Main Article Content
Background/Objectives: In this study, a model study was conducted to introduce Chetbot, which is being introduced by governments and companies through the use of artificial intelligence, into the knowledge consulting industry.
Methods/Statistical analysis: In this research analysis, we analyze how Chatbot, which is being utilized in various fields, thinks about its utilization through portal and social SNS text mining and sentiment analysis when utilized for knowledge consensus. The analysis software used the Python Ver3.8 and Textom Ver.4.5 programs, which were analyzed by scrolling through portal text for three months at Google, Naver, and Daum.
Findings: Chatbot is divided into Q&A chatbot and AI-based counseling chatbot.
Technical factors include pattern recognition, natural language processing, semantic web, text mining, and situation recognition computing. These two conditions were identified and the study was conducted.
The construction process of the Chatbot consists of user scenario definition - pre-registration - pattern registration - exception processing - query registration.
Six indicators for evaluating the performance of the Chatbot were collected by scrolling through three months of data based on Comprehensive Capabilities, User engagement, Speed, Functionality, Interoperability, and Scalability.
In order to introduce knowledge services to Chatbot, the core of chatbot accuracy can be seen as driving various algorithm tasks countless times to increase accuracy by 90%. The purpose of the service, such as which knowledge consulting tasks the chatbot will replace, was also set at the consulting stage and a knowledge consulting chatbot model was derived.
The most important thing in building knowledge consulting chatbots is response rate and accuracy. It should be designed according to the purpose of the service through text mining, etc.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.