ETRI-Knowledge Sharing Plaform

ENGLISH

성과물

논문 검색
구분 SCI
연도 ~ 키워드

상세정보

학술대회 Forecasting Time-Series Trends by Merging Structured and Unstructured Datasets
Cited 0 time in scopus Download 7 time Share share facebook twitter linkedin kakaostory
저자
박지상, 조현성, 이지성, 정교일, 김정민, 김동진
발행일
202010
출처
International Conference on Information and Communication Technology Convergence (ICTC) 2020, pp.1230-1233
DOI
https://dx.doi.org/10.1109/ICTC49870.2020.9289258
협약과제
20PS1100, 빅데이터 및 AI 기반의 투자 및 자산관리 지원 서비스 시스템 개발, 박지상
초록
This paper introduces a new approach to forecast daily stock trends by merging structured and unstructured datasets. This study intends to reveal the effectiveness of using supplemental datasets for accurate prediction of stock prices. A set of features, which is seemingly highly correlated with daily stock price variations, are selected using random forest optimization technique. Stock-relevant keywords that are extracted from news articles are converted into a time-series dataset in terms of temporal frequency. Convolution neural network (CNN) based deep learning models are generated separately for stock trading data and keyword frequencies from news articles, and two CNN models are merged together for training input datasets. The analysis results show that merging two different datasets may generate the better forecasting results than using stock trading datasets only. Additional issues for future analysis and implementations are discussed.
키워드
CNN, machine learning, stock price, structured unstructured datasets
KSP 제안 키워드
Accurate prediction, Convolution neural network(CNN), New approach, News articles, Optimization techniques(OT), Random forest, Stock prices, Stock trading, Time series, deep learning(DL), deep learning models