Evaluation approach used in Randoop, ARTOO, and KORAT Papers

This week we are discussing papers on RANDOOP, ARTOO, and KORAT. I would like you to compare the papers in terms of the strategy used to evaluate the proposed approaches. In particular, compare how the experiments were set up, what kinds of benchmarks were used, what kinds of faults were used (seeded vs real), metrics were used, how many runs were used, and whether the authors compared with other approaches. Feel free to include other comparison criteria as well.

Part A (6 points)

1. Post your thoughts on the three papers that you can find in Canvas.

Part B (4 points)

1. Respond to at least one other posting and discuss the ideas. (3 points)
2. Prepare (as a group) a list containing all the ideas proposed in your group and post it to the main discussion board. (1 point)

Grading Policy

You must actively participate in the discussion to get a grade.