TY - GEN
T1 - An Investigation into Influences of Tweet Sentiments on Stock Market Movements
AU - Hao, Zhicheng
AU - Chen-Burger, Yun-Heh Jessica
N1 - Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
PY - 2022/8/23
Y1 - 2022/8/23
N2 - The increasingly common practice by the public of using social media as an information source and basis for decision making for investment has given social media, such as Twitter, a growing influential role in people’s behaviour. We investigate the impact of Tweets on the US stock market using sentiment analysis. We examine how the sentiments in Tweets influence the price movements in S&P 500, an excellent indicator for the US stock market and economy in general. We compared the top five influential Twitter outlets and found WSJ, Bloomberg, Forbes, and Reuters consistently showed similar sentiments. We also discovered a significant two trading days’ delay in impacting S&P 500 trends where obvious agreements are found between the above outlets. An enhanced version of SentiStrength has been used where we injected finance-related lexicons underpinned by our investor’s ontology to obtain a significantly improved and consistent performance.
AB - The increasingly common practice by the public of using social media as an information source and basis for decision making for investment has given social media, such as Twitter, a growing influential role in people’s behaviour. We investigate the impact of Tweets on the US stock market using sentiment analysis. We examine how the sentiments in Tweets influence the price movements in S&P 500, an excellent indicator for the US stock market and economy in general. We compared the top five influential Twitter outlets and found WSJ, Bloomberg, Forbes, and Reuters consistently showed similar sentiments. We also discovered a significant two trading days’ delay in impacting S&P 500 trends where obvious agreements are found between the above outlets. An enhanced version of SentiStrength has been used where we injected finance-related lexicons underpinned by our investor’s ontology to obtain a significantly improved and consistent performance.
KW - Emotive words
KW - Lexicon-based analysis
KW - Sentiment analysis
KW - SentiStrength
KW - Stock market price movements
KW - Tweets
UR - http://www.scopus.com/inward/record.url?scp=85137022907&partnerID=8YFLogxK
U2 - 10.1007/978-981-19-3359-2_8
DO - 10.1007/978-981-19-3359-2_8
M3 - Conference contribution
AN - SCOPUS:85137022907
SN - 9789811933585
T3 - Smart Innovation, Systems and Technologies
SP - 87
EP - 97
BT - Agents and Multi-Agent Systems
PB - Springer
T2 - 16th KES International Conference on Agents and Multi-Agent Systems: Technologies and Applications 2022
Y2 - 20 June 2022 through 22 June 2022
ER -