Problem Event Extraction to Develop Causal Loop Representation from Texts

This research aims to extract consequent problem events as a cause-effect concept pair series, from teen-drug addiction web-boards. The extracted consequent problem events benefit for a problem analysis in a solving system through a Causal Loop representation. The research has three problems; how to determine a causative/effect event concept based on a verb phrase expression with an overlap problem between causative-verb concept set and effect-verb concept set, how to determine cause-effect concept pair series from several verb phrases, and how to develop a Causal Loop representation from the extracted cause-effect concept pair series. Therefore, we apply an event rate to solve the overlap problem. We then propose using N-WordCo to determine the cause-effect concept pair series and also use a similarity score to develop the Causal Loop representation. The research results provide a high precision of the problem event extraction from the documents.