We present TEGCER, an automated feedback tool for novice programmers. TEGCER uses supervised classification to match compilation errors in new code submissions with relevant pre-existing errors, submitted by other students before. The dense neural network used to perform this classification task is trained on 15,000+ error-repair code examples. The proposed model yields a test set classification Pred@3 accuracy of 97.7% across 212 error category labels. Using this model as its base, TEGCER presents students with the closest relevant examples of solutions for their specific error on demand. A large scale (N>230) usability study shows that students who use TEGCER are able to resolve errors more than 25% faster on average than students being assisted by human tutors.
Tue 12 Nov
16:00 - 17:40: Papers - Testing and Visualization at Cortez 1 Chair(s): Amin AlipourUniversity of Houston | ||||||||||||||||||||||||||||||||||||||||||
16:00 - 16:20 Talk | History-Guided Configuration Diversification for Compiler Test-Program GenerationACM SIGSOFT Distinguished Paper Award Junjie ChenTianjin University, Guancheng WangPeking University, Dan HaoPeking University, Yingfei XiongPeking University, Hongyu ZhangThe University of Newcastle, Lu ZhangPeking University | |||||||||||||||||||||||||||||||||||||||||
16:20 - 16:40 Talk | Data-Driven Compiler Testing and Debugging Junjie ChenTianjin University | |||||||||||||||||||||||||||||||||||||||||
16:40 - 17:00 Talk | Targeted Example Generation for Compilation Errors Umair Z. AhmedNational University of Singapore, Renuka SindhgattaQueensland University of Technology, Australia, Nisheeth SrivastavaIndian Institute of Technology, Kanpur, Amey KarkareIIT Kanpur Link to publication Pre-print | |||||||||||||||||||||||||||||||||||||||||
17:00 - 17:20 Talk | Lightweight Assessment of Test-Case Effectiveness using Source-Code-Quality Indicators Giovanni GranoUniversity of Zurich, Fabio PalombaDepartment of Informatics, University of Zurich, Harald GallUniversity of Zurich Link to publication Pre-print | |||||||||||||||||||||||||||||||||||||||||
17:20 - 17:30 Demonstration | Visual Analytics for Concurrent Java Executions Cyrille ArthoKTH Royal Institute of Technology, Sweden, Monali PandeKTH Royal Institute of Technology, Qiyi TangUniversity of Oxford | |||||||||||||||||||||||||||||||||||||||||
17:30 - 17:40 Demonstration | NeuralVis: Visualizing and Interpreting Deep Learning Models Xufan ZhangState Key Laboratory for Novel Software Technology Nanjing University, Nanjing, China, Ziyue YinState Key Laboratory for Novel Software Technology Nanjing University, Nanjing, China, Yang FengUniversity of California, Irvine, Qingkai ShiHong Kong University of Science and Technology, Jia LiuState Key Laboratory for Novel Software Technology Nanjing University, Nanjing, China, Zhenyu ChenNanjing University |