A Computational Knowledge Elicitation and Sharing System for mental health case management of the social service industry

  • Authors:
  • W. M. Wang;C. F. Cheung

  • Affiliations:
  • Knowledge Management and Innovation Research Centre, Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong;Knowledge Management and Innovation Research Centre, Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong

  • Venue:
  • Computers in Industry
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

Narrative data provide rich information and knowledge to the workers. However, existing systems mainly served as a workflow system, a reporting system, or a database system for storing this kind of information. The massive amount of unstructured narrative data makes it extremely difficult to be shared and reused. Actual knowledge sharing and reuse among the workers is still limited. This paper presents a Computational Knowledge Elicitation and Sharing System which attempts to elicit knowledge from individuals as well as a team and converts it into a structured format and shared among the team. The proposed system accomplishes several current technologies in knowledge-based system, artificial intelligence and natural language processing, which converts the narrative knowledge of knowledge workers into a concept mapping representation. With a sufficient number of narratives, patterns are revealed and an aggregate concept map for all participating members is produced. It converts the unstructured text into a more structured format which helps to summarize and share the knowledge that can be taken in handling different case management issues. Such integration is considered to be novel. A prototype system has been implemented based on the method successfully in the mental healthcare of a social service organization for handling their case management issues. An experiment has been carried out for measuring the accuracy for converting the unstructured data into the structured format. The theoretical results are found to agree well with the experimental results.