Magazine article AI Magazine

The Computational Analogy Workshop at ICCBR-16

Magazine article AI Magazine

The Computational Analogy Workshop at ICCBR-16

Article excerpt

The Computational Analogy Workshop was held October 31, 2016, as part of the ICCBR 2016 conference in Atlanta, Georgia. This report summarizes the workshop.

**********

Computational analogy and case-based reasoning (CBR) are closely related research areas. Both employ prior cases to reason in complex situations with incomplete information. Analogy research often focuses on modeling human cognitive processes, the structural alignment between a base or source and target, and adaptation or abstraction of the analogical source content. While CBR research also deals with alignment and adaptation, the field tends to focus more on retrieval, case-base maintenance, and pragmatic solutions to real-world problems. However, despite the overlap in their research goals and approaches, cross communication and collaboration between these areas have been progressively diminishing.

This workshop brought together researchers who use a variety of analogical reasoning approaches together with researchers in CBR. The workshop included 11 papers representing a range of computational analogy research methods and topics, including corpus mining for analogies, analogical proportions, analogy for natural language processing, design by analogy, analogy tutors, and instruction for analogical agents. Two invited talks set the stage for discussing the intersection of analogy and CBR. Ashok Goel (Georgia Institute of Technology) discussed the role of computational analogy in the domains of biologically inspired design, visual thinking, and cognitive robotics. Ken Forbus (Northwestern University) discussed applications of analogical reasoning in qualitative and commonsense reasoning and cognitive architectures.

Three major themes ran through the rest of the workshop: formal analogy, adaptation and abstraction, and design by analogy. Four papers dealt with formal (symbol-string) analogies. These papers examined how to predict the morphology of rare words, predict unseen words in text translation, solve symbol-string analogical equations, and translate natural language input to bash commands.

The design-by-analogy session discussed the use and development of systems that identify similar designs from a database. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.