Stock market trend prediction with sentiment analysis based on LSTM neural network

Xu Jiawei, Tomohiro Murata

Research output: Contribution to journalConference article

Abstract

—This paper aims to analyze influencing factors of stock market trend prediction and propose an innovative neural network approach to achieve stock market trend prediction. With the breakthrough of deep learning recently, there occurred lots of useful techniques for stock trend prediction. This thesis aims to propose a method of feature selection for selecting useful stock indexes and proposes deep learning model to do sentiment analysis of financial news as another influencing factor influencing stock trend. Then it proposes accurate stock trend prediction method using LSTM (Long Short-term Memory).

Original languageEnglish
Pages (from-to)475-479
Number of pages5
JournalLecture Notes in Engineering and Computer Science
Volume2239
Publication statusPublished - 2019 Jan 1
Event2019 International MultiConference of Engineers and Computer Scientists, IMECS 2019 - Kowloon, Hong Kong
Duration: 2019 Mar 132019 Mar 15

    Fingerprint

Keywords

  • Chinese Stock market
  • Deep learning
  • Feature Selection…
  • LSTM
  • Sentiment Analysis
  • Stock trend prediction

ASJC Scopus subject areas

  • Computer Science (miscellaneous)

Cite this