A Machine Learning-Based Web Application for Simplifying Data Analysis and Prediction

Abstract: In our rapidly evolving world, technology advancements continue to shape our daily lives, prompting a shift towards modern and simplified techniques to accommodate our busy lifestyles. This project focuses on the utilization of machine learning algorithms, including linear regression, logistic regression, decision trees, SVM, Naive Bayes, KNN, K-means, Random Forest, dimensionality reduction algorithms, gradient boosting algorithms, and AdaBoosting algorithms, to streamline analytical work and prediction tasks. The system offers a user-friendly web interface that facilitates the loading of CSV and Excel data, allowing users to select and apply their preferred algorithm to suit their specific requirements. The system cleans the received data using data cleaning algorithms, and the user is presented with a list of options to assign algorithms to specific columns in the file. Graphs and charts generated by Google Charts based on the output of the predictions can be downloaded by the user. Additionally, the system enables users to visually compare two Excel or CSV files using charts, aiding in data analysis and comprehension. The application is developed using Django, Google Charts, Pandas, NumPy, and a MySQL database. Users can maintain distinct accounts to access their previous analytical work history conveniently. The application supports transforming various types of data into charts, allowing users to select and download the required charts.

Keywords: Machine Learning, Big Data, Charts, Data Analysis.