A consumer-centric design approach to develop comprehensive knowledge-based systems for keyword discovery

The Internet is the largest information repository. Most information retrieval systems are based on the premise that users know the keywords for searching subjects. However, identifying the appropriate keywords has become increasingly problematic for most users. This study proposes a comprehensive knowledge-based system that assists users in discovering keywords by comprehending human feelings toward a subject. A consumer-centric design approach is employed for ontology development. The following three tasks are performed: gathering and grouping consumer psychological cognitions; translating these semantics into design elements of concept models; and, annotating the features of each concept. The comprehensive knowledge-based system allows users to input sensory descriptions as a query to infer possible resemblances of guesses to the real subject. A case study regarding the identification of dog breeds is presented to demonstrate how users obtains a dog breed name by describing what they see and feel. Empirical results demonstrate that the new design help users with little domain knowledge obtain appropriate keywords.