Raja Sooriamurthi

When Two Case Bases Are Better than One: Exploiting Multiple Case Bases

Much current CBR research focuses on how to compact, refine, and augment the contents of individual case bases, in order to distill needed information into a single concise and authoritative source. However, as deployed case-based reasoning systems become increasingly prevalent, opportunities will arise for supplementing local case bases on demand, by drawing on the case bases of other CBR systems addressing related tasks. Taking full advantage of these case bases will require multi-case-base reasoning: Reasoning not only about how to apply cases, but also about when and how to draw on particular case bases. This paper begins by considering tradeoffs of attempting to merge individual case bases into a single source, versus retaining them individually, and argues that retaining multiple case bases can benefit both performance and maintenance. However, achieving the benefits requires methods for case dispatching--deciding when to retrieve from external case bases, and which case bases to select--and for cross-case-base adaptation to revise suggested solutions from one context to apply in another. The paper presents initial experiments illustrating how these procedures may affect the benefits of using multiple case bases, and closes by delineating key research issues for multi-case-base reasoning.

Automatically Selecting Strategies for Multi-Case-Base Reasoning

Case-based reasoning (CBR) systems solve new problems by retrieving stored prior cases, and adapting their solutions to fit new circumstances. Traditionally, CBR systems draw their cases from a single local case-base tailored to their task. However, when a systems own set of cases is limited, it may be beneficial to supplement the local case-base with cases drawn from external casebases for related tasks. Effective use of external case-bases requires strategies for multi-case-base reasoning (MCBR): (1) for deciding when to dispatch problems to an external case-base, and (2) for performing cross-case-base adaptation to compensate for differences in the tasks and environments that each case-base reflects. This paper presents methods for automatically tuning MCBR systems by selecting effective dispatching criteria and cross-case-base adaptation strategies. The methods require no advance knowledge of the task and domain: they perform tests on an initial set of problems and use the results to select strategies reflecting the characteristics of the local and external case-bases. We present experimental illustrations of the performance of the tuning methods for a numerical prediction task, and demonstrate that a small sample set can be sufficient to make high-quality choices of dispatching and cross-case-base adaptation strategies.

Puzzle-Based Learning for Engineering and Computer Science

To attract, motivate, and retain students and increase their mathematical awareness and problem-solving skills, universities are introducing courses or seminars that explore puzzle-based learning. We introduce and define this learning approach with a sample syllabus and course material, describe course variations, and highlight early student feedback.

CASE DISPATCHING VERSUS CASE-BASE MERGING: WHEN MCBR MATTERS

Multi-case-base reasoning (MCBR) extends case-based reasoning to draw on multiple case bases that may address somewhat different tasks. In MCBR, an agent selectively supplements its own case-base as needed, by dispatching problems to external case-bases and using cross-case-base adaptation to adjust their solutions for inter-case-base differences. MCBR has been advocated as a means to facilitate handling large case-bases when storage is limited, or to enable use of distributed case sources. However, this raises an important question: When storage is not an issue, and the cases from all external case sources could be merged into a single case-base, is there any reason for MCBR? This article answers that question with an experimental assessment of how MCBR affects the quality of solutions generated. It demonstrates that for a given local case-base and an external case-base for a task environment that is similar to, but different from, the local task environment, MCBR can improve accuracy compared to merging the case-bases into a single case-base. This improvement holds even if the cross-case-base adaptation method used by MCBR is also applied to the external cases before merging. The article hypothesizes an explanation of this behavior in terms of the ability of MCBR to exploit the tradeoffs between similarity of problems and similarity of solution contexts. It provides experimental evidence to support this hypothesis, and also demonstrates that MCBR is a useful framework for guiding case-base maintenance by selecting cases to add to a case-base.

Introducing abstraction and decomposition to novice programmers

This paper discusses a learning exercise we use in our beginning programming classes to introduce students to the concepts of abstraction and decomposition. The assignment is to write a perpetual calendar generation program: given a month and a year the program will display the correct monthly calendar. The learning goals of the exercise include how to decompose a large problem into smaller pieces and how to specify what each piece needs to do. This exercise helps students learn the process of incremental and iterative development. More than the actual solution, the value of this exercise is in the several themes of software development that are discussed during its development. We have successfully used this assignment for several years in a variety of CS1/CS2 programming environments (Pascal, C, Java and .net) and also as a Java servlet based web application exercise. Over this period, the case-study has received very favorable feedback from students as to its interestingness and pedagogical value.

Guide to Teaching Puzzle-based Learning

This book provides insights drawn from the authors extensive experience in teaching Puzzle-based Learning. Practical advice is provided for teachers and lecturers evaluating a range of different formats for varying class sizes. Features: suggests numerous entertaining puzzles designed to motivate students to think about framing and solving unstructured problems; discusses models for student engagement, setting up puzzle clubs, hosting a puzzle competition, and warm-up activities; presents an overview of effective teaching approaches used in Puzzle-based Learning, covering a variety of class activities, assignment settings and assessment strategies; examines the issues involved in framing a problem and reviews a range of problem-solving strategies; contains tips for teachers and notes on common student pitfalls throughout the text; provides a collection of puzzle sets for use during a Puzzle-based Learning event, including puzzles that require probabilistic reasoning, and logic and geometry puzzles.

Puzzle-based learning: introducing creative thinking and problem solving for computer science and engineering (abstract only)

Puzzle-based learning (PBL) is an emerging model of teaching critical thinking and problem solving. Todays market place needs skilled graduates capable of solving real problems of innovation in a changing environment. While solving puzzles is innately fun, companies such as Google and Yahoo also use puzzles to assess the creative problem solving skills of potential employees. In this interactive workshop we will examine a range of puzzles, games, and general problem solving strategies. Participants will emerge with the needed pedagogical foundation to offer a full course on PBL or to include it as part of another course. Currently 40+ institutions around the world are offering courses based on PBL. More details are available at www.PuzzleBasedLearning.edu.au. Laptop optional.

Information Systems Application Development Courses: A Carnegie Mellon University Experience in Global Pedagogy

In this experience report, we describe recent initiatives in global undergraduate Information Systems education at Carnegie Mellon University. The entire systems development core curriculum is now offered at CMU campuses in Pittsburgh and Doha, Qatar. Courses are co-designed and delivered by faculty in both locations with an eye toward consistency of content and assessment, but also with content tuned for local sections. The collaboration and lessons learned among collaborating faculty in two example courses is described.

Data jam: introducing high school students to data science

At the present, there is a significant lack of programs or resources at the high school level to prepare students for a data driven future. Data Jam is a high school outreach program, that introduces students to data science. The program is organized by members of academia (Carnegie Mellon University, University of Pittsburgh) and industry and research (IBM, Teradata, Pittsburgh Computing Center). Over a period of four months (Oct-Feb of the school year), teachers and students explore the concepts of data science and big data via workshops, exercises, a field trip, and a team project. Data Jam is currently in its fifth year. Participation has grown from an initial pool of seven teams to twenty-five teams last year. Based on teacher and student feedback, we are pleased with the programs success. This poster discusses the goals and structure of Data Jam, its execution, participant feedback, and lessons learned.

Diagnosing NFS errors: Preliminary Findings from a Syslog Analysis of Bridges

Bridges is the current main system at the Pittsburgh Supercomputing Center. Given the complexity of the system and the volume of its use, it is a very good environment for exploring the potential of machine learning techniques in studying sub-optimal performance. This short report discusses preliminary and ongoing work of a new graduate student exploring this novel realm. Our initial focus has been on learning to predict the occurrence of NFS time out errors from preceding syslog messages.