Logo image
Back
Learning feature weights from positive cases
Journal article   Open access   Peer reviewed

Learning feature weights from positive cases

Sidath Deepal Gunawardena, Rosina O. Weber and Julia Stoyanovich
Case-Based Reasoning Research and Development, pp 134-148
13 Jun 2013
DOI: https://doi.org/10.1007/978-3-642-39056-2_10
pdf
Learning Feature Weights from Positive Cases302.18 kBDownloadView
Accepted (AM) Open Access

Abstract

Case alignment Case cohesion Density clustering Multidisciplinary collaboration Recommender systems (Information filtering) Single class learning Subspace clustering
The availability of new data sources presents both opportunities and challenges for the use of Case-based Reasoning to solve novel problems. In this paper, we describe the research challenges we faced when trying to reuse expe-riences of successful academic collaborations available online in descriptions of funded grant proposals. The goal is to recommend the characteristics of two collaborators to complement an academic seeking a multidisciplinary team; the three form a collaboration that resembles a configuration that has been success-ful in securing funding. While seeking a suitable measure for computing simi-larity between cases, we were confronted with two challenges: a problem con-text with insufficient domain knowledge and data that consists exclusively of successful collaborations, that is, it contains only positive instances. We present our strategy to overcome these challenges, which is a clustering-based approach to learn feature weights. Our approach identifies poorly aligned cases, i.e., ones that violate the assumption that similar problems have similar solutions. We use the poorly aligned cases as negatives in a feedback algorithm to learn feature weights. The result of this work is an integration of methods that makes CBR useful to yet another context and in conditions it has not been used before.

Metrics

79 File views/ downloads
17 Record Views
1 citations in Scopus

Details

Logo image