@misc{9102,
  author = {Sunil Nair},
  title = {Learning Program Models by Observing Their Behavior},
  abstract = {Testing large complex system is expensive and highly resource consuming. Studies estimate that about 30 \% to 50 \% of the resources in an average software project are typically allotted to testing alone. To add more fuel to the fire, many of these bugs remain undetected due to poor testing strategies that lead to disastrous consequences. In spite of such huge resource allocation, the unreliability is due to the fact that Software testing is typically a Manual process. Documentation is an effective way to identify the behaviour of the system, but it is hardly up to date or written by people who do not understand the system. Without a guide to effectively test the system, the human tester is forced to use intuition to select test cases that exercises the full system. The project will aim to develop a technique to automate testing of unfamiliar systems. There are already tools that can do this, but these are not particularly accurate, and can often end up producing rules that are misleading. The technique introduced - ILUSTRATOR (Inductive Learning USing Tests by RAndom generaTOR) is based on novel machine learning techniques to infer models from unfamiliar systems. Inference by induction is the key concept dealt in this project. The outcome of the project will be a rigorous, automated test case generator for unfamiliar black-box systems, supported by inductive testing and novel machine learning techniques that can be applied to a larger and realistic software system.},
  year = {2011},
  publisher = {University Of Leicester},
}