International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 10 Issue: 07 | July 2023
p-ISSN: 2395-0072
www.irjet.net
Sentiment Analysis based Stock Forecast Application Shivani R1, Devaraju B M2, Dr.Girijamma H A3 1Post Graduate Student, Department of Computer Science and Engineering, RNSIT, Karnataka, India 2Assistant Professor, Department of Computer Science and Engineering, RNSIT, Karnataka, India 3Professor, Department of Computer Science and Engineering, RNSIT, Karnataka, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - An abstract summarizes, in one paragraph
premise is that the sentiment expressed in financial news can impact investor behavior, thereby influencing stock prices. Positive sentiment may drive increased buying activity, leading to price appreciation, while negative sentiment could prompt selling and subsequent price decline. Sentiment analysis algorithms employ machine learning and NLP models to classify the sentiment of textual data accurately. These models assess the sentiment polarity (positive, negative, or neutral) and the strength of sentiment expressed. The sentiment analysis results can be combined with other fundamental and technical indicators to generate predictive models that capture market sentiment as an additional input.
(usually), the major aspects of the entire paper in the following prescribed sequence. In the contemporary era, as data continues to grow in volume and significance for businesses, manual data analysis is no longer feasible in the fast-paced world. Therefore, the adoption of artificial intelligence and data mining techniques has become imperative. Among various factors affecting stock price fluctuations, a crucial determinant is the gains or losses incurred by a company. As news is a primary source of information for most traders, it plays a pivotal role in forecasting changes in the stock market. This study focuses on sentiment classification and its influence on stock market prices. Sentiment analysis based stock prediction is a cuttingedge approach that leverages natural language processing and machine learning techniques to predict stock price movements using sentiment data from various sources. This study explores the use of sentiment analysis on financial news, social media, and other textual data to gauge market sentiment, which can serve as a leading indicator for stock price trends. There are several classifiers that could be used in sentiment analysis for stock prediction. The classifiers used in the proposed system include, Decision Tree, Logistic Regression, Naïve Bayes and LSTM.
1.1 Problem Statement In the current era, there is a challenge of accurately predicting stock prices in the financial market. The problem lies in the complex and volatile nature of the market, where stock prices are influenced by various factors, including economic indicators, company performance, and investor sentiment. The goal is to develop a reliable prediction model that can forecast future stock prices with a high degree of accuracy based on financial news. This model should leverage historical stock data, market trends, and relevant indicators to make informed predictions and assist investors in making strategic investment decisions. The aim is to overcome the inherent uncertainties and challenges associated with stock prediction, ultimately enabling users to optimize their portfolio management and maximize their investment returns.
Key Words: Machine Learning, Deep Learning, Sentiment analysis, LSTM, Naive Bayes, Sentiment, score, Logistic Regression, Decision Tree
1.INTRODUCTION Stock prediction is an intriguing field that aims to forecast the future movements of stock prices in financial markets. However, predicting stock prices accurately is challenging due to the intricacy and volatility of financial markets. Recent advancements in data science and artificial intelligence have provided new tools and approaches to improve prediction accuracy. These techniques involve the extraction and analysis of vast amounts of financial data, including historical stock prices, company news, economic indicators, and social media sentiment. Stock prediction using sentiment analysis of financial news has emerged as an innovative approach to gain insights into market trends. By harnessing natural language processing (NLP) techniques, this method involves analyzing textual data, such as news articles and social media posts, to gauge the sentiment associated with specific stocks or companies. The underlying
© 2023, IRJET
|
Impact Factor value: 8.226
2. LITERATURE SURVEY This study utilizes natural language processing techniques and recurrent neural networks (RNN), specifically long short-term memory (LSTM) networks In the field of stock forecasting. The objective is to predict stock prices using textual data. The study leverages data from Finwiz, a financial information platform, focusing on stocks of companies such as Adidas AG, Continental AG, Deutsche Lufthansa AG, Henkel AG & Co., and Siemens AG. By employing RNN with LSTM, the model aims to capture temporal dependencies and patterns in the textual data, allowing for accurate stock price predictions. The study demonstrates an accuracy rate of 80% for Adidas AG, 70% for Continental AG and Deutsche Lufthansa AG, and 65% for
|
ISO 9001:2008 Certified Journal
|
Page 956