Topics of Interest and Research
The Software Systems and Management faculty, in conjunction with colleagues in ECE, Silicon Valley companies and NASA's Ames Research Center, is engaged in a broad range of leading edge research projects across software systems and management. The Silicon Valley climate of innovation and entrepreneurship has a strong influence on our research, resulting in an emphasis on projects that can be applied to real world problems rather than purely focusing on foundational research. Silicon Valley faculty have been highly successful in obtaining research grants, publishing research results in leading conferences and archival journals, and in bringing their research into the classroom. Our location puts us in close proximity to many of the world's most innovative companies, and we have been able to draw on their top technical leaders to present their work in our weekly colloquia. In addition to direct software engineering and software management research, several other software intensive research projects engage faculty, MS and PhD students in building novel applications and systems involving sensors, networks, mobile and vehicular systems, using speech and machine learning technologies.
The Mobility Research Center was created to address the rapid growth of “smart” mobile devices as a primary means of access to the Internet, as well as a medium of communication. Research projects include methods and tools for mobile application development, and their growing interaction with social networks and cloud computing. Many of the research projects in the center use, study or enhance mobile applications and systems. Sensor-rich, mobile devices are a potentially rich source of context, enabling the creation of applications that adapt to the user’s need and situation. The Disaster Management Initiative was a direct outgrowth of the center’s activities.
The security research group at Carnegie Mellon Silicon Valley is part of Carnegie Mellon CyLab, one of the largest university-based cyber-security research and education centers in the U.S. Faculty and students work on several problems in many areas of security with a focus on the following core areas: Foundations of Security and Privacy, Web Security, Wireless Network and System Security
Innovation in Software Engineering Education
We research and apply modern education techniques to maximize transfer between the academic learning environment and the student’s professional practice. From the working professionals’ point of view, the most valuable graduate education aligns with and anticipates their evolving professional work. Toward that end, we employ a learning approach that functions as an enhanced, guided and practical version of the life-long practices that graduates will employ on the job. We have adopted a pedagogy that is based heavily on team-oriented projects, simulations, just-in-time coaching and tutorials, and industrial practicums, delivering our courses as Story-Centered Curricula.
PapersDeveloping Software Engineering Leaders at Carnegie Mellon Silicon Valley
Coaching Via Cognitive Apprenticeship
A Graduate Education in Software Management and the Software Business for Mid-Career Professionals
Exploration of Knowledge and Skills Transfer from a Formal Software Engineering Curriculum to a Capstone Practicum Project
Software Engineering Education at Carnegie Mellon University: One University; Programs Taught in Two Places
Contextualized Mobile Support for Learning by Doing in the Real World
We perform research in software verification. We investigate the use of abstraction and symbolic execution in the context of the open source Java PathFinder verification tool set, with applications to test input generation and error detection. Our main interest is in developing, extending and maintaining Symbolic PathFinder, a symbolic execution tool for Java bytecode. We are also working on using learning techniques for automating assume-guarantee compositional verification. Currently we are working on compositional techniques for probabilistic systems. Other research interests include parallelization of verification tasks and modeling and analysis for multiple statechart formalisms.
Faculty: Corina Pasareanu, Guillame Brat, Arnaud Venet
Open Source Software
We conduct research on evaluation, adoption, and use of open source software. Projects include analyzing business models for companies using free and open source software, quantitative analysis of open source projects, and . As thought-leaders in this discipline, we actively participate in the international open-source community.
Building a Business on Open Source Software
A Framework for Evaluating Managerial Styles in Open Source Projects
Evaluating Software Engineering Processes in Commercial and Community Open Source Projects
Teamwork and the Craft of Software Development
We examine effective practices to create high performing teams and we study the craft of software development and how to improve individual development practices while programmers are creating code. We are also involved with the evolution of agile development processes, widely used in Silicon Valley, especially among small entrepreneurial companies.
Papers: Towards Teaching The Craft of Software Development
Faculty: Todd Sedano
Explorations of standardized approaches to defining, collecting and interpreting measurements appropriate for effective management of different levels and phases of the software lifecycle and software business.
Software Component Reuse and Product Line Methods
Studying methods and technologies to improve the construction and use of components and frameworks, such as crowdsourcing, domain engineering, aspects and generators.
The Intelligent Systems Laboratory (ISL) develops state-of-the-art statistical learning and reasoning algorithms to build intelligent systems for real world problems. The group is developing intelligent systems for electrical power system management, speech translation, user behavior and mobile context information detection and modeling (e.g., indoor positioning and gesture detection). The research is centered on statistical learning, which draws on methods from statistics, machine learning, and data mining, is motivated by the massive amounts of data available nowadays. Many of the implementations draw on high-performance, GPU-based systems.