Value Trace Problems for Graph Theory Algorithms in Java Programming Learning Assistant System
Nobuo Funabiki, Khin Khin Zaw, Minoru Kuribayashi, Wen-Chung Kao · International Journal of Information and Education Technology · 2016
Code reading is very important in programming educations for students.Through reading and analyzing high quality codes, they can study how to write a proper code and modify it with given specification.To assist with the studies of Java code reading, we proposed the value trace problem in Web-based Java Programming Learning Assistant System (JPLAS).JPLAS has been developed to provide self-learning environments to students by our group.This value trace problem asks students to trace the actual values of important variables in a Java code implementing a fundamental data structure or an algorithm.In this paper, we study value trace problems for graph theory algorithms.First, using the Dijkstra algorithm, we analyze the requirements and points in a value trace problem for this representative graph theory algorithm.Then, we generate problems for the two graph theory algorithms to examine their problem size and the effectiveness in Java programming studies.Our evaluation results show that value trace problems for graph theory algorithms are viable learning tools for algorithm understanding and code reading whereas additional tools are necessary for code writing.