Tools
Tools
Bhatia, Danish (2017) Performance based predictive analysis of divergent classifiers for United States flight delays. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (1MB) | Preview |
Abstract
Flights are an imperative means of commutation in the transportation system for United States. Extensive reliability on them for Business or leisure, is making the flights inevitably delayed as a result of which passengers end up waiting for ages at the airport, because of which airports schedule and management gets disturbed, ultimately depleting the reputation of airlines. Accurately predicting the delay beforehand, could significantly reduce the impact of a late flight, if not completely alleviate it. Myriad researches have been done within this domain considering Machine Learning techniques and evaluating performance of a model based on Accuracy alone. However, considering only these algorithms and accuracy as a performance metric alone, is kind of biased. This research therefore, intended to propose a novel solution of analyzing the performance of four models including a Deep Learning model based upon Accuracy, Precision, Recall, F Measure and Kappa Statistics. The prime objective of this research is to get as accurate model as possible which could probably be used to assist customers to save their time. This research harnessed key insights informing that late arriving flights is the root cause of delay. Unstable weather also results in a lot of delay and eventually cancellations. This piece of information could be lucrative for the airlines as they can come up with ideas to mitigate these delays.
| Item Type: | Thesis (Masters) |
|---|---|
| Subjects: | Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science |
| Divisions: | School of Computing > Master of Science in Data Analytics |
| Depositing User: | Caoimhe Ní Mhaicín |
| Date Deposited: | 28 Aug 2018 12:53 |
| Last Modified: | 28 Aug 2018 12:53 |
| URI: | https://norma.ncirl.ie/id/eprint/3093 |
Actions (login required)
 |
View Item |