Test project integrated management device

Abstract

The invention relates to the field of test data management, and discloses a test project integrated management device which comprises a data storage layer, a data access layer, a business logic layer, an application service layer and a safety control layer. Through a relational database technology, structured storage and disaster recovery of test item data are realized; the natural language processing module can automatically extract key information from the unstructured text, so that the data processing efficiency is improved; the machine learning module can predict test results based on test conditions and contents; the application service layer provides a graphical interface, facilitates user interaction, and can realize addition, deletion, modification and check of projects; the intelligent recommendation module can provide personalized suggestions for the user according to the prediction result; the security control layer ensures the security of the data through authority management and transmission encryption; according to the invention, the efficiency, accuracy and safety of test project management are significantly improved.

Description

Technical Field

[0001] The invention relates to the field of test data management, and in particular to a test project integrated management device. Background Art

[0002] With the advancement of science and technology and the increasing complexity of test projects, the management of test projects has become increasingly important. In traditional test project management, paper or simple electronic documents are usually used to record and track the progress of test projects. However, this method has problems such as difficult information retrieval, untimely data updates, and low information visualization.

[0003] Human processing capabilities for test text data are limited, which means that a large amount of key information, such as test leaders, test locations, test institutions, test conditions, etc., need to be manually identified and extracted, which is inefficient and prone to errors.

[0004] In addition, the existing test project management methods often lack intelligent recommendation functions and cannot provide users with optimization suggestions based on historical data and current test conditions, which makes users lack data support and scientific basis when formulating and adjusting test plans. To this end, we propose an integrated test project management device. Summary of the invention

[0005] The purpose of the present invention is to provide a test item integrated management device to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions: a test project integrated management device, including a data storage layer, a data access layer, a business logic layer, an application service layer and a security control layer;

[0007] The data storage layer includes a storage module and a disaster recovery module; the storage module uses relational database technology to store test project data; the disaster recovery module uses regular backup to retain data;

[0008] The data access layer includes an interface module and a connection pool management module; the interface module realizes data reading and writing by defining a standardized data access interface; the connection pool management module manages and reuses database connections through connection pool technology;

[0009] The business logic layer includes a natural language processing module and a machine learning module; the natural language processing module uses word segmentation, part-of-speech tagging and named entity recognition technology to parse the test project text, and is used to extract structured key information from the unstructured test text data, and the key information includes the test person in charge, test location, test organization, test conditions, test content, test information, test purpose and test results in the test text; the machine learning module uses a supervised learning algorithm to train a test result prediction model, and the prediction model is used to predict the test results in advance based on the experimental conditions and experimental content;

[0010] The application service layer includes a user interface module, a project management module and an intelligent recommendation module; the user interface module uses a graphical interface design technology to provide user interaction; the project management module is used to implement the addition, deletion, modification and query of test items; the intelligent recommendation module provides users with personalized test implementation suggestions based on the prediction results of the machine learning model;

[0011] The security control layer includes a rights management module and a transmission encryption module; the rights management module manages user rights through role-based access control; the transmission encryption module uses SSL / TLS encryption technology to perform security control on transmitted data.

[0012] Preferably, the natural language processing module uses word segmentation technology to segment the test project text and divide the continuous text into independent vocabulary units; further uses part-of-speech tagging technology to assign a corresponding part-of-speech tag to each vocabulary unit to determine its grammatical function in the sentence; finally, through named entity recognition technology, the test person in charge, test location, test organization, test conditions, test content, test information, test purpose and test results are identified and extracted from the segmented and part-of-speech tagged text.

[0013] Preferably, the machine learning module constructs a training data set using historical test project data, and trains a test result prediction model using a decision tree algorithm; the specific steps of training the prediction model are:

[0014] S1: Collect and preprocess historical test project data, extract test conditions, test content and corresponding test results as features, and organize them into feature data sets;

[0015] S2: The feature dataset is randomly divided into a training set and a validation set. The training set is used to build and train the model, and the validation set is used to evaluate the model performance.

[0016] S3: Use the decision tree algorithm to build the initial test result prediction model. The decision tree algorithm uses information gain as an indicator for selecting the optimal partition attribute. The calculation formula of information gain is:

[0017] Gain(A)=Info(D)-Info A (D)

[0018] Where Gain(A) represents the gain of attribute A, Info(D) represents the entropy of sample set D, and Info A (D) represents the entropy after partitioning D according to attribute A. The entropy calculation formula is:

[0019] Info(D)=-∑p(x)log2p(x)

[0020] p(x) is the proportion of the xth category in the sample set;

[0021] S4: Use the training set to train the model, select the attribute with the largest information gain as the partition attribute of the current node, and recursively construct the subtree until the stopping condition is met;

[0022] S5: Prune the constructed initial decision tree to avoid overfitting, and use the validation set to evaluate the performance of the pruned model. If the model performance meets the predetermined requirements, it is determined as the final test result prediction model.

[0023] Preferably, the intelligent recommendation module receives the test result prediction data output by the machine learning module, and selects matching test implementation suggestions from a preset recommendation strategy library based on the prediction results and current test conditions and content. The recommendation strategy library contains optimization suggestions and implementation strategies for a variety of test scenarios. These suggestions and strategies are classified and labeled according to the test type, test conditions and test objectives, and stored in a database for rapid retrieval and matching based on specific conditions.

[0024] Preferably, the disaster recovery module pre-sets a backup cycle and a backup time point, and at the backup time point, automatically exports the test project data in the storage module as a backup file, and stores the backup file in a designated backup storage location.

[0025] Preferably, the interface module defines a standardized data access interface, including a data reading interface and a data writing interface; for the data reading interface, a data reading request is received, and corresponding test project data is queried from a relational database according to parameters in the request; for the data writing interface, a write request containing test project data is received; the test project data in the write request is verified, and the verified test project data is written into the relational database;

[0026] The connection pool module pre-creates and maintains a certain number of active database connections; when receiving a connection request, it directly allocates unused available connections; after the connection is used, it is marked as unused, and is regularly checked to close inactive connections.

[0027] Preferably, the user interface module uses WEB technology to create an interactive display interface, and provides relevant functional options for test project management on the display interface, including creating a new project, viewing a project list, editing project information, deleting a project, and searching for a project;

[0028] When the user chooses to create a new project, a form is provided for the user to enter basic information of the project, including project name, project description, test person in charge, test location, test organization, test conditions, test content, test purpose and expected test results;

[0029] When the user chooses to view the project list, the information of all test projects is retrieved from the database and displayed in a list on the user interface. Each project item displays the basic information and status of the project;

[0030] When the user chooses to edit the project information, an editing form is provided for the user to modify the relevant information of the selected test project, and the modified information is updated in the database;

[0031] When the user chooses to delete an item, all information of the selected test item is deleted from the database, and the item list is updated on the user interface;

[0032] When the user selects a search item, a search box is provided for the user to enter keywords, and matching test items are retrieved from the database based on the keywords, and the results are displayed on the user interface.

[0033] Preferably, the display interface of the user interface module receives and displays the output results of the intelligent recommendation module.

[0034] Preferably, the project management module receives and processes user requests for adding, deleting, modifying and querying test projects through a user interface, interacts with a data storage layer to update or retrieve project information, and displays corresponding operation results on a display interface.

[0035] Preferably, the permission management module predefines user roles and assigns corresponding permissions, each role corresponding to a set of specific operation permissions and data access scope; when a user logs into the system, the user's identity is verified, and the access permissions are determined based on the user's role; when the user performs an operation, it is checked whether the user's permissions allow the operation to be performed. If allowed, the operation request is released, otherwise the request is rejected and an error prompt is returned; the permission management module also provides a permission modification function, allowing the administrator to adjust the permission allocation of roles according to actual needs.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] Through the relational database technology of the data storage layer, the structured storage of the test project data is realized, making the retrieval and update of information faster and more accurate; the regular backup function of the disaster recovery module ensures the security of the data and prevents data loss. The natural language processing module of the business logic layer uses word segmentation, part-of-speech tagging and named entity recognition technology to automatically extract structured key information from unstructured test text data, greatly improving the efficiency and accuracy of data processing and reducing the possibility of human errors. The test result prediction model trained by the learning module through the supervised learning algorithm can predict the test results in advance based on the experimental conditions and experimental content, providing users with scientific reference. At the same time, the intelligent recommendation module of the application service layer provides users with personalized test implementation suggestions based on the prediction results, further optimizing the test plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is the overall structure diagram of the present invention;

[0039] Figure 2 A diagram of the steps for parsing test text for the natural language processing module;

[0040] Figure 3 Diagram of the training steps for the model to predict test results. DETAILED DESCRIPTION

[0041] 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.

[0042] See also Figure 1-3 The present invention provides a technical solution: a test project integrated management device, including a data storage layer, a data access layer, a business logic layer, an application service layer and a security control layer. The various layers communicate through standardized interfaces to ensure smooth transmission and processing of data.

[0043] The data storage layer is the data foundation of the device, including storage modules and disaster recovery modules; the storage module adopts relational database technology, such as MySQL or PostgreSQL, to realize structured storage of test project data; the disaster recovery module ensures the security and recoverability of data through a regular backup mechanism.

[0044] The data access layer is responsible for data reading and writing operations, including the interface module and the connection pool management module; the interface module defines a standardized data access interface so that upper-level applications can seamlessly access underlying data; the connection pool management module manages and reuses database connections through connection pool technology to improve the efficiency and stability of data access.

[0045] The business logic layer is the core of this device, including the natural language processing module and the machine learning module. The natural language processing module uses word segmentation, part-of-speech tagging and named entity recognition technology to deeply analyze the text data of the test project and extract structured key information, such as the person in charge of the test, the test location, etc. This information provides rich materials for subsequent data analysis and intelligent recommendation.

[0046] The machine learning module uses a supervised learning algorithm to train a test result prediction model, which can predict possible test results in advance based on test conditions and test content, providing valuable reference for experimenters.

[0047] The application service layer is directly facing users, including user interface module, project management module and intelligent recommendation module; the user interface module adopts graphical interface design technology, such as HTML5 and CSS3, to provide users with an intuitive and friendly operation experience; the project management module realizes the functions of adding, deleting, modifying and querying test projects, making it convenient for users to comprehensively manage projects.

[0048] The intelligent recommendation module provides users with personalized test implementation suggestions based on the prediction results of the machine learning model, combined with the user's historical data and current test status. These suggestions may include optimizing test conditions, adjusting test plans, etc., aiming to improve the efficiency and success rate of the test.

[0049] The security control layer is responsible for the security of the device, including the permission management module and the transmission encryption module. The permission management module uses the role-based access control (RBAC) technology to accurately manage the access rights of each user to ensure the confidentiality and integrity of the data; the transmission encryption module uses SSL / TLS encryption technology to encrypt the data transmitted in the network to prevent the data from being illegally intercepted or tampered with.

[0050] The present invention will be further described below in conjunction with Examples 1 to 3:

[0051] Embodiment 1:

[0052] See attached Figure 2, the natural language processing module uses word segmentation technology to segment the test project text and divide the continuous text into independent vocabulary units; further uses part-of-speech tagging technology to assign corresponding part-of-speech tags to each vocabulary unit to determine its grammatical function in the sentence; finally, through named entity recognition technology, the test person in charge, test location, test organization, test conditions, test content, test information, test purpose and test results are identified and extracted from the segmented and part-of-speech tagged text. This module mainly processes the unstructured text data related to the test project through word segmentation technology, part-of-speech tagging technology and named entity recognition technology to extract structured key information. The following is the specific implementation method of this module:

[0053] Word segmentation technology: The natural language processing module first uses word segmentation technology to segment the test project text. Word segmentation is the process of dividing continuous text into independent vocabulary units. It is a basic step in Chinese text processing because Chinese text does not have obvious space separators like English. The implementation of word segmentation technology can adopt existing word segmentation algorithms, such as string matching-based word segmentation methods, statistical-based word segmentation methods, or deep learning-based word segmentation methods. In this module, it is preferred to use statistical-based word segmentation algorithms, such as hidden Markov models (HMM) or conditional random fields (CRF), which can better handle the complexity and ambiguity of Chinese text. Through word segmentation technology, the test project text is divided into independent vocabulary units, which provides a basis for subsequent part-of-speech tagging and named entity recognition.

[0054] Part-of-speech tagging technology: After word segmentation is completed, the natural language processing module further uses part-of-speech tagging technology to assign corresponding part-of-speech tags to each vocabulary unit. Part-of-speech tagging is the process of determining the grammatical function of vocabulary in a sentence, which helps to understand the semantics of the text more accurately. The implementation of part-of-speech tagging technology can adopt rule-based methods, statistical methods or deep learning methods. In this module, statistical-based part-of-speech tagging algorithms are preferred, such as maximum entropy models or deep learning models. These models can learn the part-of-speech rules of vocabulary through a large amount of training data and accurately tag them. Through part-of-speech tagging technology, each vocabulary unit is assigned a corresponding part-of-speech tag, such as noun, verb, adjective, etc., which helps the accuracy of subsequent named entity recognition.

[0055] Named Entity Recognition Technology: Finally, the natural language processing module uses named entity recognition technology to identify and extract key information from the text that has been segmented and POS-tagged. Named entity recognition is the process of identifying entity names with specific meanings in the text, such as names of people, places, and institutions. In this module, deep learning-based named entity recognition algorithms are used, such as the BiLSTM-CRF model or the Transformer model. These models can be trained with a large amount of labeled data to learn the characteristics and contextual information of the entities, thereby accurately identifying the named entities in the text. Through named entity recognition technology, the natural language processing module can accurately extract key information from the test project text, including the test person in charge, test location, test organization, test conditions, test content, test information, test purpose, and test results. This information is stored and displayed in a structured form, which greatly improves the efficiency and accuracy of test project management.

[0056] Here is a specific example showing how the Natural Language Processing module works on the pilot project text data:

[0057] Example of test project text: "This test was conducted by Engineer Li at the Beijing Institute of Science and Technology. The purpose of the test was to study the performance of new materials in high temperature environments. The test conditions were a temperature of 800 degrees Celsius and a duration of 2 hours. The test content included material heating, performance observation, and data recording. The final test results showed that the new material maintained good stability at high temperatures."

[0058] Natural language processing module processing steps:

[0059] Step 1: Word segmentation: Divide the continuous text into independent vocabulary units.

[0060] Word segmentation result: "This test was conducted by Engineer Li at Beijing Institute of Science and Technology. The purpose of the test was to study the performance of new materials in a high temperature environment. The test conditions were a temperature of 800 degrees Celsius and a duration of 2 hours. The test content included material heating, performance observation and data recording. The final test results showed that the new material maintained good stability at high temperatures."

[0061] Step 2: Part-of-speech tagging: Assign a corresponding part-of-speech tag to each vocabulary unit.

[0062] Part-of-speech tagging results: "This (pronoun) / test (noun) / by (preposition) / Engineer Li (noun) / in charge of (verb) / , (punctuation) / at (preposition) / Beijing Institute of Science and Technology (noun) / conducted (verb) / . (punctuation) / test (noun) / purpose (noun) / is (verb) / research (verb) / new materials (noun) / under (preposition) / high temperature (noun) / environment (noun) / under (local word) / of (particle) / performance (noun) / performance (noun) / . (punctuation) / test (noun) / condition (noun) / for (verb) / temperature (noun) / 800 degrees Celsius (quantifier) / , (punctuation) / last ( verb) / time (noun) / 2 (quantifier) / hour (noun) / . (punctuation) / test (noun) / content (noun) / include (verb) / material (noun) / heat (verb) / , (punctuation) / performance (noun) / observation (verb) / and (conjunction) / data (noun) / record (verb) / . (punctuation) / final (adverb) / test (noun) / result (noun) / show (verb) / , (punctuation) / new material (noun) / under (preposition) / high temperature (noun) / under (directional word) / maintain (verb) / had (particle) / good (adjective) / of (particle) / stability (noun) / . (punctuation)”

[0063] Step 3: Named Entity Recognition: Identify and extract key information from the text that has been segmented and POS-tagged.

[0064] Named entity recognition results:

[0065] Test leader: Engineer Li;

[0066] Experimental location: Beijing Institute of Science and Technology;

[0067] Test purpose: To study the performance of new materials in high temperature environment;

[0068] Test conditions: temperature 800 degrees Celsius, duration 2 hours;

[0069] Test contents: material heating, performance observation and data recording;

[0070] Test results: The new material maintains good stability at high temperatures;

[0071] Through this example, we can see how the natural language processing module extracts structured key information from unstructured test project text through word segmentation, part-of-speech tagging, and named entity recognition techniques.

[0072] Embodiment 2:

[0073] See attached Figure 3The machine learning module uses historical test project data to construct a training data set and trains a test result prediction model through a decision tree algorithm; the specific steps of training the prediction model are:

[0074] S1: Before starting training, you first need to prepare the training data. This data is the historical test project data after preprocessing and feature extraction, including test conditions, test content and corresponding test results. The data is organized into a format suitable for model training, such as CSV or TFRecord.

[0075] S2: In order to ensure that the model has good generalization ability and avoid overfitting, the prepared data needs to be randomly divided into a training set and a validation set. The training set is used to train the model, while the validation set is used to evaluate the performance of the model during the training process and make necessary adjustments.

[0076] Before training begins, the decision tree model needs to be initialized, including setting parameters such as the depth of the tree and the minimum number of samples for leaf nodes. The settings of these parameters will affect the complexity and prediction performance of the model.

[0077] S3: Use the decision tree algorithm to build the initial test result prediction model. The decision tree algorithm uses information gain as an indicator for selecting the optimal partition attribute. The calculation formula of information gain is:

[0078] Gain(A)=Info(D)-Info A (D)

[0079] Where Gain(A) represents the gain of attribute A, Info(D) represents the entropy of sample set D, and Info A (D) represents the entropy after partitioning D according to attribute A. The entropy calculation formula is:

[0080] Info(D)=-∑p(x)log2p(x)

[0081] p(x) is the proportion of the xth category in the sample set; the specific calculation steps are as follows:

[0082] 1. Statistical category distribution: First, perform category statistics on the samples in the data set D to determine the number of samples in each category.

[0083] 2. Calculate the category proportion: Based on the number of samples in each category, calculate the proportion p(x) of the category in the data set. The proportion is calculated by dividing the number of samples in the category by the total number of samples in the data set.

[0084] 3. Apply the entropy formula: Substitute the proportion of each category p(x) into the entropy calculation formula to calculate the entropy value of the data set. Here, a logarithmic function (usually with base 2) is used to calculate the entropy contribution of each category, and they are added together to get the entropy of the entire data set.

[0085] S4: Use the training set to train the model, select the attribute with the largest information gain as the partition attribute of the current node, and recursively build a subtree until the stopping condition is met; specifically, it includes the following three steps:

[0086] 1. Select the best partition attribute: For each attribute, calculate its information gain. Information gain is a measure based on entropy, which indicates how much "information" or "purity" can be improved by partitioning a data set according to a certain attribute. At each node of the decision tree, the algorithm calculates the information gain of all possible partition attributes and selects the attribute with the largest information gain as the partition attribute of the current node to ensure that the model can make decisions based on the most important features.

[0087] 2. Recursively construct subtrees: Divide the data set into subsets according to the selected partitioning attributes, and recursively perform the above process on each subset until the stopping condition is met (such as reaching the preset tree depth, the number of samples in the node is less than the preset threshold, etc.).

[0088] 3. Processing leaf nodes: When the recursion reaches a leaf node, the node will contain a group of samples with the same or similar features. At this time, the algorithm will calculate the average or mode of the test results of these samples and use it as the prediction result of the leaf node.

[0089] S5: After the decision tree is built, pruning is required to avoid overfitting caused by overly complex models. Pruning can start from the bottom and simplify the tree, reducing the complexity of the model by removing certain branches. During the pruning process, the predictive performance and complexity of the model are comprehensively considered to reach a balance point.

[0090] During the training process, the validation set is used to evaluate the model regularly. The accuracy of the model is evaluated by comparing the prediction performance of the model on the validation set with the consistency of the actual test results. If the model performance is poor or overfitting occurs, the model needs to be adjusted, such as adjusting the parameters of the decision tree, changing the selection strategy of the partition attribute, etc. When the model shows good performance on both the training set and the validation set, the model training can be considered complete, and the model parameters can be saved for subsequent use.

[0091] Here is an example of building and training a trial outcome prediction model using Python as the programming language:

[0092] from sklearn.tree import DecisionTreeClassifier

[0093] from sklearn.model_selection import train_test_split

[0094] from sklearn.metrics import accuracy_score

[0095] import pandas as pd

[0096] # Load data

[0097] #The data set is a CSV file containing features (feature1, feature2, ...) and labels (result)

[0098] data=pd.read_csv('trial_data.csv')

[0099] #Separate features and labels

[0100] X = data.drop('result', axis = 1) # Feature column

[0101] y = data['result'] #label column

[0102] # Divide into training set and test set

[0103] X_train, X_test, y_train, y_test=train_test_split(X, y, test_size=0.2, random_state=42)

[0104] # Initialize the decision tree classifier

[0105] #You can set the maximum depth of the decision tree, the minimum number of samples for leaf nodes, and other parameters

[0106] clf=DecisionTreeClassifier(max_depth=5, random_state=42)

[0107] #Train model

[0108] #Use the features and labels of the training set to train the decision tree model

[0109] clf.fit(X_train, y_train)

[0110] #Predict the results of the test set

[0111] #Use the trained model to predict the test set

[0112] y_pred = clf.predict(X_test)

[0113] #Evaluate model performance

[0114] #Calculate the accuracy of the model on the test set

[0115] accuracy=accuracy_score(y_test,y_pred)

[0116] print(f'Model accuracy:{accuracy}')

[0117] The above code first loads the test data stored in the CSV file through the pandas library. This data contains multiple features (such as test conditions, test content, etc.) and the corresponding test results. After loading the data, the features are separated from the labels (test results) for subsequent processing.

[0118] In order to evaluate the performance of the model, the original data set is divided into a training set and a test set. The training set is used to train the model, while the test set is used to verify the predictive ability of the model. The train_test_split function is used to randomly divide the data according to the set ratio (such as 80% for training set and 20% for test set).

[0119] Use the DecisionTreeClassifier class to initialize a decision tree classifier. During this process, you can set the relevant parameters of the decision tree, such as the maximum depth and the minimum number of samples in the leaf nodes, to control the complexity of the model. Next, use the features and labels of the training set to train the model so that the model can learn the rules and patterns in the data. After training, use the test set to predict the model. Through the predict method, the model will predict the samples in the test set according to the learned rules and output the prediction results. In order to evaluate the performance of the model, the prediction results are compared with the true labels of the test set to calculate the accuracy of the model. The accuracy is an important indicator to measure the correctness of the model's prediction. The higher the value, the stronger the prediction ability of the model.

[0120] The intelligent recommendation module receives the test result prediction data output by the machine learning module, and selects matching test implementation suggestions from a preset recommendation strategy library based on the prediction results and current test conditions and content. The recommendation strategy library contains optimization suggestions and implementation strategies for a variety of test scenarios. These suggestions and strategies are classified and labeled according to the test type, test conditions and test objectives, and stored in a database for rapid retrieval and matching based on specific conditions.

[0121] The main function of the intelligent recommendation module is to provide users with personalized test implementation suggestions based on the prediction results of the machine learning module. This module receives the test result prediction data output by the machine learning module, and selects matching test implementation suggestions from the preset recommendation strategy library in combination with the current test conditions and content, so as to assist users in optimizing the test plan and improving the test efficiency and success rate; the implementation examples of this module are as follows:

[0122] 1. Data Reception and Processing

[0123] 1. Receive prediction data: The intelligent recommendation module first receives the test result prediction data from the machine learning module. These data are the output of the prediction model trained by the supervised learning algorithm based on the experimental conditions and experimental content.

[0124] 2. Data preprocessing: Perform necessary preprocessing on the received prediction data, including data cleaning, formatting conversion, etc., to ensure the accuracy and availability of the data.

[0125] 2. Construction of Recommended Strategy Library

[0126] 1. Strategy collection and classification: Collect optimization suggestions and implementation strategies for various test scenarios. These suggestions and strategies can come from the successful experience of historical test projects, suggestions from industry experts, relevant literature research, etc. These suggestions and strategies should be classified according to the test type, test conditions and test purpose.

[0127] 2. Labeling and storage: Add corresponding labels to each recommended strategy, such as "improve test efficiency", "reduce costs", "optimize test conditions", etc., so that they can be quickly retrieved and matched according to specific needs later. Store these labeled recommended strategies in the database to form a recommended strategy library.

[0128] 3. Recommendation Algorithm Implementation

[0129] 1. Condition matching: According to the conditions and content of the current experiment and the prediction results of the machine learning module, the matching strategy is retrieved from the recommended strategy library. The matching process can consider a variety of factors, such as the type of experiment, the purpose of the experiment, the confidence of the prediction results, etc.

[0130] 2. Strategy sorting and selection: For multiple recommended strategies that are matched, they are sorted according to a certain sorting algorithm (such as based on the historical success rate of the strategy) and the best recommended strategy is selected as the output.

[0131] 3. Personalized recommendation: Combine the user's historical preferences and behavior data to personalize the recommendation strategy to meet the needs of different users. For example, for users who pay more attention to test efficiency, strategies that can shorten the test cycle can be recommended first.

[0132] Embodiment 3:

[0133] The disaster recovery module pre-sets the backup cycle and backup time point. At the backup time point, the test project data in the storage module is automatically exported as a backup file, and the backup file is stored in the specified backup storage location; the disaster recovery module can automatically export the test project data in the storage module as a backup file, and store these backup files in the specified backup storage location, thereby preventing data loss or damage. The user can set the backup cycle through the configuration interface of the disaster recovery module according to actual needs, such as daily, weekly or monthly, etc. This setting determines the frequency of the disaster recovery module to perform backup operations. Within the set backup cycle, the user can specify a specific backup time point, such as 2 a.m. every day or 10 p.m. every Sunday, etc. This setting ensures that the backup is performed during a period of low system load to reduce the impact on normal business operations. When the predetermined backup time point is reached, the disaster recovery module will automatically trigger the data export operation. By docking with the interface of the storage module, the disaster recovery module can obtain the latest test project data and export it to a standard backup file format, such as an SQL file, a CSV file or other user-specified format. The exported backup file will be automatically stored in the specified backup storage location. This location can be a local disk, a network storage device, or a cloud storage service, etc., depending on the user's configuration and needs. The disaster recovery module maintains a list of backup files, recording information such as the time, file name, and storage location of each backup. Users can view and manage these backup files through the management interface of the disaster recovery module. If necessary, users can select specific backup files for data recovery through the disaster recovery module. The disaster recovery module will assist users in importing data from backup files into the storage module to recover lost or damaged data.

[0134] The interface module defines a standardized data access interface, including a data reading interface and a data writing interface; for the data reading interface, a data reading request is received, and corresponding test project data is queried from a relational database according to the parameters in the request; for the data writing interface, a write request containing test project data is received; the test project data in the write request is verified, and the verified test project data is written into the relational database; the interface module defines a standardized data access interface so that upper-level applications can access the database in a unified and standardized manner.

[0135] Data reading interface implementation:

[0136] a. Receive data read requests: The interface module provides one or more data read interface functions or methods for upper-layer applications to call. These interfaces receive data read requests, which usually contain parameters such as query conditions, sorting methods, and return fields.

[0137] b. Query processing: The interface module constructs the corresponding SQL query statement based on the parameters in the request. This process needs to consider security issues such as SQL injection to ensure the security and legality of the query parameters.

[0138] c. Execute query: obtain database connection through the connection pool module and execute the constructed SQL query statement.

[0139] d. Result return: Encapsulate the query results into a standard data structure (such as JSON, XML, etc.) and return them to the upper-level application.

[0140] Data writing interface implementation:

[0141] a. Receive data write request: The interface module provides data write interface functions or methods. These interfaces receive write requests containing test item data.

[0142] b. Data verification: Before writing data, the interface module needs to verify the test item data in the request to ensure the integrity, legality and correctness of the format of the data.

[0143] c. Write: Obtain database connection through the connection pool module and write the verified test project data into the relational database. Possible data conflicts and exceptions need to be handled during the writing process.

[0144] d. Write feedback: returns the result of the write operation, including whether it is successful, error information, etc.

[0145] The connection pool module is responsible for managing database connections. It improves the efficiency and response speed of database access by pre-creating and maintaining a certain number of active database connections. When the system starts, the connection pool is initialized according to the configuration, including setting the maximum number of connections, the minimum number of connections, the connection timeout and other parameters of the connection pool. A specified number of database connections are created in advance and kept active. When the interface module requests a database connection, the connection pool module checks whether there is an available connection. If so, an unused connection is directly allocated to the interface module. When the interface module has finished using the connection, it needs to return the connection to the connection pool. The connection pool module marks the received connection as unused for subsequent requests. The connection pool module needs to regularly detect the connection status in the connection pool, close inactive or timed-out connections, and add new connections to maintain the size of the connection pool. When handling database connection exceptions, the connection pool module needs to have a retry mechanism and error handling capabilities to ensure the stability of the upper-layer application.

[0146] The user interface module is an important part of the application service layer in the test project integrated management device. Its main function is to provide an intuitive and easy-to-use graphical interface so that users can easily manage test projects. This module uses WEB technologies such as HTML, CSS and JavaScript to create an interactive display interface.

[0147] The user interface adopts a clear layout design to ensure that all functions and information are clear at a glance. The main areas include the project list display area, the operation button area, the form filling area, and the intelligent recommendation result display area. The interface provides a series of functional options, including "Create a new project", "View project list", "Edit project information", "Delete project" and "Search project". Each option corresponds to a specific function page or pop-up window. The functions are implemented as follows:

[0148] Create a new project: When the user selects "Create a new project", the interface jumps to a page containing a form. The form includes fields such as project name, project description, test person in charge, test location, test organization, test conditions, test content, test purpose and expected test results. After the user fills in the form, the data is stored in the database through the data access layer.

[0149] View Project List: When you select "View Project List", the user interface module retrieves the information of all test projects from the database through the data access layer and displays it in a list. Each project entry includes basic information such as project name, description, status, and operation buttons (such as edit, delete).

[0150] Edit project information: When the user clicks the "Edit" button of a project item, an edit form pops up containing the current information of the project. The user can modify the relevant information and submit the update to the database.

[0151] Delete item: When you click the "Delete" button, the system pops up a confirmation dialog box to ask the user whether to delete the item. After confirmation, all information of the item will be deleted from the database and the item list will be updated.

[0152] Search project: After entering keywords in the search box, the system retrieves matching test projects from the database and displays the results in a list on the user interface.

[0153] The user interface module is also responsible for receiving and displaying the output of the intelligent recommendation module. After the machine learning module predicts the test results based on the experimental conditions and content, the intelligent recommendation module will generate personalized test implementation suggestions. These suggestions will be presented to users in an intuitive way through the user interface module to help users optimize the test plan.

[0154] The project management module receives and processes user requests for adding, deleting, modifying and querying test items through the user interface, interacts with the data storage layer to update or retrieve project information, and displays the corresponding operation results on the display interface. The project management module receives user operation instructions through the user interface, interacts with the data storage layer after parsing and processing, realizes the update or retrieval of project information, and promptly feeds back the operation results to the user. In this process, the project management module strictly follows the security and authority control requirements to ensure that only authorized users are allowed to perform corresponding operations.

[0155] The permission management module predefines user roles and assigns corresponding permissions, and each role corresponds to a set of specific operation permissions and data access scopes; when a user logs into the system, the user's identity is verified and the access permissions are determined based on the user's role; when the user performs an operation, it is checked whether the user's permissions allow the operation to be performed. If allowed, the operation request is released, otherwise the request is rejected and an error prompt is returned; the permission management module also provides a permission modification function, allowing administrators to adjust the permission allocation of roles according to actual needs.

[0156] The permission management module first defines a series of user roles, such as administrator, ordinary user, visitor, etc., and assigns a set of specific operation permissions and data access scopes to each role. These permissions and scopes clearly define the operations that each role can perform in the system and the data that can be accessed. When a user tries to log in to the system, the permission management module will require the user to provide valid authentication information, such as username and password. The permission management module will compare the received authentication information with the user information stored in the system to verify the user's identity. If the authentication is successful, the module will assign the corresponding access rights to the user according to the user's role; if the authentication fails, the user's login will be denied and an error prompt will be returned. After the user logs in to the system, whenever the user tries to perform an operation (such as viewing, adding, modifying or deleting test items), the permission management module will perform a permission check. The permission management module will check the user's role and permissions to determine whether he is qualified to perform the operation. If the user has the corresponding permissions, the operation request will be released; otherwise, the request will be denied and an error prompt will be returned; in order to improve the security of the system, the permission management module will also perform secondary verification on sensitive operations, such as entering a verification code or performing fingerprint recognition. The permission management module provides a permission modification function, allowing administrators to adjust the permission allocation of roles according to actual needs. Administrators can add, delete or modify the permissions of existing roles through the user interface access rights management function. All permission modification operations will be recorded in the system log for subsequent problem tracking and auditing. At the same time, the module will also conduct a strict verification and confirmation process for permission modifications to prevent unauthorized modifications.

[0157] 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.

[0158] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A test project integrated management device, characterized in that: It includes data storage layer, data access layer, business logic layer, application service layer and security control layer; The data storage layer includes a storage module and a disaster recovery module; the storage module uses relational database technology to store test project data; the disaster recovery module uses regular backup to retain data; The data access layer includes an interface module and a connection pool management module; the interface module realizes data reading and writing by defining a standardized data access interface; the connection pool management module manages and reuses database connections through connection pool technology; The business logic layer includes a natural language processing module and a machine learning module; the natural language processing module uses word segmentation, part-of-speech tagging and named entity recognition technology to parse the test project text, and is used to extract structured key information from the unstructured test text data, and the key information includes the test person in charge, test location, test organization, test conditions, test content, test information, test purpose and test results in the test text; the machine learning module uses a supervised learning algorithm to train a test result prediction model, and the prediction model is used to predict the test results in advance based on the experimental conditions and experimental content; The application service layer includes a user interface module, a project management module and an intelligent recommendation module; The user interface module uses graphical interface design technology to provide user interaction; the project management module is used to realize the addition, deletion, modification and query of test items; the intelligent recommendation module provides users with personalized test implementation suggestions based on the prediction results of the machine learning model; The security control layer includes a rights management module and a transmission encryption module; The authority management module manages user permissions through role-based access control; the transmission encryption module uses SSL / TLS encryption technology to perform security control on transmitted data.

2. A test project integrated management device according to claim 1, characterized in that: The natural language processing module uses word segmentation technology to segment the test project text and divide the continuous text into independent vocabulary units; further uses part-of-speech tagging technology to assign a corresponding part-of-speech tag to each vocabulary unit to determine its grammatical function in the sentence; finally, through named entity recognition technology, the test person in charge, test location, test organization, test conditions, test content, test information, test purpose and test results are identified and extracted from the text that has been segmented and tagged with part-of-speech.

3. A test project integrated management device according to claim 2, characterized in that: The machine learning module constructs a training data set using historical test project data, and trains a test result prediction model using a decision tree algorithm; the specific steps for training the prediction model are: S1: Collect and preprocess historical test project data, extract test conditions, test content and corresponding test results as features, and organize them into feature data sets; S2: The feature dataset is randomly divided into a training set and a validation set. The training set is used to build and train the model, and the validation set is used to evaluate the model performance. S3: Use the decision tree algorithm to build the initial test result prediction model. The decision tree algorithm uses information gain as an indicator for selecting the optimal partition attribute. The calculation formula of information gain is: Gain(A)=Info(D)-Info A (D) Where Gain(A) represents the gain of attribute A, Info(D) represents the entropy of sample set D, and Info A (D) represents the entropy after partitioning D according to attribute A. The entropy calculation formula is: Info(D)=-Σp(x)log2p(x) p(x) is the proportion of the xth category in the sample set; S4: Use the training set to train the model, select the attribute with the largest information gain as the partition attribute of the current node, and recursively construct the subtree until the stopping condition is met; S5: Prune the constructed initial decision tree to avoid overfitting, and use the validation set to evaluate the performance of the pruned model. If the model performance meets the predetermined requirements, it is determined as the final test result prediction model.

4. The test project integrated management device according to claim 3, characterized in that: The intelligent recommendation module receives the test result prediction data output by the machine learning module, and selects matching test implementation suggestions from a preset recommendation strategy library based on the prediction results and current test conditions and content. The recommendation strategy library contains optimization suggestions and implementation strategies for a variety of test scenarios. These suggestions and strategies are classified and labeled according to the test type, test conditions and test objectives, and stored in a database for rapid retrieval and matching based on specific conditions.

5. The test project integrated management device according to claim 1, characterized in that: The disaster recovery module pre-sets a backup cycle and a backup time point, and at the backup time point, automatically exports the test project data in the storage module as a backup file, and stores the backup file in a designated backup storage location.

6. A test project integrated management device according to claim 5, characterized in that: The interface module defines a standardized data access interface, including a data reading interface and a data writing interface; for the data reading interface, a data reading request is received, and according to the parameters in the request, corresponding test project data is queried from the relational database; for the data writing interface, a write request containing test project data is received; the test project data in the write request is verified, and the verified test project data is written into the relational database; The connection pool module pre-creates and maintains a certain number of active database connections; when receiving a connection request, it directly allocates unused available connections; after the connection is used, it is marked as unused, and is regularly checked to close inactive connections.

7. The test project integrated management device according to claim 1, characterized in that: The user interface module uses WEB technology to create an interactive display interface, and provides relevant functional options for test project management on the display interface, including creating a new project, viewing a project list, editing project information, deleting a project, and searching for a project; When the user chooses to create a new project, a form is provided for the user to enter basic information of the project, including project name, project description, test person in charge, test location, test organization, test conditions, test content, test purpose and expected test results; When the user chooses to view the project list, the information of all test projects is retrieved from the database and displayed in a list on the user interface. Each project item displays the basic information and status of the project; When the user chooses to edit the project information, an editing form is provided for the user to modify the relevant information of the selected test project, and the modified information is updated in the database; When the user chooses to delete an item, all information of the selected test item is deleted from the database, and the item list is updated on the user interface; When the user selects a search item, a search box is provided for the user to enter keywords, and matching test items are retrieved from the database based on the keywords, and the results are displayed on the user interface.

8. The test project integrated management device according to claim 7, characterized in that: The display interface of the user interface module receives and displays the output results of the intelligent recommendation module.

9. The test project integrated management device according to claim 7, characterized in that: The project management module receives and processes user requests for adding, deleting, modifying and querying test projects through the user interface, interacts with the data storage layer to update or retrieve project information, and displays corresponding operation results on the display interface.

10. The test project integrated management device according to claim 1, characterized in that: The authority management module predefines user roles and assigns corresponding permissions. Each role corresponds to a set of specific operation permissions and data access scopes. When a user logs into the system, the user's identity is verified and the access permissions are determined based on the user's role. When a user performs an operation, the module checks whether the user's permissions allow the operation to be performed. If so, the operation request is released. Otherwise, the request is rejected and an error prompt is returned. The permission management module also provides permission modification function, allowing administrators to adjust the permission allocation of roles according to actual needs.

Images