Track: High School STEM Competition

Abstract

In text data big data analysis, the majority of raw data is large and atypical, and analysis techniques cannot be applied. Therefore, the collected raw data removes unnecessary data through heuristic purification and old terms through purification. After that, the vocabulary frequency is calculated, visualized through word cloud technology, the core task is extracted, and the result is analyzed after the information is made. In this study, we propose a new method for improving unused terms using the external unused terms Set (DB) of Word Cloud, and derive the problems and usefulness of this method through real-world case analysis. Through this verification result, we present the utility for practical application of word cloud analysis using the proposed subdivision method.

Keywords

artificial intelligence, AI, deep learning, big data and Word Cloud