Random Forest:
It is a supervised algorithm for classification and regression, operated by a huge number of decision trees. Actually, it returns the most selected class in a classification problem or the mean in a regression problem. So as they are a lot, it deals with overfitting a single decision tree.


In essence, we want to predict how much occupied the rack is going to be. In essence, these decisions will be based mostly on items like Time, Weekday, Weather, Location of the rack on the campus and more TBC.

The algorithm for the realization of the Decision trees is the ID3, however, as it is only able to deal with categorical features, and as this project will consider numerical values as well, then a type of quantization strategy may be needed, in order to turn those features categorical and apply ID3 algorithm again. This strategy is called C4.5. In essence, a threshold is taken and 1 and 0 are gotten according to the behavior regarding the threshold.
ARIMA is an acronym that stands for Auto-Regressive Integrated Moving Average composed of:
AR Autoregression: A model that uses the dependent relationship between an observation and some number of lagged observations
I Integrated: The use of differencing of raw observations in order to make the time series stationary
MA Moving Average: A model that uses the dependency between an observation and a residual error from a moving average model applied to lagged observations
