mferdous2012/Site-Down-Prediction

Predictive analysis using machine learning algorithms to forecast future site down based on historical data. Initially single feature linear regression used to build the model. The popular time series used as feature in linear regression with necessary hyper-parameter tuning. Some good and promising result achieved after running the model with very high accuracy in some of the cases. In future have a plan to further improve the model with more features and high training data.