Commonsense-based topic modeling
Open Access
- 11 August 2013
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1-8
Abstract
Topic modeling is a technique used for discovering the abstract 'topics' that occur in a collection of documents, which is useful for tasks such as text auto-categorization and opinion mining. In this paper, a commonsense knowledge based algorithm for document topic modeling is presented. In contrast to probabilistic models, the proposed approach does not involve training of any kind and does not depend on word co-occurrence or particular word distributions, making the algorithm effective on texts of any length and composition. 'Semantic atoms' are used to generate feature vectors for document concepts. These features are then clustered using group average agglomerative clustering, providing much improved performance over existing algorithms.Keywords
This publication has 20 references indexed in Scilit:
- Intention awareness: improving upon situation awareness in human-centric environmentsHuman-centric Computing and Information Sciences, 2013
- Common Sense Knowledge for Handwritten Chinese Text RecognitionCognitive Computation, 2012
- Probabilistic topic modelsCommunications of the ACM, 2012
- Sentic Web: A New Paradigm for Managing Social Media Affective InformationCognitive Computation, 2011
- Part-of-Speech Tagging from 97% to 100%: Is It Time for Some Linguistics?2011
- DBpedia: A Nucleus for a Web of Open Data2007
- A Survey of Clustering Data Mining Techniques2006
- ConceptNet — A Practical Commonsense Reasoning Tool-KitBT Technology Journal, 2004
- A model of textual affect sensing using real-world knowledge2003
- Another stemmerACM SIGIR Forum, 1990