Consultation service and evaluation system for whole-process project management

Abstract

The invention relates to the technical field of project management, and discloses a consultation service and evaluation system for whole-process project management, and the system is characterized in that an information input module is used for inputting basic information of a consultation project by a user through a Web interface or an application program interface; the data pre-processing module is used for performing data pre-processing on the basic information of the project, cleaning and standardizing the information, and obtaining pre-processed project information; the scheme generation module is used for generating a project resource configuration scheme by adopting a particle swarm algorithm according to the preprocessed project information, and collecting project process data according to data nodes of the project resource configuration scheme; the risk assessment module is used for inputting the project process data into a risk assessment model, predicting the risk of each stage of the project through the risk assessment model, and outputting a prediction result; the data feedback module is used for generating risk coping strategy suggestions based on the prediction result, integrating data generated in the evaluation process and feeding back the data to the user; the project success rate is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of project management, and in particular to a consulting service and evaluation system for full-process project management. Background Art

[0002] With the expansion of project scale, the increase of complexity and the prevalence of global cooperation, traditional project management methods are facing unprecedented challenges. Project team members are distributed in different regions, information transmission is delayed, and communication costs are rising, resulting in inefficient decision-making and frequent misunderstandings. In a complex and changing project environment, it is a major problem to accurately assess resource requirements in real time and optimize allocation strategies. During the project execution, various changes may lead to many unforeseen risks. Traditional methods often rely on experience and judgment and lack a systematic risk identification, assessment and response mechanism. Summary of the invention

[0003] The purpose of the present invention is to solve the above problems and to design a consulting service and evaluation system for full-process project management.

[0004] The present invention provides a consulting service and evaluation system for full-process project management, which includes an information input module, a data preprocessing module, a solution generation module, a risk assessment module and a data feedback module, wherein:

[0005] An information input module is used for a user to input basic information of a consulting project through a web interface or an application program interface, wherein the basic information of the project includes a project name, budget, expected completion date, and key milestones;

[0006] A data preprocessing module is used to perform data preprocessing on the basic information of the project, clean and standardize the information, and obtain preprocessed project information;

[0007] A scheme generation module is used to generate a project resource allocation scheme based on the preprocessed project information by using a particle swarm algorithm, and to collect project progress data based on the data nodes of the project resource allocation scheme, wherein the project progress data includes project progress data, cost data, and working time data;

[0008] A risk assessment module, used to input the project process data into a risk assessment model, predict the risks of each stage of the project through the risk assessment model, and output the prediction results;

[0009] The data feedback module is used to generate risk response strategy suggestions based on the prediction results and integrate the data generated during the evaluation process to feedback to the user.

[0010] Optionally, in a first implementation of the present invention, the data preprocessing module includes a data cleaning submodule, an initial filling submodule, a further filling submodule and a data standardization submodule, wherein:

[0011] The data cleaning submodule is used to obtain basic information of the project, perform data cleaning on the basic information of the project to remove duplicate data, and obtain cleaned basic information;

[0012] The initial filling submodule is used to detect missing values ​​in the cleaned basic information and to initially fill the missing values ​​in the cleaned basic information using the average ratio method;

[0013] The further filling submodule is used to further fill the missing values ​​through the collaborative prediction method after clustering processing using the clustering algorithm to obtain the basic information after filling;

[0014] The data standardization submodule is used to perform data standardization processing on the filled basic information by using minimum-maximum normalization to obtain the preprocessed project information.

[0015] Optionally, in a second implementation of the present invention, the initial filling submodule includes a detection unit, a selection unit, a first calculation unit, a second calculation unit and a first filling unit, wherein:

[0016] A detection unit, used to perform data missing value detection on the cleaned basic information to obtain a detection result, wherein the detection result includes the detected missing value;

[0017] A selection unit, configured to select a to-be-filled attribute variable according to the detection result, wherein the to-be-filled attribute variable has at least one missing value;

[0018] The first calculation unit is used to select a variable with a complete value and the highest correlation with the attribute variable to be filled C1 as the reference attribute variable C2, and calculate the average ratio corresponding to the variable based on the attribute variable to be filled and the reference attribute variable

[0019] The second calculation unit is used to calculate the filling value of the cleaned basic information according to the average ratios C1 / C2 and C2 / C1 to obtain the filling value and

[0020] The first filling unit is used to repeatedly fill the missing values ​​in the cleaned basic information to complete the initial filling process of the missing values.

[0021] Optionally, in a third implementation of the present invention, the further filling submodule includes a clustering processing unit, a similarity calculation unit, a third calculation unit, an acquisition unit and a second filling unit, wherein:

[0022] A clustering processing unit, used for clustering the basic information after the initial filling processing by using a fuzzy C-means clustering algorithm to obtain a clustering result;

[0023] A similarity calculation unit is used to calculate t based on the clustering result. pq Indicates the similarity between two data points in a cluster:

[0024]

[0025] In the formula, O p,q represents the task set, T p,i 、T q,i They represent the bias values ​​corresponding to task i, object p, and object q respectively, represents the average bias value of object p and object q in all tasks;

[0026] A third calculation unit is used to store the calculation result in a result data set, and select i labels with the maximum similarity in the result data set to construct a nearest neighbor set by taking the object similarity as an indicator;

[0027] An acquisition unit is used to obtain a filling value G of the basic information after the initial filling process in the nearest neighbor set. p,i :

[0028]

[0029] The second padding unit is used to further pad the basic information after the initial padding process according to the padding value to obtain the padded basic information.

[0030] Optionally, in a fourth implementation of the present invention, the solution generation module includes an initialization submodule, a first update submodule, a second update submodule and a judgment submodule, wherein:

[0031] The initialization submodule is used to initialize the population and initial parameters, calculate the objective function value of each particle as the fitness value, and initialize the fitness value of each particle and the fitness value of the initial population;

[0032] The first updating submodule is used to calculate the similarity of each particle and the clustering degree of the population in the contemporary era, and update the cognitive parameters, wherein the cognitive parameters at least include inertia weight, individual cognitive parameters, social cognitive parameters and flight cognitive parameters

[0033] The second update submodule is used to update the population to update the particle speed and position, calculate the fitness value of the updated population, and compare the fitness of the current population with that of the previous generation. If the new solution dominates the old solution, the fitness value is replaced. If the new solution is dominated by the old solution, the fitness value of the previous generation is used. If the new and old solutions do not dominate each other, a random choice is made as to whether to replace them.

[0034] The judgment submodule is used to update the population fitness value using an adaptive grid strategy, to determine whether the current number of iterations has reached the maximum number of iterations, if not, to continue iterating, if satisfied, to output the optimal solution, and to generate a project resource allocation plan based on the optimal solution.

[0035] Optionally, in a fifth implementation of the present invention, the risk assessment module includes an input submodule, a convolution submodule, an information interaction submodule, a promotion submodule and an output submodule, wherein:

[0036] An input submodule, used for inputting the project process data into a risk assessment model, wherein the risk assessment model is constructed based on a Transformer network framework;

[0037] The convolution submodule is used to extract the spatial features of the project process data through convolution kernel and sliding convolution operations, and obtain the spatial receptive field and deeper spatial features through the superposition operation of multiple CNN convolution layers;

[0038] The information interaction submodule is used to take the spatial feature sequence extracted by the CNN convolution operation as input, and extract the temporal features in the spatial feature matrix at time t through the information interaction between the input gate, forget gate, and output gate in the LSTM, as well as the iterative transmission of the unit state;

[0039] The enhancement submodule is used to fuse and reduce the dimensionality of the time series features after LSTM processing, process the dependency between features at any position in the input sequence through the self-attention mechanism, and use the encoding and decoding structure to improve the expression ability of high-dimensional feature output;

[0040] The output submodule is used to obtain multi-classification results based on the output high-dimensional features using the Softmax function to output the prediction results of the risk assessment model.

[0041] Optionally, in a sixth implementation of the present invention, the method for implementing a consulting service and evaluation system for full-process project management includes the following steps:

[0042] The user inputs basic information of the consulting project through a web interface or an application program interface, wherein the basic information of the project includes project name, budget, expected completion date, and key milestones;

[0043] Performing data preprocessing on the basic information of the project, cleaning and standardizing the information, and obtaining preprocessed project information;

[0044] Generate a project resource allocation plan using a particle swarm algorithm based on the preprocessed project information, and collect project progress data based on data nodes of the project resource allocation plan, wherein the project progress data includes project progress data, cost data, and work hour data;

[0045] Inputting the project process data into the risk assessment model, predicting the risks of each stage of the project through the risk assessment model, and outputting the prediction results;

[0046] Generate risk response strategy recommendations based on the prediction results, and integrate the data generated during the evaluation process to feed back to the user.

[0047] Optionally, in a seventh implementation of the present invention, performing data preprocessing on the basic information of the project, cleaning and standardizing the information, and obtaining preprocessed project information includes:

[0048] Obtaining basic information of the project, performing data cleaning on the basic information of the project to eliminate duplicate data, and obtaining cleaned basic information;

[0049] Perform data missing value detection on the cleaned basic information, and use the average ratio method to initially fill in the missing values ​​in the cleaned basic information;

[0050] After clustering processing using clustering algorithm, the missing values ​​are further filled through collaborative prediction method to obtain basic information after filling;

[0051] The minimum-maximum normalization is used to perform data standardization on the filled basic information to obtain the preprocessed project information.

[0052] Optionally, in an eighth implementation of the present invention, the generating a project resource configuration scheme by using a particle swarm algorithm according to the preprocessed project information, and collecting project process data according to data nodes of the project resource configuration scheme include:

[0053] Initialize the population and initial parameters, calculate the objective function value of each particle as the fitness value, initialize the fitness value of each particle and the fitness value of the initial population;

[0054] Calculate the similarity and population clustering of each particle in the current era, and update the cognitive parameters, where the cognitive parameters at least include inertia weight, individual cognitive parameters, social cognitive parameters, and flight cognitive parameters.

[0055] Perform population update to update particle speed and position, calculate the fitness value of the updated population, and compare the fitness of the current population with that of the previous generation. If the new solution dominates the old solution, the fitness value is replaced. If the new solution is dominated by the old solution, the fitness value of the previous generation is used. If the new and old solutions do not dominate each other, randomly choose whether to replace them.

[0056] Adopt adaptive grid strategy to update population fitness value, judge whether the current number of iterations reaches the maximum number of iterations, if not, continue to iterate, if satisfied, output the optimal solution, and generate project resource allocation plan based on the optimal solution.

[0057] In the technical solution provided by the present invention, the user inputs basic information of the consulting project through a Web interface or an application program interface; data preprocessing is performed on the basic information of the project, and the information is cleaned and standardized to obtain preprocessed project information; a project resource allocation plan is generated based on the preprocessed project information using a particle swarm algorithm, and project process data is collected based on data nodes of the project resource allocation plan; the project process data is input into a risk assessment model, and the risks of each stage of the project are predicted through the risk assessment model, and the prediction results are output; risk response strategy recommendations are generated based on the prediction results, and the data generated in the assessment process are integrated and fed back to the user; the present invention realizes the reasonable allocation and efficient use of resources, reduces waste, effectively reduces the impact of uncertainty factors in the project execution process, and improves the project success rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the following detailed description of the preferred embodiment.The drawings are only for the purpose of illustrating the preferred embodiments and are not to be construed as limiting the invention.

[0059] Figure 1 A schematic diagram of the structure of a consulting service and evaluation system for full-process project management provided by an embodiment of the present invention;

[0060] Figure 2 A schematic diagram of the structure of a data preprocessing module of a consulting service and evaluation system for full-process project management provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0061] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, device, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0062] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 The structural diagram of the consulting service and evaluation system for the whole process project management provided by the embodiment of the present invention includes an information input module, a data preprocessing module, a solution generation module, a risk assessment module and a data feedback module, wherein:

[0063] The information input module 101 is used for a user to input basic information of a consulting project through a web interface or an application program interface, wherein the basic information of the project includes a project name, budget, expected completion date, and key milestones;

[0064] The data preprocessing module 102 is used to perform data preprocessing on the basic information of the project, clean and standardize the information, and obtain the preprocessed project information;

[0065] A scheme generation module 103 is used to generate a project resource allocation scheme using a particle swarm algorithm according to the preprocessed project information, and collect project process data according to the data nodes of the project resource allocation scheme, wherein the project process data includes project progress data, cost data, and working time data;

[0066] The risk assessment module 104 is used to input the project process data into the risk assessment model, predict the risks of each stage of the project through the risk assessment model, and output the prediction results;

[0067] The data feedback module 105 is used to generate risk response strategy suggestions based on the prediction results and integrate the data generated during the evaluation process to feedback to the user.

[0068] In this embodiment, please refer to Figure 2 As shown, the data preprocessing module 102 includes a data cleaning submodule, an initial filling submodule, a further filling submodule and a data standardization submodule, wherein:

[0069] The data cleaning submodule 1021 is used to obtain basic information of the project, perform data cleaning on the basic information of the project to remove duplicate data, and obtain cleaned basic information;

[0070] The initial filling submodule 1022 is used to detect missing values ​​in the cleaned basic information and to initially fill the missing values ​​in the cleaned basic information using the average ratio method;

[0071] A further filling submodule 1023 is used to further fill in the missing values ​​after clustering processing using a clustering algorithm through a collaborative prediction method to obtain basic information after filling;

[0072] The data standardization submodule 1024 is used to perform data standardization processing on the filled basic information by using minimum-maximum normalization to obtain pre-processed project information.

[0073] In this embodiment, the initial filling submodule includes a detection unit, a selection unit, a first calculation unit, a second calculation unit and a first filling unit, wherein:

[0074] A detection unit, used to perform data missing value detection on the cleaned basic information to obtain a detection result, wherein the detection result includes the detected missing value;

[0075] A selection unit, used for selecting attribute variables to be filled according to the detection result, wherein there is at least one missing value in the attribute variables to be filled;

[0076] The first calculation unit is used to select a variable with a complete value and the highest correlation with the attribute variable to be filled C1 as the reference attribute variable C2, and calculate the average ratio corresponding to the variable based on the attribute variable to be filled and the reference attribute variable

[0077] The second calculation unit is used to calculate the filling value of the cleaned basic information according to the average ratios C1 / C2 and C2 / C1 to obtain the filling value and

[0078] The first filling unit is used to repeatedly fill the missing values ​​in the cleaned basic information to complete the initial filling process of the missing values.

[0079] In this embodiment, the further filling submodule includes a clustering processing unit, a similarity calculation unit, a third calculation unit, an acquisition unit and a second filling unit, wherein:

[0080] A clustering processing unit, used for clustering the basic information after the initial filling processing by using a fuzzy C-means clustering algorithm to obtain a clustering result;

[0081] Similarity calculation unit, used based on clustering results, assuming t pq Indicates the similarity between two data points in a cluster:

[0082]

[0083] In the formula, O p,q represents the task set, T p,i 、T q,i They represent the bias values ​​corresponding to task i, object p, and object q respectively, represents the average bias value of object p and object q in all tasks;

[0084] The third calculation unit is used to store the calculation result in the result data set, and select the i labels with the maximum similarity in the result data set to construct the nearest neighbor set by taking the object similarity as an indicator;

[0085] The acquisition unit is used to obtain the filling value G of the basic information after the initial filling process in the nearest neighbor set. p,i :

[0086]

[0087] The second padding unit is used to further pad the basic information after the initial padding process according to the padding value to obtain the padded basic information.

[0088] In this embodiment, the solution generation module includes an initialization submodule, a first update submodule, a second update submodule and a judgment submodule, wherein:

[0089] The initialization submodule is used to initialize the population and initial parameters, calculate the objective function value of each particle as the fitness value, and initialize the fitness value of each particle and the fitness value of the initial population;

[0090] The first update submodule is used to calculate the similarity of each particle and the clustering degree of the population in the contemporary era, and update the cognitive parameters, where the cognitive parameters at least include inertia weight, individual cognitive parameters, social cognitive parameters and flight cognitive parameters

[0091] The second update submodule is used to update the population to update the particle speed and position, calculate the fitness value of the updated population, and compare the fitness of the current population with that of the previous generation. If the new solution dominates the old solution, the fitness value is replaced. If the new solution is dominated by the old solution, the fitness value of the previous generation is used. If the new and old solutions do not dominate each other, a random choice is made as to whether to replace them.

[0092] The judgment submodule is used to update the population fitness value using an adaptive grid strategy, to determine whether the current number of iterations has reached the maximum number of iterations, if not, to continue iterating, if satisfied, to output the optimal solution, and to generate a project resource allocation plan based on the optimal solution.

[0093] In this embodiment, the risk assessment module includes an input submodule, a convolution submodule, an information interaction submodule, a promotion submodule and an output submodule, wherein:

[0094] The input submodule is used to input the project process data into the risk assessment model, where the risk assessment model is built based on the Transformer network framework;

[0095] The convolution submodule is used to extract the spatial features of the project process data through convolution kernel and sliding convolution operations, and obtain the spatial receptive field and deeper spatial features through the superposition of multiple CNN convolution layers;

[0096] The information interaction submodule is used to take the spatial feature sequence extracted by the CNN convolution operation as input, and extract the temporal features in the spatial feature matrix at time t through the information interaction between the input gate, forget gate, and output gate in the LSTM, as well as the iterative transmission of the unit state;

[0097] The enhancement submodule is used to fuse and reduce the dimensionality of the time series features after LSTM processing, process the dependency between features at any position in the input sequence through the self-attention mechanism, and use the encoding and decoding structure to improve the expression ability of high-dimensional feature output;

[0098] The output submodule is used to obtain multi-classification results based on the output high-dimensional features using the Softmax function to output the prediction results of the risk assessment model.

[0099] Through the implementation of the above scheme, the system includes an information input module, a data preprocessing module, a scheme generation module, a risk assessment module and a data feedback module; the present invention realizes the rational allocation and efficient utilization of resources, reduces waste, effectively reduces the impact of uncertain factors in the project execution process, and improves the project success rate.

[0100] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A consulting service and evaluation system for full-process project management, characterized by: The consulting service and evaluation system for the whole process project management includes an information input module, a data preprocessing module, a solution generation module, a risk assessment module and a data feedback module, wherein: An information input module is used for a user to input basic information of a consulting project through a web interface or an application program interface, wherein the basic information of the project includes a project name, budget, expected completion date, and key milestones; A data preprocessing module is used to perform data preprocessing on the basic information of the project, clean and standardize the information, and obtain preprocessed project information; A scheme generation module is used to generate a project resource allocation scheme based on the preprocessed project information by using a particle swarm algorithm, and to collect project progress data based on the data nodes of the project resource allocation scheme, wherein the project progress data includes project progress data, cost data, and working time data; A risk assessment module, used to input the project process data into a risk assessment model, predict the risks of each stage of the project through the risk assessment model, and output the prediction results; The data feedback module is used to generate risk response strategy suggestions based on the prediction results and integrate the data generated during the evaluation process to feedback to the user.

2. A consulting service and evaluation system for full-process project management as claimed in claim 1, characterized in that: The data preprocessing module includes a data cleaning submodule, an initial filling submodule, a further filling submodule and a data standardization submodule, wherein: The data cleaning submodule is used to obtain basic information of the project, perform data cleaning on the basic information of the project to remove duplicate data, and obtain cleaned basic information; The initial filling submodule is used to detect missing values ​​in the cleaned basic information and to initially fill the missing values ​​in the cleaned basic information using the average ratio method; The further filling submodule is used to further fill the missing values ​​through the collaborative prediction method after clustering processing using the clustering algorithm to obtain the basic information after filling; The data standardization submodule is used to perform data standardization processing on the filled basic information by using minimum-maximum normalization to obtain the preprocessed project information.

3. A consulting service and evaluation system for full-process project management as claimed in claim 2, characterized in that: The initial filling submodule includes a detection unit, a selection unit, a first calculation unit, a second calculation unit and a first filling unit, wherein: A detection unit, used to perform data missing value detection on the cleaned basic information to obtain a detection result, wherein the detection result includes the detected missing value; A selection unit, configured to select a to-be-filled attribute variable according to the detection result, wherein the to-be-filled attribute variable has at least one missing value; The first calculation unit is used to select a variable with a complete value and the highest correlation with the attribute variable to be filled C1 as the reference attribute variable C2, and calculate the average ratio corresponding to the variable based on the attribute variable to be filled and the reference attribute variable The second calculation unit is used to calculate the filling value of the cleaned basic information according to the average ratios C1 / C2 and C2 / C1 to obtain the filling value and The first filling unit is used to repeatedly fill the missing values ​​in the cleaned basic information to complete the initial filling process of the missing values.

4. A consulting service and evaluation system for full-process project management as claimed in claim 2, characterized in that: The further filling submodule includes a cluster processing unit, a similarity calculation unit, a third calculation unit, an acquisition unit and a second filling unit, wherein: A clustering processing unit, used for clustering the basic information after the initial filling processing by using a fuzzy C-means clustering algorithm to obtain a clustering result; A similarity calculation unit is used to calculate t based on the clustering result. pq Indicates the similarity between two data points in a cluster: In the formula, O p,q represents the task set, T p,i , T q,i They represent the bias values ​​corresponding to task i, object p, and object q respectively, represents the average bias value of object p and object q in all tasks; A third calculation unit is used to store the calculation result in a result data set, and select i labels with the maximum similarity in the result data set to construct a nearest neighbor set by taking the object similarity as an indicator; An acquisition unit is used to obtain a filling value G of the basic information after the initial filling process in the nearest neighbor set. p,i : The second padding unit is used to further pad the basic information after the initial padding process according to the padding value to obtain the padded basic information.

5. A consulting service and evaluation system for full-process project management as claimed in claim 1, characterized in that: The solution generation module includes an initialization submodule, a first update submodule, a second update submodule and a judgment submodule, wherein: The initialization submodule is used to initialize the population and initial parameters, calculate the objective function value of each particle as the fitness value, and initialize the fitness value of each particle and the fitness value of the initial population; The first updating submodule is used to calculate the similarity of each particle and the clustering degree of the population in the contemporary era, and update the cognitive parameters, wherein the cognitive parameters at least include inertia weight, individual cognitive parameters, social cognitive parameters and flight cognitive parameters The second update submodule is used to update the population to update the particle speed and position, calculate the fitness value of the updated population, and compare the fitness of the current population with that of the previous generation. If the new solution dominates the old solution, the fitness value is replaced. If the new solution is dominated by the old solution, the fitness value of the previous generation is used. If the new and old solutions do not dominate each other, a random choice is made as to whether to replace them. The judgment submodule is used to update the population fitness value using an adaptive grid strategy, to determine whether the current number of iterations has reached the maximum number of iterations, if not, to continue iterating, if satisfied, to output the optimal solution, and to generate a project resource allocation plan based on the optimal solution.

6. A consulting service and evaluation system for full-process project management as claimed in claim 1, characterized in that: The risk assessment module includes an input submodule, a convolution submodule, an information interaction submodule, a promotion submodule and an output submodule, wherein: An input submodule, used for inputting the project process data into a risk assessment model, wherein the risk assessment model is constructed based on a Transformer network framework; The convolution submodule is used to extract the spatial features of the project process data through convolution kernel and sliding convolution operations, and obtain the spatial receptive field and deeper spatial features through the superposition operation of multiple CNN convolution layers; The information interaction submodule is used to take the spatial feature sequence extracted by the CNN convolution operation as input, and extract the temporal features in the spatial feature matrix at time t through the information interaction between the input gate, forget gate, and output gate in the LSTM, as well as the iterative transmission of the unit state; The enhancement submodule is used to fuse and reduce the dimensionality of the time series features after LSTM processing, process the dependency between features at any position in the input sequence through the self-attention mechanism, and use the encoding and decoding structure to improve the expression ability of high-dimensional feature output; The output submodule is used to obtain multi-classification results based on the output high-dimensional features using the Softmax function to output the prediction results of the risk assessment model.

7. A consulting service and evaluation system for full-process project management as claimed in claim 1, characterized in that: The method of implementing the consulting service and evaluation system of the whole process project management includes the following steps: The user inputs basic information of the consulting project through a web interface or an application program interface, wherein the basic information of the project includes project name, budget, expected completion date, and key milestones; Performing data preprocessing on the basic information of the project, cleaning and standardizing the information, and obtaining preprocessed project information; Generate a project resource allocation plan using a particle swarm algorithm based on the preprocessed project information, and collect project progress data based on data nodes of the project resource allocation plan, wherein the project progress data includes project progress data, cost data, and work hour data; Inputting the project process data into the risk assessment model, predicting the risks of each stage of the project through the risk assessment model, and outputting the prediction results; Generate risk response strategy recommendations based on the prediction results, and integrate the data generated during the evaluation process to feed back to the user.

8. A consulting service and evaluation system for full-process project management as claimed in claim 7, characterized in that: The data preprocessing of the basic information of the project, cleaning and standardizing the information, and obtaining the preprocessed project information includes: Obtaining basic information of the project, performing data cleaning on the basic information of the project to eliminate duplicate data, and obtaining cleaned basic information; Perform data missing value detection on the cleaned basic information, and use the average ratio method to initially fill in the missing values ​​in the cleaned basic information; After clustering processing using clustering algorithm, the missing values ​​are further filled through collaborative prediction method to obtain basic information after filling; The minimum-maximum normalization is used to perform data standardization on the filled basic information to obtain the preprocessed project information.

9. A consulting service and evaluation system for full-process project management as claimed in claim 7, characterized in that: The method of generating a project resource configuration scheme by using a particle swarm algorithm according to the preprocessed project information and collecting project process data according to the data nodes of the project resource configuration scheme includes: Initialize the population and initial parameters, calculate the objective function value of each particle as the fitness value, initialize the fitness value of each particle and the fitness value of the initial population; Calculate the similarity and population clustering of each particle in the current era, and update the cognitive parameters, where the cognitive parameters at least include inertia weight, individual cognitive parameters, social cognitive parameters, and flight cognitive parameters. Perform population update to update particle speed and position, calculate the fitness value of the updated population, and compare the fitness of the current population with that of the previous generation. If the new solution dominates the old solution, the fitness value is replaced. If the new solution is dominated by the old solution, the fitness value of the previous generation is used. If the new and old solutions do not dominate each other, randomly choose whether to replace them. Adopt adaptive grid strategy to update population fitness value, judge whether the current number of iterations reaches the maximum number of iterations, if not, continue to iterate, if satisfied, output the optimal solution, and generate project resource allocation plan based on the optimal solution.

Images