Abstract
This study aims to investigate the impact of social sentiment on firm performance similarity in financial markets. We analyze social sentiment towards a firms’ business using opinion mining techniques, and develop a social relation revealing firms’ business similarity based on the output. We found that when social trend and social sentiment about a firm vary on social media, people are likely to change their view of the firm. We found social trend to be most influential followed by negative sentiment. We use stock trade volume as an indicator of firms’ performance in two different channels (traditional and mobile), and develop social relations showing firms’ business similarity based on the output. Our findings suggest that social relations inferred from social sentiment are related to firm performance similarity. More interestingly, the relationship is stronger in the mobile channel, but it is even negative in the traditional channel.
Originalsprog | Engelsk |
---|---|
Titel | ICIS 2017 Proceedings |
Antal sider | 18 |
Udgivelsessted | Atlanta, GA |
Forlag | Association for Information Systems. AIS Electronic Library (AISeL) |
Publikationsdato | 2017 |
Status | Udgivet - 2017 |
Udgivet eksternt | Ja |
Begivenhed | 38th International Conference on Information Systems: Transforming Society with Digital Innovation, ICIS 2017: Transforming Society with Digital Innovation - Coex Convention Center , Seoul, Sydkorea Varighed: 10 dec. 2017 → 13 dec. 2017 Konferencens nummer: 38 https://icis2017.aisnet.org/ |
Konference
Konference | 38th International Conference on Information Systems: Transforming Society with Digital Innovation, ICIS 2017 |
---|---|
Nummer | 38 |
Lokation | Coex Convention Center |
Land/Område | Sydkorea |
By | Seoul |
Periode | 10/12/2017 → 13/12/2017 |
Internetadresse |
Navn | Proceedings of the International Conference on Information Systems |
---|---|
Nummer | 8 |
ISSN | 0000-0033 |
Emneord
- Social relations
- Social trend
- Social sentiment
- Firm performance similarity
- Decision tree
- QAP
- ERGM
- Bayesian