Curiosity - "Find the Unknown"

To comply with security and airline protocol, air travelers should arrive at the airport well in advance of their flight. Without realizing the probability of adverse conditions at the time of the scheduled departure, they may experience inconvenient delays at the airport. Delays can be short and relatively easy to manage, or they can cause long hours of waiting in crowded airports. Flight delays can even cause forced overnight stays at local hotels or inside the terminal. Travelers could benefit from knowing the likelihood of a delay as it could help them prepare for the wait time. Can an app be developed that predicts the impact of weather on airplane departure times?

Our solution was based on the concept of applying Machine learning and data mining to predict airline delays. We used Supervised Classification model to tackle the problem. The three algorithms chosen for the task was Naive Bayyes, Support Vector Machines and Simple Logistic Regression.The goal was to apply and test the following algorithms on the dataset and measure the accuracy of prediction. Furthermore, the maximum accuracy obtained by our program on our test set was 72% which is not bad for a supervised machine learning program. We took a step further to create a small java program using all the three algorithms with the aid of Weka,R,java,XLSTAT,ApacheHadoop JavaMachinelearninglibrary,LibSVM and other libraries. The results were quite promising and we wanted to know if we can predict the amount of time the flight will be delayed. Turns out using K-means clustering we can approximately predict the amount of time a flight will be delayed due to weather based on the previous data obtained. Our goal is to solve the headache of customers having to wait for unknown amount of minutes/hours at the airport due to flight delays. With the aid of our software, we have taken the initiative to tackle the problem and using software as a tool to benefit humanity.

Team Curiosity-" To infinity and beyond."

Softwares used - R, Weka, XLSTAT.RapidMiner Studio.

Online platforms used - Microsoft Azure Machine learning platform.

IDE used - Eclipse.

Websites used - www.stackoverflow.com, www.google.com.