Optical-based Limit Order Book Modelling using Deep Neural Networks

Conference proceedings article

Authors/Editors

POJ TANGAMCHIT

Strategic Research Themes

Machine Learning (Big Data Analytics)

Publication Details

Author list: Laowatanachai P., Tangamchit P.

Publisher: Hindawi

Publication year: 2020

Start page: 1

End page: 4

Number of pages: 4

ISBN: 9781728130767

ISSN: 0146-9428

eISSN: 1745-4557

URL: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85084978133&doi=10.1109%2fiEECON48109.2020.229466&partnerID=40&md5=e19a14b098cfbcd4591083d69cce455d

Languages: English-Great Britain (EN-GB)

View in Web of Science | View on publisher site | View citing articles in Web of Science

Abstract

We used a deep neural network to create a model of order books' behaviors in a stock market using their VDO snapshots as an input. The snapshots were taken from a stock market application in time series format. Google's Tesseract OCR was used to extract price data from these snapshots. A long short-term memory (LSTM) neural network was used to learn the price behaviors in order to predict their future trends, i.e. up, down, or neutral. The result showed that the system achieved an accuracy of 68.96% despite the noise from the OCR and the sampling effect of the snapshots. © 2020 IEEE.

Keywords

Deep neural networks, Predictive models, Stock markets, Time Series Analysis