Social Media and Stock Market Prediction: A Big Data Approach

• DOI: 10.32604/cmc.2021.014253
• Publication Date: 2021-02-23T02:30:14Z
• Authors: Mazhar Javed Awan; Mohd Shafry Mohd ...; Haitham Nobanee; Ashna Munawar; Awais Yasin; Azlan Mohd Zain A...
• Keywords: Stock (firearms)

Big data is the collection of large datasets from traditional and digital sources to identify trends patterns. The quantity variety computer are growing exponentially for many reasons. For example, retailers building vast databases customer sales activity. Organizations working on logistics financial services, public social media sharing a sentiments related price products. Challenges big include volume in both structured unstructured data. In this paper, we implemented several machine learning models through Spark MLlib using PySpark, which scalable, fast, easily integrated with other tools, has better performance than models. We studied stocks 10 top companies, whose historical stock prices, such as linear regression, generalized random forest, decision tree. naive Bayes logistic regression classification Experimental results suggest that provide an accuracy 80%–98%. experimental tree did not well predict share movements market.