INTRODUCTION Predicting how the securities exchange will perform is perhaps the most troublesome thing to do. There are countless elements involved with the forecast: actual variables versus mental, levelheaded, and silly conduct, and so forth This large number of viewpoints joins to make share costs unstable and truly challenging to foresee with a serious level of exactness. We can use various machine learning algorithms to predict stock prices like Random Forest, Support Vector Machine, KNN, linear regression, etc.
1.2 OBJECTIVES • Machine learning is the ability of computers to learn new things based on previous data, experiences, and observations. • The algorithm analyzes historical stock data. It compares the data with current stock prices. • It predicts the mean squared error and gives us the prediction value. • The results can then be compared to the real-life performance of stocks to check whether the algorithm performed accurately or if it still needs improvements.
1.3 METHODOLOGY TO BE FOLLOWED The first step in this process is to import all the required libraries. We will use several libraries here, including some machine-learning libraries. A library is a collection of modules. It contains a bundle of codes that can be used repeatedly. Next, we read or load the CSV file. The dataset will have several rows and columns containing raw data. We will be using this data to make our prediction. Datasets also contain some unwanted fields which won’t be required or used. So, we can remove those unwanted columns. The next step is important. Here, we split the information into
In this project, we split the data into 75:25 ratios for training and testing sets. After splitting the data, we create the regressor from Sci-kit and do linear regression. Lastly, we make the predictions and evaluate the results.
