.png){kind=icon}
返回
Test case classification via few-shot learning
.png){kind=icon}
PRE
AI
DOI：10.1016/j.infsof.2023.107228.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Context: Crowdsourced testing can reduce testing costs and improve testing efficiency. However, crowdsourced testing generates massive test cases, requiring testers to select high-quality test cases for execution. Conse-quently, crowdsourced test cases require much effort to perform labeling due to the costly manual labor and domain knowledge.Objective: Existing methods usually fail to consider the crowdsourced testing scenario's inadequate and imbalanced data issues. We aim to effectively and efficiently classify many crowdsourced test cases for developers to alleviate manual efforts. Method: In this paper, we propose a test case classification approach based on few-shot learning and test case augmentation to address the limitations mentioned above. The proposed approach generates new test cases by the large pre-trained masked language model and extracts embedding representation by training word embedding models. Then a Bidirectional Long Short-Term Memory (BiLSTM)-based classifier is designed to perform test case classification by extracting the in-depth features. Besides, we also apply the attention mechanism to assign high weights to words that represent the test case category by lexicon matching. Results: To verify the effectiveness of the classification framework, we select 1659 test cases from three crowdsourced testing projects to conduct in-usability evaluation experiments. The experimental results show that the proposed approach has a higher accuracy and precision rate than existing classification methods.Conclusion: It can be concluded that (1) the proposed approach is an effective test case classification technique for crowdsourced testing; (2) the proposed approach is practical to help developers select high-quality test cases quickly and effectively.
Keyword:
Test case classification
Few-shot learning
Attention mechanism
期刊
IF:
4.3
论文数:
3.8K
被引数:
7.7K
机构
引用论文
Using unsupervised clustering approach to train the Support Vector Machine for text classification使用无监督聚类方法训练支持向量机进行文本分类.png)
NEUROCOMPUTING
NEUROCOMPUTINGIF6.5
PREAI
分享收藏分享收藏
Substrate-dependent shift of optimum pH in porcine pancreatic .alpha.-amylase-catalyzed reactionsBiochemistry
BiochemistryIF0
PREAI
分享收藏分享收藏
查看更多内容