Experimental Performance Analysis of Machine Learning Algorithms

Abstract

AbstractMachine Learning models and algorithms have become quite common these days. Deep Learning and Machine Learning algorithms are utilized in various projects, and now, it has opened the door to several opportunities in various fields of research and business. However, identifying the appropriate algorithm for a particular program has always been an enigma, and that necessitates to be solved ere the development of any machine learning system. Let’s take the example of the Stock Price Prediction system, it is used to identify the future asset prediction of a industry or other financial aspects traded on a related transaction. Now, it is a daunting task to find the right algorithm or model for such a purpose that can predict accurate values. There are several other systems such as recommendation systems, sales prediction of a mega-store, or predicting what are the chances of a driver meeting an accident based on his past records and the road they’ve taken. These problem statements require to be built using the most suitable algorithm and identifying them is a necessary task. This is what the system does, it compares a set of machine learning algorithms while determining the appropriate algorithm for the selected predictive system using the required data sets. The objective is to develop an interface that can be used to display the result matrix of different machine learning algorithms after being exposed to different datasets with different features. Besides that, one can determine the most suitable (or optimal) models for their operations, using these fundamentals. For experimental performance analysis several technologies and tools are used including Python, Django, Jupyter Notebook, Machine Learning, Data Science methodologies, etc. The comparative performance analysis of best known five time series forecasting machine learning algorithms viz. linear regression, K – nearest neighbor, Auto ARIMA, Prophet, and Support Vector Machine is done. Stock market, earth and sales forecasting data is used for analysis.