the node, is the predicted probability.

[0068] It should be noted that the model in the present application does not replace the rule base, but is used as a "supplement and optimization" to solve the limitations of the rule base: 1. Improve the recognition accuracy of fuzzy scenarios: For example, the rule base is difficult to judge the case of "non-standard column name expression", at this time the model can accurately identify "non-standard column name expression but substantial compliance", and avoid misjudgment of the rule base as "missing column"; 2. Optimize the priority of correction and self-adaptive adjustment: When multiple violations exist at the same time, the model can learn the priority of "correcting column sequence first, then checking data format" through historical correction cases; At the same time, the model will count the cases that are still returned by the tenderer after correction, and automatically feed back the suggestion of "adding regional rules" to the rule base, forming a closed loop of "correction-feedback-optimization".

[0069] Step 104: fuse the first preliminary correction result and the second preliminary correction result, obtain conflict-free data, and combine the conflict-free data according to the logical order of the road and bridge engineering bidding file standard to generate a road and bridge engineering target bidding file.

[0070] Specifically, when multiple writers edit, real-time synchronization of scene sub-base rules; conflict warning: such as "cost group uses blue border table (violation, should be black)", "safety group misses 'risk level' column", instant pop-up warning and prompt correction scheme; permission control: only "format administrator" can modify industry rules to avoid misoperation of ordinary writers. Logical merging: merge sub-files according to the standard order of road and bridge engineering bidding files (engineering overview → construction organization design → bill of quantities → safety production scheme); secondary correction: final calibration of "cross-module format consistency" of the merged document.

[0071] Collaboration: receive conflict-free data from the collaborative conflict pre-dissipation module, complete merging and secondary correction, and transmit to the output module.

[0072] In a specific embodiment, the synergy mechanism of the rule base and the machine learning model: both are not independent, but work together through the process of "rule screening, model judgment, and result mutual feeding", to ensure the comprehensiveness and accuracy of identification and correction: first step: rule base screening: first, through the rule base to handle "deterministic violations", quickly solve 80% of the basic format problems, reduce the model calculation amount; second step: model judgment: for the fuzzy scene that the rule base "cannot judge", call the model for secondary identification, output accurate results; third step: result mutual feeding: the "new violation type" identified by the model will automatically generate "rule addition suggestion", which is confirmed by the administrator and input into the rule base; the "high-frequency violation rule" of the rule base will be input into the model as "key training samples" to improve the identification speed of the model for this type of problem.

[0073] In some embodiments, the hardware environment for implementing this solution is: PC end: recommended configuration CPU Intel Core i7, memory 16GB, hard disk 500GB (SSD), supporting single parsing of 100MB or more engineering quantity drawing PDF; server end: using dual Intel Xeon E5 processor, memory 64GB, RAID 5 array (2TB or more), supporting 50 people simultaneous collaborative editing (such as "construction group 10 people + cost group 20 people + safety group 20 people"); cloud: based on Ali Cloud ECS (4 cores 8G or more) deployment, supporting elastic expansion of computing resources according to the bidding peak period (such as the end of the quarter). Software environment: operating system: PC end supports Windows 10 / 11 (needs to be compatible with Guanglianda software), server end supports Windows Server 2019; supporting software: install JRE 1.8, Python 3.9 (for industry element recognition model), add Excel format conversion component, road and bridge engineering OCR term recognition plug-in (supporting "list number" "construction process" and other terms accurate extraction); integration compatibility: through API interface and road and bridge enterprise commonly used systems (bid management system, cost software) integration, realize "cost software export → format correction → bid system storage" seamless connection of the whole process.

[0074] It should be understood that the size of the serial number of each step in the above embodiments does not mean the order of execution, and the execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0075] The beneficial effects achieved by the present application are as follows: 1. Solve the pain points of road and bridge industry special formats and improve the compliance rate of bidding: customized correction of industry elements such as bill of quantities and construction process diagram, which can improve the format compliance rate of bidding documents from 70% to more than 98% through manual processing, and avoid the invalidation of bidding qualification due to industry format deviation.

[0076] 2. Dynamic response to the requirements of the tenderer, shorten the preparation cycle: when the tenderer temporarily supplements the format requirements, the rule update and file revision are completed within 10 minutes, compared with the 2-3 days of adjustment time of manual, the response cycle is greatly shortened, and the submission on time is ensured.

[0077] 3. Pre-digestion of conflicts, reduce the cost of joint revision: real-time warning of format deviation when multiple groups of editors, no need for special personnel to revise after joint revision, manual cost is reduced by 70%, and the preparation cycle is shortened from an average of 10 days to 4 days or less.

[0078] 4. Strong industry adaptability, expand application scenarios: support file import of Guanglianda, Tongwang and other road and bridge cost software, can adapt to multiple types of engineering bidding scenarios such as highway, bridge and tunnel, and can be integrated with the existing bid document system of the enterprise, the practicality and expansibility far exceed that of general document processing tools.

[0079] As shown in Figure 2 The following is an embodiment of the road and bridge engineering bidding file revision and joint revision system provided by the embodiment of the present disclosure, which belongs to the same inventive concept as the road and bridge engineering bidding file revision and joint revision method described above. Details not described in the embodiment of the road and bridge engineering bidding file revision and joint revision system can be referred to the embodiment of the road and bridge engineering bidding file revision and joint revision method.

[0080] The road and bridge engineering bidding file revision and joint revision system comprises: An input module configured to receive a road and bridge engineering original bidding file and tenderer format requirement data, and synchronize the tenderer format requirement data to a road and bridge engineering sub-scene format sub-database; An element analysis module configured to extract basic elements and identify road and bridge industry-specific elements from the received road and bridge engineering original bidding file, and output an analysis result containing industry element information; A revision module configured to call rules of the road and bridge engineering sub-scene format sub-database, revise the analysis result according to road and bridge industry-specific formats, and output a first preliminary revision result; revise the analysis result by using a machine learning model to obtain a second preliminary revision result; A joint revision module configured to fuse the first preliminary revision result and the second preliminary revision result to obtain conflict-free data, and combine the conflict-free data according to a standard logical sequence of road and bridge engineering bidding files to generate a road and bridge engineering target bidding file.

[0081] In an example, refer to Figure 3As shown, it is an example diagram of a road and bridge engineering bidding file revision and combination system provided by an embodiment of the present application. The input module supports single file upload, multiple sub-file upload, template selection: the user can select a specific template. The document parsing module: parses the uploaded single file or multiple sub-files. The format knowledge base: provides document format related knowledge for guiding the processing process. The intelligent revision engine: uses the information in the specification configuration and version library to intelligently revise the document. The combination module: combines and processes multiple files or templates. The specification configuration and version library: provides specification configuration and version management for different bidders / projects, supports document processing and revision. The output module: exports the processed document in Word or PDF format. The system integration interface: integrates with other systems through API or internal platform. Review and leave marks: provide revision rules and difference visualization to facilitate review and track modification records.

[0082] Figure 4 It is a hardware structure schematic diagram of an electronic device for implementing various embodiments of the present application.

[0083] The road and bridge engineering bidding file revision and combination method provided by the embodiment of the present application can be applied to an electronic device. Those skilled in the art can understand that the electronic device structure involved in the embodiment of the present application does not constitute a limitation on the electronic device, and the electronic device can include more or fewer components than the diagram, or combine certain components, or different component arrangements. In the embodiment of the present application, the electronic device includes but is not limited to a laptop computer, a desktop computer, a workstation, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of the present application described herein and / or claimed.

[0084] The electronic device can include a processor, an external memory interface, an internal memory, a universal serial bus (USB) interface, a charge management module, a power management module, a battery, a wireless communication module, an audio module, a speaker, a microphone, a sensor module, a key, a camera, a display screen, and a SIM card interface, etc.

[0085] It can be understood that the structure shown in the embodiment of the present application does not constitute a specific limitation on the electronic device. In other embodiments of the present application, the electronic device can include more or fewer components than the diagram, or combine certain components, or split certain components, or different component arrangements. The components shown in the diagram can be implemented in hardware, software, or a combination of software and hardware.

[0086] The processor can include one or more processing units, such as: the processor can include a central processing unit (CPU), an application processor (AP), a modem processor, a graphics processing unit (GPU), an image signal processor (ISP), a controller, a memory, a video code, a digital signal processor (DSP), a baseband processor, and / or a neural-network processing unit (NPU), etc. Among them, different processing units can be independent devices, or can be integrated in one or more processors.

[0087] Among them, the processor can be the nerve center and command center of the electronic device. The controller can generate operation control signals according to instruction operation codes and timing signals, complete the control of fetching instructions and executing instructions.

[0088] The processor can also be provided with a memory for storing instructions and data. In some embodiments, the memory in the processor is a cache memory. The memory can save instructions or data that the processor has just used or repeatedly uses. If the processor needs to use the instructions or data again, it can be directly called from the memory. Avoiding repeated access, reducing the waiting time of the processor, thus improving the efficiency of the system.

[0089] The external memory interface can be used to connect an external storage card, such as a MicroSD card, to realize the expansion of the storage capacity of the electronic device. The external storage card communicates with the processor through the external memory interface to realize the data storage function. For example, files such as music and video are saved in the external storage card.

[0090] The internal memory can be used to store computer executable program code, which includes instructions. The processor executes various functional applications and data processing of the electronic device by running the instructions stored in the internal memory. The internal memory can include a program storage area and a data storage area. The internal memory can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, a universal flash storage (UFS), etc.

[0091] The wireless communication function of the electronic device can be realized through an antenna, a wireless communication module, a modem processor, and a baseband processor, etc.

[0092] The wireless communication module can provide solutions for wireless local area networks (WLAN) (such as wireless fidelity (Wi-Fi) networks), Bluetooth (BT), global navigation satellite systems (GNSS), frequency modulation (FM), near field communication (NFC), infrared technology (IR), and the like.

[0093] The electronic device can implement audio functions and the like through an audio module, a speaker, a receiver, a microphone, a headset interface, an application processor, and the like.

[0094] The electronic device can implement a photographing function through an ISP, a camera, a video codec, a GPU, a display screen, an application processor, and the like.

[0095] The electronic device can implement a display function through a GPU, a display screen, an application processor, and the like.

[0096] The GPU is a microprocessor for image processing, connected to a display screen and an application processor. The GPU is used to perform mathematical and geometric calculations for graphics rendering. The processor can include one or more GPUs that execute program instructions to generate or change display information.

[0097] The display screen is used to display images, videos, and the like. The display screen includes a display panel.

[0098] In the storage medium provided in the present application, a program product capable of realizing the method for correcting and merging the road and bridge engineering bidding file is stored.

[0099] The method for correcting and merging the road and bridge engineering bidding file includes: receiving a road and bridge engineering original bidding file and tenderer format requirement data, and synchronizing the tenderer format requirement data to a road and bridge engineering sub-scene format library; performing basic element extraction and road and bridge industry exclusive element identification on the received road and bridge engineering original bidding file, and outputting an analysis result containing industry element information; calling rules of the road and bridge engineering sub-scene format library, performing road and bridge industry special format correction on the analysis result, and outputting a first preliminary correction result; using a machine learning model to correct the analysis result, and obtaining a second preliminary correction result; fusing the first preliminary correction result and the second preliminary correction result, obtaining no-conflict data, and merging the no-conflict data according to a road and bridge engineering bidding file standard logical sequence to generate a road and bridge engineering target bidding file.

[0100] In some possible implementation manners, the subject name bridge engineering bidding document revision and finalization method and system of the present disclosure can be implemented in the form of a program product, which includes program codes for causing a terminal device to perform the steps described in the above "Exemplary Method" section of the present specification according to various exemplary embodiments of the present disclosure when the program product is run on the terminal device.

[0101] The storage medium of the present disclosure can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium may, for example, be but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination thereof. More specific examples (non-exhaustive list) of the readable storage medium include an electrical connection having one or more wires, a portable disc, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0102] The above description of disclosed embodiments enables one of ordinary skill in the art to make or use the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for revising and compiling tender documents for road and bridge engineering projects, characterized in that, include: Receive the original tender documents and the format requirements data of the tendering party for road and bridge engineering projects, and synchronize the format requirements data of the tendering party to the sub-library of road and bridge engineering scenarios. The system extracts basic elements and identifies industry-specific elements from the received original tender documents for road and bridge engineering projects, and outputs parsing results containing industry element information. The rules of the road and bridge engineering scenario-specific format sub-library are called to correct the parsing results for road and bridge industry-specific formats, and the first preliminary correction result is output. A machine learning model is used to correct the analysis results, resulting in a second preliminary corrected result; By integrating the first preliminary correction result and the second preliminary correction result, conflict-free data is obtained, and the conflict-free data is merged according to the standard logical order of road and bridge engineering tender documents to generate the target tender document for road and bridge engineering.

2. The method for revising and compiling road and bridge engineering tender documents according to claim 1, characterized in that, The rules of the road and bridge engineering scenario-specific format sub-library are called to perform road and bridge industry-specific format correction on the parsed results, and output the first preliminary correction result, including: Based on the analysis results, the location information of basic elements and road and bridge industry-specific elements in the original tender documents of road and bridge projects is determined; Based on the type of basic elements and road and bridge industry-specific elements and the engineering scenario, the corresponding rules are called from the road and bridge engineering scenario-specific format sub-library; Compare the actual format of basic elements and road and bridge industry-specific elements with the standard format according to the judgment logic in the rules, and output the compliance or non-compliance result and the type of non-compliance; Select a correction scheme that matches the type of violation, correct the violation result, and obtain the first preliminary correction result.

3. The method for revising and compiling road and bridge engineering tender documents according to claim 1, characterized in that, The step of using a machine learning model to correct the analysis results and obtain a second preliminary corrected result includes: For the quantity table element type in the original tender documents of road and bridge projects, the Transformer model is selected to perform semantic recognition of table column names to avoid omissions due to different column name expressions in the sub-library of road and bridge project scenarios. For the types of construction image elements in the original tender documents of road and bridge projects, a target detection model is selected to identify whether the images contain numbered watermarks and areas with insufficient resolution. For the process flow chart element type in the original tender documents of road and bridge engineering: use graph neural networks to learn the logic of road and bridge construction processes and help identify process logic errors.

4. The method for revising and compiling road and bridge engineering tender documents according to claim 3, characterized in that, For the quantity table element type in the original tender documents of road and bridge engineering projects, the Transformer model is selected to perform semantic recognition of table column names to avoid missed judgments due to different column name expressions in the sub-library of road and bridge engineering scenario formats, including: For the input quantity table, use pandas to read the header row of the quantity table and get the column name text; The obtained column name text is input into the Transformer model, which outputs the confidence scores of multiple categories and selects the category with the highest confidence score as the predicted category. The Transformer model has the following structure: input layer, BERT pre-trained layer (freezing the first 6 layers to retain the underlying semantic features), pooling layer (outputting a 768-dimensional vector representing the overall semantics of the column name), dropout layer, first fully connected layer (compressing the feature dimension), second fully connected layer (corresponding to 12 standard column name categories), and output layer (using the Softmax function to output the confidence score of each category). After the Transformer model outputs the predicted category, it calls the road and bridge engineering scenario-specific format sub-library to perform column name correction. Specifically, if the Transformer model identifies the project code as a list number, and the standard format requirement for the list number in the road and bridge engineering scenario-specific format sub-library is 12-digit encoding, it checks whether the current project code format meets the 12-digit encoding requirement. If it does not meet the requirement, a correction prompt will automatically pop up, indicating that the column name project code belongs to the list number category and needs to be supplemented with a 12-digit encoding. If it does meet the requirement, it is marked as semantically compliant with column name, and proceeds to the subsequent table column order correction stage.

5. The method for revising and compiling road and bridge engineering tender documents according to claim 3, characterized in that, For the construction image elements in the original tender documents of road and bridge engineering projects, an object detection model is selected to identify whether the images contain numbered watermarks and areas with insufficient resolution, including: The input construction images are normalized in size, standardized in pixels, and noise suppressed to obtain the target construction images. The ResNet-50 network is used to extract deep visual features from the target construction image, and a sliding window is applied to the feature map to generate candidate regions containing the construction target. For the generated candidate regions containing construction targets, first align the features, then classify them as either numbered watermarked or insufficient resolution regions and fine-tune the bounding boxes to adapt to the precise positioning requirements of the construction targets. For the classification results, retain the target boxes whose category probability is greater than the preset probability threshold, and remove overlapping boxes using non-maximum suppression. Output the bounding boxes and categories of the numbered watermark and the region with insufficient resolution.

6. The method for revising and compiling road and bridge engineering tender documents according to claim 3, characterized in that, For the process flow chart elements in the original tender documents for road and bridge engineering: Graph neural networks are used to learn the logic of road and bridge construction processes, assisting in the identification of process logic errors, including: Use OCR to identify the process text boxes in the process flow diagram and use them as the node set of the target diagram, and identify the arrow lines between the process text boxes and use them as the edge set of the target diagram. By integrating the node set and edge set, the process flow diagram is transformed into structured data consisting of nodes and edges; The structured data of nodes and edges is input into the input layer of the GAT graph neural network, and a linear transformation is performed on the node features to enhance the expression of node features. Calculate the attention weight of a node to its neighboring nodes, and based on this attention weight, aggregate the features of the neighboring nodes and update the node's features; The updated final features of all nodes are processed by global mean pooling to obtain the graph-level features of the entire process flow diagram. The graph-level features are then input into the fully connected layer to output the probability distribution of logical error types.

7. The method for revising and compiling road and bridge engineering tender documents according to claim 6, characterized in that, Calculate the attention weights of a node to its neighboring nodes, and based on these attention weights, aggregate the features of the neighboring nodes to update the node's features, including: The formula for calculating the correlation of node pairs is as follows: ； In the formula, a represents the attention weight vector, and Z i Z j Representing nodes respectively and neighboring nodes Features after linear transformation Z represents i and Z j splicing; The formula for normalizing attention weights is as follows: ； In the formula, N( i ) is a node v i The set of all neighboring nodes; The feature update formula for a node is as follows: ； In the formula, This is the activation function.

8. A system for revising and compiling tender documents for road and bridge engineering projects, characterized in that, include: The input module is used to receive the original tender documents and the format requirements of the tendering party for road and bridge engineering projects, and to synchronize the format requirements of the tendering party to the sub-library of road and bridge engineering scenarios. The element parsing module is used to extract basic elements and identify industry-specific elements from the received original tender documents for road and bridge engineering projects, and output parsing results containing industry element information. The correction module is used to call the rules of the road and bridge engineering scenario-specific format sub-library, correct the parsing results for road and bridge industry-specific formats, and output the first preliminary correction result; A machine learning model is used to correct the analysis results, resulting in a second preliminary corrected result; The merge module is used to merge the first preliminary correction result and the second preliminary correction result to obtain conflict-free data, and merge the conflict-free data according to the standard logical order of road and bridge engineering tender documents to generate the target tender document for road and bridge engineering.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method for revising and compiling road and bridge engineering tender documents as described in any one of claims 1 to 7.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for revising and compiling road and bridge engineering tender documents as described in any one of claims 1 to 7.

Images