Proposed Research
Time is the most precious resource that human being has. According to a survey an average American is spending 2340 mins in traffic and conditions are worst in many well-developed cities like Los Angeles, where on an average commuter are spending 6120 hours in traffic jams. These non-productive hours could be used for many novel and humanitarian work. The goal of this research is to predict the traffic week in advance which will help commuter to plan his/her days better. This will lead us to have a better work-life balance, less traffic, better ecosystem in terms of pollution, fuel saving and more productivity. This paper will be using metro city traffic data set for research and analysis. It’s a time series dataset and hence various time series algorithm like - Autocorrelation function (ACF), Partial autocorrelation function (PACF), Auto Regressive Moving Average Model (ARMA) and Auto Regression Integration Moving Average Model (ARIMA) will be analyzed. Along with this research paper will also analyze a comparative study on conventional time series and deep learning algorithm (i.e. long short-term memory) deep learning algorithm.