Concept Identification Using Co-Occurrence Graph

Abstract

In a community setting, Utilitarian Knowledge or “Knowledge that works” are routinely diffused through social media interactions. The aggregation of this knowledge is a divergent process, where common knowledge gets segregated into several local worlds of utilitarian knowledge. To capture and represent this knowledge, several data models have been proposed. One of the model organizes concepts (atomic elements) in a hierarchy namely concept hierarchy (“is-a”) in which concepts are added manually at the most appropriate level inside the hierarchy. To minimize manual intervention in entity resolution, this article proposes entity resolution based on co-occurrence graph and continuous learning, thereby eliminating the bottleneck of manual concept entry. While traditional Supervised Learning methods require sufficient training data beforehand which is not available in a community setting at start, Continuous Learning method could be useful which can acquire new behaviours and can evolve as the community data evolves.