Revealing Relations between Open and Closed Answers in Questionnaires through Text Clustering Evaluation
2008 (English)In: Proceedings of the Sixth International Language Resources and Evaluation (LREC'08), 2008, 1-7 p.Conference paper (Refereed)
Open answers in questionnaires contain valuable information that is very time-consuming to analyze manually. We present a method forhypothesis generation from questionnaires based on text clustering. Text clustering is used interactively on the open answers, and the usercan explore the cluster contents. The exploration is guided by automatic evaluation of the clusters against a closed answer regarded as acategorization. This simplifies the process of selecting interesting clusters. The user formulates a hypothesis from the relation betweenthe cluster content and the closed answer categorization. We have applied our method on an open answer regarding occupation comparedto a closed answer on smoking habits. With no prior knowledge of smoking habits in different occupation groups we have generated thehypothesis that farmers smoke less than the average. The hypothesis is supported by several separate surveys. Closed answers are easyto analyze automatically but are restricted and may miss valuable aspects. Open answers, on the other hand, fully capture the dynamicsand diversity of possible outcomes. With our method the process of analyzing open answers becomes feasible.
Place, publisher, year, edition, pages
2008. 1-7 p.
Computer and Information Science
IdentifiersURN: urn:nbn:se:kth:diva-10123ISI: 000324028901137OAI: oai:DiVA.org:kth-10123DiVA: diva2:209206
LREC´08, 28-30 May, Marrakesh, Marocco
QC 201008062009-03-242009-03-242014-09-29Bibliographically approved