JIT-Smart: A Multi-task Learning Framework for Just-in-Time Defect Prediction and Localization

Just-in-time defect prediction (JIT-DP) is used to predict the defect-proneness of a commit and just-in-time defect localization (JIT-DL) is used to locate the exact buggy positions (defective lines) in a commit. Recently, various JIT-DP and JIT-DL techniques have been proposed, while most of them use a post-mortem way (e.g., code entropy, attention weight, LIME) to achieve the JIT-DL goal based on the prediction results in JIT-DP. These methods do not utilize the label information of the defective code lines during model building. In this paper, we propose a unified model JIT-Smart, which makes the training process of just-in-time defect prediction and localization tasks a mutually reinforcing multi-task learning process. Specifically, we design a novel defect localization network (DLN), which explicitly introduces the label information of defective code lines for supervised learning in JIT-DL with considering the class imbalance issue. To further investigate the accuracy and cost-effectiveness of JIT-Smart, we compare JIT-Smart with 7 state-of-the-art baselines under 5 commit-level and 5 line-level evaluation metrics in JIT-DP and JIT-DL. The results demonstrate that JIT-Smart is statistically better than all the state-of-the-art baselines in JIT-DP and JIT-DL. In JIT-DP, at the median value, JIT-Smart achieves F1-Score of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> mml:mrow mml:mn0.475</mml:mn> </mml:mrow> </mml:math> , AUC of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> mml:mrow mml:mn0.886</mml:mn> </mml:mrow> </mml:math> , Recall@20%Effort of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> mml:mrow mml:mn0.823</mml:mn> </mml:mrow> </mml:math> , Effort@20%Recall of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> mml:mrow mml:mn0.01</mml:mn> </mml:mrow> </mml:math> and Popt of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> mml:mrow mml:mn0.942</mml:mn> </mml:mrow> </mml:math> and improves the baselines by <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> mml:mrow mml:mn19</mml:mn> mml:mn.89%</mml:mn> mml:mi-</mml:mi> mml:mn702</mml:mn> mml:mn.74%</mml:mn> </mml:mrow> </mml:math> , <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> mml:mrow mml:mn1</mml:mn> mml:mn.23%</mml:mn> mml:mi-</mml:mi> mml:mn31</mml:mn> mml:mn.34%</mml:mn> </mml:mrow> </mml:math> , <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> mml:mrow mml:mn9</mml:mn> mml:mn.44%</mml:mn> mml:mi-</mml:mi> mml:mn33</mml:mn> mml:mn.16%</mml:mn> </mml:mrow> </mml:math> , <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> mml:mrow mml:mn21</mml:mn> mml:mn.6%</mml:mn> mml:mi-</mml:mi> mml:mn53</mml:mn> mml:mn.82%</mml:mn> </mml:mrow> </mml:math> and <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> mml:mrow mml:mn1</mml:mn> mml:mn.94%</mml:mn> mml:mi-</mml:mi> mml:mn34</mml:mn> mml:mn.89%</mml:mn> </mml:mrow> </mml:math> , respectively. In JIT-DL, at the median value, JIT-Smart achieves Top-5 Accuracy of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> mml:mrow mml:mn0.539</mml:mn> </mml:mrow> </mml:math> and Top-10 Accuracy of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> mml:mrow mml:mn0.396</mml:mn> </mml:mrow> </mml:math> , Recall@ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> mml:mrow mml:mn20</mml:mn> mml:mi%</mml:mi> mml:msub mml:mrow mml:mtextEffort</mml:mtext> </mml:mrow> mml:mrow mml:mil</mml:mi> mml:mii</mml:mi> mml:min</mml:mi> mml:mie</mml:mi> </mml:mrow> </mml:msub> </mml:mrow> </mml:math> of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> mml:mrow mml:mn0.726</mml:mn> </mml:mrow> </mml:math> , Effort@ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> mml:mrow mml:mn20</mml:mn> mml:mi%</mml:mi> mml:msub mml:mrow mml:mtextRecall</mml:mtext> </mml:mrow> mml:mrow mml:mil</mml:mi> mml:mii</mml:mi> mml:min</mml:mi> mml:mie</mml:mi> </mml:mrow> </mml:msub> </mml:mrow> </mml:math> of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> mml:mrow mml:mn0.087</mml:mn> </mml:mrow> </mml:math> and <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> mml:mrow mml:msub mml:mrow mml:mtextIFA</mml:mtext> </mml:mrow> mml:mrow mml:mil</mml:mi> mml:mii</mml:mi> mml:min</mml:mi> mml:mie</mml:mi> </mml:mrow> </mml:msub> </mml:mrow>