Search

CN-121981100-A - Project node data processing method and related equipment based on contract deviation analysis

CN121981100ACN 121981100 ACN121981100 ACN 121981100ACN-121981100-A

Abstract

The application provides a project item data processing method and related equipment based on contract deviation analysis, the method comprises the steps of obtaining contract deviation analysis results and project attribute data of a target project, determining complexity coefficients of the target project based on the number of tasks, the number of participating departments and the number of delivery categories in the project attribute data, determining a project item summary template frame of the target project according to the complexity coefficients, wherein the project item summary template frame comprises a plurality of combinations of content modules and different complexity coefficients correspond to different combinations, converting contract deviation analysis results into natural language description texts by using a natural language generation model, establishing mapping relations between the natural language description texts and the content modules in the project item summary template frame, and further filling the natural language description texts into the corresponding content modules to generate project item summary data. The application can improve the pertinence of the project junction report content and the bearing degree of effective information.

Inventors

  • LI XIAOLONG
  • ZHANG JIANPO

Assignees

  • 北京神州光大科技有限公司

Dates

Publication Date
20260505
Application Date
20260128

Claims (10)

  1. 1. The project node data processing method based on contract deviation analysis is characterized by comprising the following steps of: Acquiring contract deviation analysis results of a target project and project attribute data, wherein the project attribute data comprises the number of tasks, the number of participating departments and the number of delivery types; Determining a complexity coefficient of the target item based on the number of tasks, the number of participating departments, and the number of delivery categories; determining a junction item summary template frame of the target item according to the complexity coefficient, wherein the junction item summary template frame comprises a plurality of combinations of content modules, and different complexity coefficients correspond to different combinations; Converting the contract deviation analysis result into a natural language description text by using a natural language generation model; establishing a mapping relation between the natural language description text and a plurality of content modules in the junction item summary template frame; And filling the natural language description text into a corresponding content module by utilizing the mapping relation to generate project junction item summary data.
  2. 2. The method for processing project summary data based on contract bias analysis according to claim 1, wherein after determining the project summary template frame of the target project according to the complexity coefficient, further comprising: acquiring the project budget amount and the project planning period of the target project; If the project budget amount is greater than or equal to a preset amount threshold value or the project planning period is greater than or equal to a preset period threshold value, splitting a content module representing deviation analysis in the junction item summarization template frame to obtain a progress deviation analysis sub-module, a cost deviation analysis sub-module and a quality deviation analysis sub-module; And if the project budget amount is smaller than the preset amount threshold and the project planning period is smaller than the preset period threshold, performing module simplifying processing on at least one content module in the item summary template frame.
  3. 3. The method for processing project summary data based on contract bias analysis according to claim 1, wherein after determining the project summary template frame of the target project according to the complexity coefficient, further comprising: Extracting an industry type tag of the target item from the item attribute data, wherein the industry type tag comprises one of software development, construction engineering and medical service; searching an industry exclusive template matched with the industry type label in a preset industry template library; Adding the target content module in the industry-specific template to a combination of a plurality of content modules of the tie summary template frame.
  4. 4. The method for processing project node data based on contract bias analysis according to claim 1, wherein the step of obtaining the result of contract bias analysis for the target project includes: Acquiring original deviation values of the target item in a plurality of dimensions, wherein the dimensions comprise a progress dimension, a cost dimension, a quality dimension, a risk dimension and a resource allocation dimension; carrying out numerical normalization processing on the original deviation values to obtain normalized deviation indexes corresponding to each dimension; Determining a target weight coefficient corresponding to each normalized deviation index based on the project stage and the project type of the target project; and carrying out weighted summation on the normalized deviation indexes by using the target weight coefficient to obtain the item health index of the target item, wherein the contract deviation analysis result comprises the item health index.
  5. 5. The method for processing project node data based on contract bias analysis according to claim 4, wherein the determining a target weight coefficient corresponding to each of the normalized bias indexes based on a project stage and a project type of the target project includes: Calling an initial weight coefficient from a preset historical project weight configuration database according to the industry type of the target project; Vectorizing the item type, the item stage, the current deviation degree of the target item and the historical deviation restoration effect to generate a state space vector, wherein the current deviation degree and the historical deviation restoration effect are determined based on a plurality of normalized deviation indexes; Inputting the state space vector to a preset deep reinforcement learning model to obtain a weight adjustment amplitude for each normalized deviation index; and correcting the initial weight coefficient by using the weight adjustment amplitude to obtain a target weight coefficient corresponding to each normalized deviation index.
  6. 6. The method for processing project node data based on contract deviation analysis according to claim 4, wherein after performing numerical normalization processing on the original deviation values to obtain normalized deviation indexes corresponding to each dimension, the method further comprises: Constructing a causal graph model between a plurality of project execution influence factors of the target project and the normalized deviation index; identifying confounding factors in the causal graph model using a backdoor adjustment criterion, the confounding factors being variables that affect both the project execution influencing factors and the normalized deviation indicators; determining the causal contribution degree of each project execution influence factor to the normalized deviation index by controlling the confounding factors; sorting the execution influence factors of the plurality of items according to the causal contribution degree; And determining a contract deviation root cause of the target item based on the sorting result of the project execution influence factors, wherein the contract deviation analysis result comprises the contract deviation root cause.
  7. 7. The method for project node data processing based on contract bias analysis according to claim 6, further comprising, before constructing a causal graph model between a plurality of project execution influencing factors of the target project and the normalized bias indexes: the method comprises the steps of obtaining structured execution data of a target item and unstructured management text, wherein the unstructured management text comprises meeting summary and incoming and outgoing mails; extracting semantic features of the unstructured management text by using a preset pre-training language model to obtain text semantic feature vectors, and carrying out association rule mining processing on the structured execution data to obtain structured data association feature vectors; Vector splicing and fusion are carried out on the text semantic feature vector and the structured data association feature vector, and a multi-mode evidence chain vector is obtained; and determining a plurality of project execution influence factors based on the multi-modal evidence chain vector.
  8. 8. A project organization data processing system based on contract bias analysis, characterized in that the project organization data processing system based on contract bias analysis is operated in an electronic device for executing the project organization data processing method based on contract bias analysis according to any one of claims 1 to 7.
  9. 9. An electronic device comprising a memory, a processor and a computer program stored on the memory, the processor executing the computer program to implement the project junction data processing method based on contract bias analysis of any one of claims 1 to 7.
  10. 10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program configured to be executed by a processor to implement the item node data processing method based on contract bias analysis according to any one of claims 1 to 7.

Description

Project node data processing method and related equipment based on contract deviation analysis Technical Field The application relates to the technical field of project items, in particular to a project item data processing method and related equipment based on contract deviation analysis. Background With the penetration of digital transformation, the project junction item is increasingly valued as a core link for evaluating the performance quality and sedimentation experience of a project contract. Currently, enterprises commonly use project management systems to collect execution data, combine contract terms to perform basic acceptance analysis, and generate junction item documents to assist performance assessment and subsequent decisions so as to aim at improving the standardization level of project management. In the related art, the scheme for generating the junction report is mainly based on a preset general template for data filling, and a statistical chart and index values are presented through a fixed format. The processing logic following the static framework often causes the analysis module of the key dimension to be missing or the key to be fuzzy, or causes a large number of redundant empty items to appear in the junction item report, so that the pertinence of the report content and the bearing degree of the effective information are reduced. Disclosure of Invention The embodiment of the application provides a project node data processing method and related equipment based on contract deviation analysis, aiming at improving the pertinence of project node report contents and the bearing degree of effective information. In a first aspect, an embodiment of the present application provides a method for processing project node data based on contract deviation analysis, where the method for processing project node data based on contract deviation analysis includes: Acquiring contract deviation analysis results of a target project and project attribute data, wherein the project attribute data comprises the number of tasks, the number of participating departments and the number of delivery types; Determining a complexity coefficient of the target item based on the number of tasks, the number of participating departments, and the number of delivery categories; determining a junction item summary template frame of the target item according to the complexity coefficient, wherein the junction item summary template frame comprises a plurality of combinations of content modules, and different complexity coefficients correspond to different combinations; Converting the contract deviation analysis result into a natural language description text by using a natural language generation model; establishing a mapping relation between the natural language description text and a plurality of content modules in the junction item summary template frame; And filling the natural language description text into a corresponding content module by utilizing the mapping relation to generate project junction item summary data. In the embodiment, the complexity coefficient is determined based on the project attribute data such as the task number, and the like, so that a junction item summarization template frame containing specific content module combinations is adapted, the contract deviation analysis result is converted into a natural language description text by using a natural language generation model and is mapped and filled into a corresponding module to generate project junction item summarization data, the dynamic adjustment of a report structure along with the project complexity and the accurate display of deviation data are realized, and the pertinence of report contents and the bearing degree of effective information are improved. In an embodiment, after determining the junction item summary template frame of the target item according to the complexity coefficient, the method further includes: acquiring the project budget amount and the project planning period of the target project; If the project budget amount is greater than or equal to a preset amount threshold value or the project planning period is greater than or equal to a preset period threshold value, splitting a content module representing deviation analysis in the junction item summarization template frame to obtain a progress deviation analysis sub-module, a cost deviation analysis sub-module and a quality deviation analysis sub-module; And if the project budget amount is smaller than the preset amount threshold and the project planning period is smaller than the preset period threshold, performing module simplifying processing on at least one content module in the item summary template frame. In the embodiment, the content module for representing the deviation analysis is split into a plurality of sub-modules or subjected to module reduction processing based on the project budget amount and the project planning period, so that the structural granularity o