Improving Responsiveness Conversation of Thai Chatbot through Sentiment Analysis Classification Techniques
Sumitra Nuanmeesri1, Lap Poomhiran2

1Sumitra Nuameesri*, Assistant Professor, Department of Information Technology Suan Sunandha Rajabhat University, Thailand.
2Lap Poomhiran, Ph.D. Department of Information Technology, King Mongkut’s University of Technology North Bangkok, Thailand.
Manuscript received on November 27, 2019. | Revised Manuscript received on December 15, 2019. | Manuscript published on December 30, 2019. | PP: 3733-3737  | Volume-9 Issue-2, December, 2019. | Retrieval Number:  C4676029320/2019©BEIESP | DOI: 10.35940/ijeat.C4676.129219
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (

Abstract: Nowadays, internet and social media are play and important role for the business and marketing. Especially, the social media marketing drives the businesses with fierce competition. if there is communication between a large number of customers, it is necessary to have the staff to coordinate thoroughly Resulting in higher expenses as well. Chatbot can be solve this problem by action like a human to deliver a suitable message for their customers. This paper proposes the techniques for analyzing the sentiments that coexist with chat messages or the conversations. Naïve Bayes, K-Nearest Neighbor, and Support Vector Machine techniques were used to classify the sentiments based on Cross-Industry Standard Process for Data Mining. As a result, the highest accuracy is produced by Support Vector Machine with value at 94.60% for improving the chatbot able to communicate effectively with sticker messages.
Keywords: Chatbot, Classification, Conversation, Sentiment analysis, Social media marketing.