Trade the tweet: Social media text mining and sparse matrix factorization for stock market prediction

Sun, Andrew; Lachanski, Michael; Fabozzi, Frank J.

doi:10.1016/j.irfa.2016.10.009

Cited by 105 publications

(39 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In the SMF, abnormal stock price returns are correlated with features extracted from corporate annual reports. In agreement with related studies [73,74], we used 2000 unigrams to build a latent factor model. A sparse group lasso regularization term was included to eliminate irrelevant unigrams.…”

Section: Resultsmentioning

confidence: 78%

See 1 more Smart Citation

Combining bag-of-words and sentiment features of annual reports to predict abnormal stock returns

Hájek

2017

Neural Comput & Applic

View full text Add to dashboard Cite

Automated textual analysis of firm-related documents has become an important decision support tool for stock market investors. Previous studies tended to adopt either dictionary-based or machine learning approach. Nevertheless, little is known about their concurrent use. Here we use the combination of financial indicators, readability, sentiment categories and bag-of-words (BoW) to increase prediction accuracy. This paper aims to extract both sentiment and BoW information from the annual reports of U.S. firms. The sentiment analysis is based on two commonly used dictionaries, namely a general dictionary Diction 7.0 and a finance-specific dictionary proposed by Loughran and McDonald [1]. The BoW are selected according to their tf-idf. We combine these features with financial indicators to predict abnormal stock returns using a multi-layer perceptron neural network with dropout regularization and rectified linear units. We show that this method performs similarly as Naïve Bayes and outperforms other machine learning algorithms (Support Vector Machine, C4.5 decision tree, and k-nearest neighbour classifier) in predicting positive/negative abnormal stock returns in terms of ROC. We also show that the quality of the prediction significantly increased when using the correlation-based feature selection of BoW. This prediction performance is robust to industry categorization and event window.

show abstract

Section: Resultsmentioning

confidence: 78%

“…with other state-of-the-art methods, we selected multinomial inverse regression (MNIR) [71,72] and sparse matrix factorization (SMF) [73,74]. The MNIR uses multinomial regression to map from BoW to the class space via relevant variables.…”

Section: Resultsmentioning

confidence: 99%

Combining bag-of-words and sentiment features of annual reports to predict abnormal stock returns

Hájek

2017

Neural Comput & Applic

View full text Add to dashboard Cite

show abstract

“…Sun and Fabozzi [32] tested their model using the majority of stocks listed on the S&P 500 index. They did not evaluate sentiment, but they analyzed textual information from microblogs.…”

Section: Sentiment Analysis and Social Media In The Stock Marketmentioning

confidence: 99%

Does Twitter affect Stock Market Decisions?Financial Sentiment Analysis in Pandemic Seasons: A Comparative Study of H1N1 and COVID-19

Valle-Cruz

Fernandez-Cortez

López-Chau

et al. 2020

Preprint

View full text Add to dashboard Cite

Abstract Backgroud:Investors are always playing with the fears and desires of buyers and sellers. Stock exchange markets are not the exception. Financial sentiment analysis allows us to understand the effect of reactions and emotions on social media in the stock market. In this research, we analyze Twitter data and financial indices to answer the question: How do polarity generated by the posts on Twitter influence financial indices behavior in pandemic seasons? Methods:The study is based on the sentiment analysis of influential Twitter accounts in this field and its relationship with the behavior of important financial indices. To achieve this, we tested four lexicons to detect polarity on Twitter. Results:Our findings shows that the period in which the markets reacted was 6 to 13 days after the information was shared and disseminated on Twitter in the COVID-19 season, and 1 to 2 day for H1N1 season. Furthermore, in our analysis, we found that the lexicons that got the best results for sentiment analysis on Twitter were S140 and Affin.Conclusions:Financial sentiment analysis is an important technique to forecasting stock market and polarity is the most widely used technique in the financial area. There is a relationship between the polarity in Twitter and the financial indexes behavior. The most influential Twitter accounts during the pandemic season were The New York Times, Bloomberg, CNN News, and Investing, presenting a very high relation between sentiments on Twitter and the stock market behavior.

show abstract

“…Rather than considering the sentiment content of social media information, Sun, Lachanski, and Fabozzi (2016) focus on the usefulness of textual information extracted from social media to predict the stock returns. In a different approach, the authors -based on a sample of 45 million messages posted on StockTwits platform during the period (2011-2015) -firstly create a dictionary of terms via examining the top words of highest frequency for each year.…”

Section: Twitter Sentiment/information and Stock Market Performancementioning

confidence: 99%

Twitter, Investor Sentiment and Capital Markets: What Do We Know?

Ali¹

2018

IJEF

View full text Add to dashboard Cite

Nowadays, the social media play a central role not only in "de-asymmetrizing" the information between firms and investors but also in influencing the emotional response to this information. The social media have provided firms with the opportunity to construct their image and stimulate significant attention and positive emotional responses (i.e. celebrity firm). Investors also become no longer passive participants; they can now communicate, re-tweet, comment, mention, react to information and express their sentiment/views. Theoretically, this should exert a positive impact on information diffusion and so the market efficiency. However, as the social media also significantly influence the public mood and emotional response to any new information/news, several behavioral explanations contradicting with the concept of market efficiency (e.g. investor sentiment and herding behavior) become more reasonable. The study aims at providing a literature review and synthesis of research on the impact of social media sentiment in the context of capital markets, scrutinizing the theoretical understanding of this impact, underlining the methodological challenges related to extracting the sentiment, and reviewing the main empirical findings on the impact in the context of Twitter and StockTwits, which will enable researchers to evaluate and classify existing studies, obtain useful insight into the theoretical understanding of the impact of social media sentiment, hence spurring further theoretical and empirical research.

show abstract

Trade the tweet: Social media text mining and sparse matrix factorization for stock market prediction

Cited by 105 publications

References 14 publications

Combining bag-of-words and sentiment features of annual reports to predict abnormal stock returns

Combining bag-of-words and sentiment features of annual reports to predict abnormal stock returns

Does Twitter affect Stock Market Decisions?Financial Sentiment Analysis in Pandemic Seasons: A Comparative Study of H1N1 and COVID-19

Twitter, Investor Sentiment and Capital Markets: What Do We Know?

Contact Info

Product

Resources

About