Yu Zhao

Dr. Yu Zhao

Jiangsu Normal University
orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 October 2024
Verified
Conference Paper
Open access
YZYu Zhao
Yu Zhao
LGLina Gong
Lina Gong
ZHZhiqiu Huang
Zhiqiu Huang
YWYongwei Wang
Yongwei Wang
Mingqiang WeiMingqiang Wei
Mingqiang Wei
FWFei Wu
Fei Wu
Vulnerability detection is garnering increasing attention in software engineering, since code vulnerabilities possibly pose significant security. Recently, reusing various code pre-trained models (e.g., CodeBERT, CodeT5, and CodeGen) has become common for code embedding without providing reasonable justifications in vulnerability detection. The premise for casually utilizing pre-trained models (PTMs) is that the code embeddings generated by different PTMs would generate a similar impact on the performance. Is that TRUE? To answer this important question, we systematically investigate the effects of code embedding generated by ten different code PTMs on the performance of vulnerability detection, and get the answer, i.e., that is NOT true. We observe that code embedding generated by various code PTMs can indeed influence the performance and selecting an embedding technique based on parameter scales and embedding dimension is not reliable. Our findings highlight the necessity of quantifying and evaluating the characteristics of code embedding generated by various code PTMs to understand the effects. To achieve this goal, we analyze the numerical representation and data distribution of code embedding generated by different PTMs to evaluate differences and characteristics. Based on these insights, we propose Coding-PTMs, a recommendation framework to assist engineers in selecting optimal code PTMs for their specific vulnerability detection tasks. Specifically, we define thirteen code embedding metrics across three dimensions (i.e., statistics, norm, and distribution) for constructing a specialized code PTM recommendation dataset. We then employ a Random Forest classifier to train a recommendation model and identify the optimal code PTMs from the candidate model zoo. We encourage engineers to use our Coding-PTMs to evaluate the characteristics of code embeddings generated by candidate code PTMs on the performance and recommend optimal code PTMs for code embedding in their vulnerability detection tasks.

17 February 2022 in Symmetry
Verified
Journal Article
Open access
PDF file
YZYu Zhao
Yu Zhao
YZYi Zhu
Yi Zhu
QYQiao YU
Qiao YU
XCXiaoying Chen
Xiaoying Chen
Software testing is the main method for finding software defects at present, and symmetric testing and other methods have been widely used, but these testing methods will cause a lot of waste of resources. Software defect prediction methods can reasonably allocate testing resources by predicting the defect tendency of software modules. Cross-project defect prediction methods have huge advantages when faced with missing datasets. However, most cross-project defect prediction methods are designed based on the settings of a single source project and a single target project. As the number of public datasets continues to grow, the number of source projects and defect information is increasing. Therefore, in the case of multi-source projects, this paper explores the problems existing when using multi-source projects for defect prediction. There are two problems. First, in practice, it is not possible to know in advance which source project is used to build the model to obtain the best prediction performance. Second, if an inappropriate source project is used in the experiment to build the model, it can lead to lower performance issues. According to the problems found in the experiment, the paper proposed a multi-source-based cross-project defect prediction method MSCPDP. Experimental results on the AEEEM dataset and PROMISE dataset show that the proposed MSCPDP method effectively solves the above two problems and outperforms most of the current state-of-art cross-project defect prediction methods on F1 and AUC. Compared with the six cross-project defect prediction methods, the F1 median is improved by 3.51%, 3.92%, 36.06%, 0.49%, 17.05%, and 9.49%, and the ACU median is improved by −3.42%, 8.78%, 0.96%, −2.21%, −7.94%, and 5.13%.

1 January 2023 in IET Software
Verified
Journal Article
Open access
PDF file
YZYu Zhao
Yu Zhao
ZHZhiqiu Huang
Zhiqiu Huang
LGLina Gong
Lina Gong
YZYi Zhu
Yi Zhu
QYQiao YU
Qiao YU
YGYuxiang Gao
Yuxiang Gao
The performance of software defect prediction (SDP) models determines the priority of test resource allocation. Researchers also use interpretability techniques to gain empirical knowledge about software quality from SDP models. However, SDP methods designed in the past research rarely consider the impact of data transformation methods, simple but commonly used preprocessing techniques, on the performance and interpretability of SDP models. Therefore, in this paper, we investigate the impact of three data transformation methods (Log, Minmax, and Z ‐score) on the performance and interpretability of SDP models. Through empirical research on (i) six classification techniques (random forest, decision tree, logistic regression, Naive Bayes, K‐nearest neighbors, and multilayer perceptron), (ii) six performance evaluation indicators (Accuracy, Precision, Recall, F1, MCC, and AUC), (iii) two interpretable methods (permutation and SHAP), (iv) two feature importance measures (Top‐ k feature rank overlap and difference), and (v) three datasets (Promise, Relink, and AEEEM), our results show that the data transformation methods can significantly improve the performance of the SDP models and greatly affect the variation of the most important features. Specifically, the impact of data transformation methods on the performance and interpretability of SDP models depends on the classification techniques and evaluation indicators. We observe that log transformation improves NB model performance by 7%–61% on the other five indicators with a 5% drop in Precision. Minmax and Z ‐score transformation improves NB model performance by 2%–9% across all indicators. However, all three transformation methods lead to substantial changes in the Top‐5 important feature ranks, with differences exceeding 2 in 40%–80% of cases (detailed results available in the main content). Based on our findings, we recommend that (1) considering the impact of data transformation methods on model performance and interpretability when designing SDP approaches as transformations can improve model accuracy, and potentially obscure important features, which lead to challenges in interpretation, (2) conducting comparative experiments with and without the transformations to validate the effectiveness of proposed methods which are designed to improve the prediction performance, and (3) tracking changes in the most important features before and after applying data transformation methods to ensure precise and traceable interpretability conclusions to gain insights. Our study reminds researchers and practitioners of the need for comprehensive considerations even when using other similar simple data processing methods.

20 August 2021 in Future Internet
Verified
Journal Article
Open access
PDF file
YZYu Zhao
Yu Zhao
YZYi Zhu
Yi Zhu
QYQiao YU
Qiao YU
XCXiaoying Chen
Xiaoying Chen
Traditional research methods in software defect prediction use part of the data in the same project to train the defect prediction model and predict the defect label of the remaining part of the data. However, in the practical realm of software development, the software project that needs to be predicted is generally a brand new software project, and there is not enough labeled data to build a defect prediction model; therefore, traditional methods are no longer applicable. Cross-project defect prediction uses the labeled data of the same type of project similar to the target project to build the defect prediction model, so as to solve the problem of data loss in traditional methods. However, the difference in data distribution between the same type of project and the target project reduces the performance of defect prediction. To solve this problem, this paper proposes a cross-project defect prediction method based on manifold feature transformation. This method transforms the original feature space of the project into a manifold space, then reduces the difference in data distribution of the transformed source project and the transformed target project in the manifold space, and finally uses the transformed source project to train a naive Bayes prediction model with better performance. A comparative experiment was carried out using the Relink dataset and the AEEEM dataset. The experimental results show that compared with the benchmark method and several cross-project defect prediction methods, the proposed method effectively reduces the difference in data distribution between the source project and the target project, and obtains a higher F1 value, which is an indicator commonly used to measure the performance of the two-class model.

8 July 2021 in IEEE Access
Verified
Journal Article
Open access
XCXiaoying Chen
Xiaoying Chen
YZYi Zhu
Yi Zhu
YZYu Zhao
Yu Zhao
JWJinyong Wang
Jinyong Wang
AAAnarbekov Altynbek
Anarbekov Altynbek
The wide application of Cyber Physical System (CPS) makes the security of CPS more and more concerned. As the key factors affecting the safety of CPS, space and time have also become the current research hotspot. The space and time safety of CPS requires that CPS arrives at the specified place at the specified time, time and space should meet the safety requirements of the CPS in the current CPS environment. We call the behavior space-time compositions. In order to solve the problem that CPS lacks the method of modeling and verification of space-time compositions, a hybrid Architecture Analysis & Design Language (AADL) modeling and model transformation method for CPS space-time compositions verification is proposed. Firstly, space-time description capability is extended in the AADL behavior annex and Hybrid AADL (HAADL) is proposed. Secondly, differential equations and space-time compositions vector are introduced in Process Algebra to propose Hybrid Space-Time Communication Sequential Processes (HS-TCSP). Furthermore, the Hybrid AADL is transformed to HS-TCSP. Finally, an example of an aircraft collision avoidance system is used to verify the effectiveness of the method.

11 February 2025 in Expert Systems with Applications
Verified
Journal Article
Closed access
YZYu Zhao
Yu Zhao
LGLina Gong
Lina Gong
YYYaoshen Yu
Yaoshen Yu
ZHZhiqiu Huang
Zhiqiu Huang
Mingqiang WeiMingqiang Wei
Mingqiang Wei