Performance evaluation of deep learning stock price by chart type for buying policy verification
- Authors
- Song, Yoojeong; Lee, Jongwoo
- Issue Date
- Nov-2018
- Publisher
- IOS Press
- Keywords
- Buying policy; Stock chart analysis; Stock prediction; Technical analysis
- Citation
- Frontiers in Artificial Intelligence and Applications, v.309, pp 646 - 652
- Pages
- 7
- Journal Title
- Frontiers in Artificial Intelligence and Applications
- Volume
- 309
- Start Page
- 646
- End Page
- 652
- URI
- https://scholarworks.sookmyung.ac.kr/handle/2020.sw.sookmyung/4779
- DOI
- 10.3233/978-1-61499-927-0-646
- ISSN
- 0922-6389
1535-6698
- Abstract
- Recently, deep learning has been widely applied to various fields, and various researches are being conducted in the stock price prediction. The study of stock price prediction is mainly composed of studying the structure of learning model, performance evaluation according to input features, and generating profit rate using stock price prediction model. Stocks range from very volatile stocks to stable stocks. Therefore, appropriate prediction models will be needed for each stock group. In this paper, we collected stocks by chart type using advanced filtering technology. There are three types of charts, and deep learning were conducted for each type. We experimented to see which chart types have the best prediction performance. As a result, the Nosedive chart shape showed the highest prediction performance in the others © 2018 The authors and IOS Press. All rights reserved.
- Files in This Item
-
Go to Link
- Appears in
Collections - ICT융합공학부 > IT공학전공 > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.