Sentiment Analysis Module for User-Friendly Tourist Information

Overview

General integrated search services typically provide results sorted only by recency or accuracy, which highlights the growing need for more user-friendly service models. There is a demand for sorting methods based on sentiment analysis—such as ranking results by positive or negative sentiment— and ensuring that documents of high public interest are displayed at the top. To address this, the Korea Tourism Organization (KTO) aims to demonstrate a user-friendly tourist information service by applying sentiment analysis modules, thereby providing an environment where these technologies can be effectively applied to social data.

Project description

The system utilizes information about tourist attractions from blogs, lifestyle, and cultural news, classifying data into eight specific categories: Satisfaction, Taste, Price, Scenery, Service, Convenience, Transportation, and Cleanliness. Each category is analyzed for positive and negative keywords; for example, if a keyword like "atmospheric" or "scenic" is detected in the Scenery category, it is classified as positive. To implement this, we selected sentiment-based keywords to develop an analysis module and created a demo page where these modules can be tested in connection with existing social data.

Key technological features

• Development of AI-based Sentiment Analysis Modules: Selecting sentiment-specific keywords and establishing a sentiment analysis API
• Application of SearchGPT-based Search Services: Supporting sentiment analysis of social data related to tourist attractions using SearchGPT and providing search results integrated with social data analytics

Expected outcomes

By selecting sentiment keywords for key tourist information and applying analysis modules, we have provided a search service that is both user-friendly and highly relatable to the public. This initiative is expected to contribute to a virtuous cycle of increasing tourist demand and fostering cultural development at various attractions. Looking ahead, the service has high potential for expansion into various fields, such as providing translation services for foreign tourists or designing personalized regional travel planners.

Launch year: 2023

Attribution

Case study courtesy of the National Information Society Agency, Republic of Korea, Apolitical's Content Partner. Sourced from Use Cases of Public AI Service, Vol.2, https://eng.nia.or.kr/site/nia_eng/ex/bbs/View.do?cbIdx=31975&bcIdx=29182&parentSeq=29215