Chatbot development in data representation for diabetes education

• Main Author: Abbas Saliimi, Lokman
• Format: Thesis
• Language: English
• Published: 2011
• Subjects: QA76 Computer software
• Online Access: http://umpir.ump.edu.my/id/eprint/17818/1/Chatbot%20development%20in%20data%20representation%20for%20diabetes%20education.pdf; http://umpir.ump.edu.my/id/eprint/17818/
• Institution: Universiti Malaysia Pahang

This thesis deals with an investigation towards developing an application package named E-CARE to function as Computer Aided Education (CAE) in the domain of Diabetes. E-CARE contains two respective applications, 1) E-CARE multimedia content, and 2) SQL-based chatbot named ViDi (acronyms for Virtual Diabetes physician). Alongside the development of ViDi chatbot, several new approaches (algorithms and techniques) had been proposed. Those are 1) Vpath, 2) Sequence Words Deleted (SWD), 3) Extension and Prerequisite, 4) One-Match and All-Match Categories (OMAMC), 5) Synonyms and Root-words and lastly 6) General Words Percentage (GWP). Vpath and Extension and Prerequisite are techniques that enable relations between responses (previous and next responses). SWD and OMAMC are proposed to enhance the process of keywords/pattern matching for chatbot. Synonyms and Root-words are proposed as an additional component for chatbot that deals with Malay Language (Bahasa Malaysia), and GWP is proposed to become a supplementary component for the process of selecting final response in case where matching process is producing multiple responses. Measurement has being done by comparing the results/outcomes of proposed approaches towards previously developed chatbots (focus on A.L.I.C.E.’ as mostly referred chatbot and VPbot as SQL-based chatbot in medical domain). Significance of results is represent by several comparison tables on issues related to each proposed approaches’ purpose. As for overall research investigation, results for each area in which E-CARE has contributed seems adequate. Multimedia content component help in the development of precise CAE in regard to specific requirements gathered from previous prerequisite study, while ViDi component contributed to several areas within the processes architecture surrounding chatbot technology.