Project evaluation management system and method based on big data
The project evaluation management system is built through big data technology, which solves the problems of time-consuming, labor-intensive and error in manual evaluation, and realizes efficient and accurate project evaluation management, improving system performance and data security.
Patent Information
- Application Number
- CN202411030653.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-30
- Publication Date
- 2025-01-03
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing project evaluations mainly rely on manual completion, which leads to time-consuming and labor-intensive assessment and analysis errors, and lacks an efficient and accurate management system.
Design a project evaluation management system based on big data, including processing control module, management analysis module and transmission storage module. Through technologies such as preprocessing, word embedding, feature extraction, feature fusion, model training and blockchain storage, project evaluation data is optimized. Processing and analysis.
It effectively improves the efficiency and accuracy of project information analysis and processing, reduces system operation load, reduces analysis and processing errors, and improves the accuracy and data security of project evaluation.
Smart Images
Figure CN119250709A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of big data technology, and in particular to a project evaluation management system and method based on big data. Background Art
[0002] Project evaluation is an important part of the project management process. Project evaluation includes two stages, namely, evaluation before project establishment and acceptance evaluation after project completion. Through the two-stage evaluation, the project management process can be strictly implemented and the project implementation process is more standardized. Project evaluation is to evaluate the project before project establishment from four aspects: the preliminary preparation of the project, the basic conditions of the project, the necessity of project construction and the feasibility of the project, based on the content of the project application form and other relevant materials provided by the project application user. This step determines whether the project can be approved. The main task of the acceptance evaluation after the project is completed is to conduct project performance evaluation, and to accept and evaluate the project from aspects such as the project construction effect, technical effect, network and information security, and system usage. Through project evaluation, projects can be effectively managed in a unified manner to promote the development of network information construction. However, at present, project evaluation is completed manually, which not only takes a lot of time and energy, but also has evaluation and analysis errors. Therefore, it is very necessary to design a project evaluation management system and method based on big data with high efficiency and high accuracy. Summary of the invention
[0003] The purpose of the present invention is to provide a project evaluation management system and method based on big data to solve the problems raised in the above background technology.
[0004] In order to solve the above technical problems, the present invention provides the following technical solutions: a project evaluation management system based on big data, comprising: a processing control module, a management analysis module and a transmission storage module, the processing control module is connected to the management analysis module through a network, the management analysis module is connected to the transmission storage module through a network, the processing control module is used for processing and controlling project evaluation, the management analysis module is used for optimizing management and analysis of project evaluation, and the transmission storage module is used for transmitting and storing data information.
[0005] According to the above technical solution, the processing control module includes a preprocessing module, a word embedding module and a feature extraction module. The preprocessing module is network-connected to the word embedding module, and the word embedding module is network-connected to the feature extraction module. The preprocessing module is used to preprocess project evaluation data information, the word embedding module is used to perform word embedding management of project evaluation data information, and the feature extraction module is used to perform feature extraction control of data information.
[0006] According to the above technical solution, the management and analysis module includes a feature fusion module, a network layer module and a training and analysis module. The feature fusion module is connected to the network layer module, and the network layer module is connected to the training and analysis module. The feature fusion module is used to perform fusion control of feature information, the network layer module is used to perform network layer control for project evaluation, and the training and analysis module is used to perform training and analysis processing of the model.
[0007] According to the above technical solution, the transmission and storage module includes an information transmission module and a backup storage module. The information transmission module and the backup storage module are connected to the network. The information transmission module is used for transmission management of project evaluation information, and the backup storage module is used for backup storage of data information.
[0008] According to the above technical solution, the project evaluation management method based on big data includes the following steps:
[0009] Step S1: During the project evaluation process of the enterprise, the evaluation preprocessing of the project data information is performed;
[0010] Step S2: After completing the preprocessing of project data information evaluation, word embedding and feature extraction control are performed;
[0011] Step S3: further optimizing and analyzing the project evaluation;
[0012] Step S4: Transmission and backup storage control of the data information of the completed project evaluation.
[0013] According to the above technical solution, step S1 further includes the following steps:
[0014] After receiving the entered project information, the project information evaluation is preprocessed. Jieba word segmentation is used through the selected dictionary to divide the project information text into words, and then all the words are put into the set vocabulary list for corresponding word segmentation. At the same time, the stop word list is used to remove stop words from the project information. It is determined whether each word in the vocabulary list is in the stop word list. If it is in the stop word list, the word is deleted.
[0015] According to the above technical solution, step S2 further includes the following steps:
[0016] Step S21: After completing the preprocessing of the project information evaluation, the Skip-gram model is used to perform word embedding and convert the project information into a recognizable vector representation. Specifically, the Skip-gram model treats each word in the project information as a central word and predicts the context of the entire text. In this way, the word vector of the context is continuously optimized. A sliding window is used so that each prediction and analysis behavior occurs within the sliding window. The calculation formula of the Skip-gram model prediction probability P is as follows:
[0017]
[0018] Where, t = im,...,i,...,i+m, m is the sliding window size, w i Indicates the current word, e(w i ) is the current vector word representation, w t w i Context words, and im≤t≤i+m;
[0019] Step S22: Use the BiLSTM model to extract the deep semantic features of the text, input the text vector matrix obtained after text embedding into the BiLSTM model, and the input sequence enters the forward LSTM and backward LSTM models respectively. In the LSTM model, the current input is set as the data as the forget gate. After determining the information to be discarded through the forget gate processing, the extracted feature data is obtained.
[0020] According to the above technical solution, step S3 further includes the following steps:
[0021] Step S31: After the feature extraction of the project information is completed through the BiLSTM model, the feature extraction is further performed through the CNN model, and the text feature vectors obtained by the CNN and BiLSTM models are vector-concatenated. The dimension of the text feature vector after concatenation will increase, so the dimension is further reduced through a 1×1 convolutional neural network, and the concatenated text feature vector is input into the 1×1 convolutional neural network. The number of output channels of the convolutional neural network is set to 100. After the concatenated vector passes through the convolution layer, a fused text feature vector is obtained;
[0022] Step S32: After the text feature vectors are concatenated by the CNN and BiLSTM models, the attention mechanism is further used to perform model optimization processing;
[0023] Step S33: After embedding the text through the BERT model to obtain a text representation vector containing rich semantic information, the BiLSTM+Attention model is used to fuse the text context semantic information, optimize the text feature vector, and obtain the final text feature vector.
[0024] According to the above technical solution, step S4 further includes the following steps:
[0025] Step S41: transmitting the feature vector of the project information evaluation process to the workbench for further evaluation and analysis by the staff;
[0026] Step S42: The project evaluation information that has completed vector processing is backed up and stored on the blockchain to effectively ensure data information security.
[0027] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention, by providing a processing control module, a management analysis module and a transmission storage module, can effectively improve the efficiency and accuracy of the analysis and processing of project information, and reduce the operating load of the system for project information analysis and processing. It can also effectively improve the accuracy of project information evaluation and avoid the generation of analysis and processing errors; through the attention mechanism, the project information evaluation processing model can be more focused and fully analyze and process text feature information, reduce the focus on non-feature information, and improve the performance of the model. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:
[0029] Figure 1 It is a schematic diagram of the system module composition of the present invention. DETAILED DESCRIPTION
[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0031] See also Figure 1 The present invention provides a technical solution: a project evaluation management system based on big data, which includes: a processing control module, a management analysis module and a transmission storage module. The processing control module is connected to the management analysis module through a network, and the management analysis module is connected to the transmission storage module through a network. The processing control module is used for processing and controlling project evaluation, the management analysis module is used for optimizing management and analysis of project evaluation, and the transmission storage module is used for transmitting and storing data information.
[0032] The processing control module includes a preprocessing module, a word embedding module and a feature extraction module. The preprocessing module is connected to the word embedding module through a network, and the word embedding module is connected to the feature extraction module through a network. The preprocessing module is used for preprocessing project evaluation data information, the word embedding module is used for word embedding management of project evaluation data information, and the feature extraction module is used for feature extraction control of data information.
[0033] The management and analysis module includes a feature fusion module, a network layer module and a training and analysis module. The feature fusion module is connected to the network layer module, and the network layer module is connected to the training and analysis module. The feature fusion module is used to perform fusion control of feature information, the network layer module is used to perform network layer control for project evaluation, and the training and analysis module is used to perform training and analysis processing of the model.
[0034] The transmission and storage module includes an information transmission module and a backup storage module. The information transmission module and the backup storage module are connected to the network. The information transmission module is used for transmission management of project evaluation information, and the backup storage module is used for backup storage of data information.
[0035] The project evaluation management method based on big data includes the following steps:
[0036] Step S1: During the project evaluation process of the enterprise, the evaluation preprocessing of the project data information is performed;
[0037] Step S2: After completing the preprocessing of project data information evaluation, word embedding and feature extraction control are performed;
[0038] Step S3: further optimizing and analyzing the project evaluation;
[0039] Step S4: Transmission and backup storage control of the data information of the completed project evaluation.
[0040] Step S1 further comprises the following steps:
[0041] After receiving the input project information, preprocessing for project information evaluation is carried out. Jieba segmentation is performed using the selected dictionary to divide the project information text into words. Then all the words are put into the set vocabulary for corresponding segmentation processing. At the same time, the stop word list is used to remove stop words from the project information. It is judged whether each word in the vocabulary is in the stop word list. If it is in the stop word list, the word is deleted. In this step, segmentation is a process of first dispersing and then aggregating. First, the text is divided into words, and then all the words are put into the vocabulary. And in the project information text, the text usually contains some meaningless modal particles, auxiliary words, etc., such as words like "de", "le", "a", "ne", etc. Retaining these stop words not only has no effect but also increases the analysis and operation load. Therefore, through this step, the analysis and processing efficiency and accuracy of project information can be effectively improved, and the operation load of the system for project information analysis and processing can be reduced.
[0042] Step S2 further includes the following steps:
[0043] Step S21: After completing the preprocessing of the project information evaluation, word embedding is performed through the Skip-gram model to convert the project information into a recognizable vector representation. Specifically, the Skip-gram model treats each word in the project information as a central word once, predicts the context of the whole text, and continuously optimizes the word vectors of the context in this way. A sliding window is adopted so that every prediction and analysis behavior occurs within the sliding window. The calculation formula for the prediction probability P of the Skip-gram model is as follows:
[0044]
[0045] In the formula, t = i - m,..., i,..., i + m, m is the size of the sliding window, w i represents the current word, e(w i ) is the current vector word representation, w t is the w i context word, and i - m ≤ t ≤ i + m; Since traditional one-hot encoding sorts the vocabulary in the corpus according to certain rules, assuming the size of the corpus is 100, then the dimension of the generated word vector is 100. If the corpus is larger, the dimension of the word vector will also be larger. When using such word vectors in the project information evaluation process, it will greatly increase the analysis volume of the project information evaluation, and using one-hot encoding lacks semantic information of the context, seriously affecting the accuracy of the project information. Therefore, through this step, the accuracy of the project information evaluation can be effectively improved, and the generation of analysis and processing errors can be avoided;
[0046] Step S22: Use the BiLSTM model to extract the deep semantic features of the text, input the text vector matrix obtained after text embedding into the BiLSTM model, and the input sequence enters the forward LSTM and backward LSTM models respectively. In the LSTM model, the current input is set as the data as the forget gate. After determining the information to be discarded through the forget gate processing, the extracted feature data is obtained.
[0047] Step S3 further comprises the following steps:
[0048] Step S31: After the feature extraction of the project information is completed through the BiLSTM model, the feature extraction is further performed through the CNN model, and the text feature vectors obtained by the CNN and BiLSTM models are vector-concatenated. The dimension of the text feature vector after concatenation will increase, so the dimension is further reduced through the 1×1 convolutional neural network, and the concatenated text feature vector is input into the 1×1 convolutional neural network. The number of output channels of the convolutional neural network is set to 100. After the concatenated vector passes through the convolution layer, the fused text feature vector is obtained; in this step, the 1×1 convolutional neural network can change the feature dimension by changing the number of output channels;
[0049] Step S32: After the concatenation of the text feature vectors is completed by the CNN and BiLSTM models, the attention mechanism is further used to optimize the model; the attention mechanism enables the model to focus on and fully analyze and process the text feature information, reduce the focus on non-feature information, and improve the performance of the model;
[0050] Step S33: After embedding the text through the BERT model to obtain a text representation vector containing rich semantic information, the BiLSTM+Attention model is used to fuse the text context semantic information, optimize the text feature vector, and obtain the final text feature vector.
[0051] Step S4 further comprises the following steps:
[0052] Step S41: The feature vector of the project information evaluation and processing is transmitted to the workbench for further evaluation and analysis by the staff, which effectively improves the efficiency and accuracy of the staff's evaluation of the project information;
[0053] Step S42: The project evaluation information that has completed vector processing is backed up and stored on the blockchain to effectively ensure data information security.
[0054] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0055] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. Project evaluation management system based on big data, including: The processing control module, the management and analysis module and the transmission and storage module are characterized in that: the processing control module and the management and analysis module are connected to the network, the management and analysis module and the transmission and storage module are connected to the network, the processing control module is used to perform processing control of project evaluation, the management and analysis module is used to perform optimization management and analysis of project evaluation, and the transmission and storage module is used to transmit and store data information.
2. The project evaluation management system based on big data according to claim 1 is characterized in that: The processing control module includes a preprocessing module, a word embedding module and a feature extraction module. The preprocessing module is network-connected to the word embedding module, and the word embedding module is network-connected to the feature extraction module. The preprocessing module is used to preprocess project evaluation data information, the word embedding module is used to perform word embedding management of project evaluation data information, and the feature extraction module is used to perform feature extraction control of data information.
3. The project evaluation management system based on big data according to claim 1 is characterized in that: The management and analysis module includes a feature fusion module, a network layer module and a training and analysis module. The feature fusion module is connected to the network layer module, and the network layer module is connected to the training and analysis module. The feature fusion module is used to perform fusion control of feature information, the network layer module is used to perform network layer control for project evaluation, and the training and analysis module is used to perform training and analysis processing of the model.
4. The project evaluation management system based on big data according to claim 1 is characterized in that: The transmission and storage module includes an information transmission module and a backup storage module. The information transmission module and the backup storage module are connected to the network. The information transmission module is used for transmission management of project evaluation information, and the backup storage module is used for backup storage of data information.
5. The project evaluation management system based on big data according to claim 1 is characterized in that: The project evaluation management method based on big data comprises the following steps: Step S1: During the project evaluation process of the enterprise, the evaluation preprocessing of the project data information is performed; Step S2: After completing the preprocessing of project data information evaluation, word embedding and feature extraction control are performed; Step S3: further optimizing and analyzing the project evaluation; Step S4: Transmission and backup storage control of the data information of the completed project evaluation.
6. The project evaluation management method based on big data according to claim 5 is characterized by: The step S1 further comprises the following steps: After receiving the entered project information, the project information evaluation is preprocessed. Jieba word segmentation is used through the selected dictionary to divide the project information text into words, and then all the words are put into the set vocabulary list for corresponding word segmentation. At the same time, the stop word list is used to remove stop words from the project information. It is determined whether each word in the vocabulary list is in the stop word list. If it is in the stop word list, the word is deleted.
7. The project evaluation management method based on big data according to claim 5 is characterized by: The step S2 further comprises the following steps: Step S21: After completing the preprocessing of the project information evaluation, the Skip-gram model is used to perform word embedding and convert the project information into a recognizable vector representation. Specifically, the Skip-gram model treats each word in the project information as a central word and predicts the context of the entire text. In this way, the word vector of the context is continuously optimized. A sliding window is used so that each prediction and analysis behavior occurs within the sliding window. The calculation formula of the Skip-gram model prediction probability P is as follows: Where, t = im,...,i,...,i+m, m is the sliding window size, w i Indicates the current word, e(w i is the current vector word representation, w t w i Context words, and im≤t≤i+m; Step S22: Use the BiLSTM model to extract the deep semantic features of the text, input the text vector matrix obtained after text embedding into the BiLSTM model, and the input sequence enters the forward LSTM and backward LSTM models respectively. In the LSTM model, the current input is set as the data as the forget gate. After determining the information to be discarded through the forget gate processing, the extracted feature data is obtained.
8. The project evaluation management method based on big data according to claim 5 is characterized by: The step S3 further comprises the following steps: Step S31: After the feature extraction of the project information is completed through the BiLSTM model, the feature extraction is further performed through the CNN model, and the text feature vectors obtained by the CNN and BiLSTM models are vector-concatenated. The dimension of the text feature vector after concatenation will increase, so the dimension is further reduced through a 1×1 convolutional neural network, and the concatenated text feature vector is input into the 1×1 convolutional neural network. The number of output channels of the convolutional neural network is set to 100. After the concatenated vector passes through the convolution layer, a fused text feature vector is obtained; Step S32: After the concatenation of the text feature vectors is completed by the CNN and BiLSTM models, the attention mechanism is further used to optimize the model; the attention mechanism enables the model to focus on and fully analyze and process the text feature information, reduce the focus on non-feature information, and improve the performance of the model; Step S33: After embedding the text through the BERT model to obtain a text representation vector containing rich semantic information, the BiLSTM+Attention model is used to fuse the text context semantic information, optimize the text feature vector, and obtain the final text feature vector.
9. The project evaluation management method based on big data according to claim 5 is characterized by: The step S4 further comprises the following steps: Step S41: transmitting the feature vector of the project information evaluation process to the workbench for further evaluation and analysis by the staff; Step S42: The project evaluation information that has completed vector processing is backed up and stored on the blockchain to effectively ensure data information security.