Title: Experimental Program Analysis Speaker: Joseph Ruthruff, Sandia National Laboratories Date/Time: September 23, 2008, 10:00 am – 11:00 am Location: CSRI Building, Room 90 (Sandia NM) Brief Abstract: Program analysis techniques are used by software engineers to deduce and infer targeted characteristics of software systems for tasks such as testing, debugging, impact analysis, maintenance, and program comprehension. Experimental program analysis is a new form of program analysis that incorporates characteristics of formal experimentation. This talk will provide an overview of the experimental program analysis paradigm. Building on an intuitive assessment of common program analysis activities as well as principles and methodologies underlying the use of experimentation in other fields, I provide definitions of experimental program analysis, illustrate them by example, and explore the applicability of experimental program analysis in various software engineering problems. In particular, I will discuss in detail the use of experimental program analysis in a screening methodology to build logistic regression models for making predictions about static analysis warnings in work conducted at Google. The contributions of this and other studies derived from this work suggests that this paradigm offers a promising new direction for program analysis research.CSRI POC: Patty Hough, 925-294-1518 and Brian Adams, 505-284-8845 |