International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 04 | Apr 2024
p-ISSN: 2395-0072
www.irjet.net
Stock Market Prediction Using LSTM and Machine Learning Techniques Dr. Chethan H K , Ms. Chaya Devi SN , Mr. Vinod Kumar C ,Mr. Vijay M 1 Professor, Dept. of Computer Science and Engineering, Maharaja Institute of Technology, Thandavapura
2,3,4, 5 Students, Dept of Computer Science and Engineering, Maharaja Institute of Technology, Thandavapura
---------------------------------------------------------------------***--------------------------------------------------------------------Assisting in the decision to purchase, sell, or hold Abstract - The financial market has seen significant
has been the main goal of stock market forecasting. By utilizing the Streamline in this application we can, make dynamic graphs of a particular company's financial data. Furthermore, we can forecast future stock prices using a machine learning algorithm. Any corporation of one's choosing (with a recognized stock code) may use this online program.
involvement from the stock market. A minor commodity can experience varying effects as a result of stock market fluctuations. Investors drawn to the company by its market value or stock price for the company's expansion. Therefore, making an accurate stock price prediction might be quite beneficial. As a result, scholars have focused on this subject and have created numerous models and studies over the years. This research proposes a new framework for stock price prediction that combines many models (Long Short Term Memory (LSTM) model, Support Vector Regression (SVR) model, Linear Regression model, and Sentimental Analysis) into a hybrid model.
1.1 Objective Our stock market prediction web app aims to give consumers precise predictions and insights so they can make wise investing choices. Our platform uses real-time data analysis and sophisticated machine learning algorithms to forecast future stock movements with high accuracy. This allows users to take advantage of opportunities and reduce risk in the ever-changing financial markets. Furthermore, our intuitive interface and adaptable features guarantee smooth navigation and customized suggestions based on individual investing objectives and risk tolerances, enabling customers to successfully traverse the intricacies of the stock market.
It is evident from the simulation results that our suggested strategy can predict future stock trends with this hybrid model when the hyper parameters are adjusted appropriately. Online datasets for stock markets with open, high, low, and closing prices are used to conduct the assessments.
Key Words: Long Short Term Memory (LSTM), Support
Vector Regression (SVR), Linear Regression, Sentimental Analysis.
2. PROPOSED SYSTEM
1. INTRODUCTION
The data for our proposed method will come from yfinance, which is an internet source. Using the presented approaches, an assembling model (i.e., Linear, SVR, LSTM, and regression. The algorithms are selected based on how well they performed, as determined by the results of the literature review. Since stocks are thought to be a combination of linear and non-linearity, several techniques are applied to improve accuracy. Sentiment analysis has also been conducted in conjunction with this model to test polarity. In addition to being more beneficial for SVM, we can learn from both positive and negative tweets, which can assist us in obtaining professional guidance.
These days, investing in stocks seems to be an additional source of revenue. While some invest as a primary source of income, others do it as a retirement plan. By retaining their shares locally under the company's name, they strengthen their bonds with their employees. Even though stocks are prone to volatility, discerning investors can make more informed decisions about which firm to invest their money in by visualizing share prices across a range of characteristics. With today's technology, one of the most prominent fields is data visualization. Since advances in machine learning, stock market analysis and forecasting have become a significant and developing trend. These analysis techniques are used by brokerage businesses, financial institutions, the banking industry, and other sectors in order to obtain familiarity with stock scores. Businesses employ this to shield themselves from the possibility that their investments will cause their share price to decline. Government agencies in both developed and emerging nations utilize stock analysis to boost their economy since it influences the pricing of other goods on the market. Giving the unstable enterprises security assurances has been a crucial component of the monitoring.
© 2024, IRJET
|
Impact Factor value: 8.226
|
ISO 9001:2008 Certified Journal
|
Page 1046