INTERNATIONAL JOURNAL OF SCIENTIFIC DEVELOPMENT AND RESEARCH International Peer Reviewed & Refereed Journals, Open Access Journal ISSN Approved Journal No: 2455-2631 | Impact factor: 8.15 | ESTD Year: 2016
open access , Peer-reviewed, and Refereed Journals, Impact factor 8.15
In this paper, stock prediction has been achieved using machine learning techniques and sentiment analysis. Use of web scraping techniques is done on tweets from twitter to collect large amounts of data required for sentiment analysis. Furthermore, implementation of modern machine learning techniques like Logistic regression for analysis and Random forests for making accurate decisions has been taken into account. This has given promising results with an accuracy of 91.96%. For a long time, stock market prediction has been an area of research. The general assumption is that stock market trends take a random path. However, the research is getting closer to reliably predicting the stock market. The research is becoming more promising than ever, and it is getting very close to proving that the stock market responds to external stimuli. The aim here is to see if public opinion influences market sentiment. The scraped data from Twitter is then analysed to perform sentiment analysis. The confusion matrix technique has been used to match expected values to test values in order to determine the project's accuracy. The research establishes that it is possible to capture public sentiments through a complex corpus like twitter. The stock market prices have been predicted successfully with 91.96% accuracy establishing that market sentiment is indeed dependent on public sentiment.
Keywords:
natural language processing, nlp, sentiment analysis, stock market
Cite Article:
"Sentiment Analysis using Natural Language Processing", International Journal of Science & Engineering Development Research (www.ijsdr.org), ISSN:2455-2631, Vol.6, Issue 7, page no.237 - 241, July-2021, Available :http://www.ijsdr.org/papers/IJSDR2107038.pdf
Downloads:
000337070
Publication Details:
Published Paper ID: IJSDR2107038
Registration ID:193474
Published In: Volume 6 Issue 7, July-2021
DOI (Digital Object Identifier):
Page No: 237 - 241
Publisher: IJSDR | www.ijsdr.org
ISSN Number: 2455-2631
Facebook Twitter Instagram LinkedIn