Please use this identifier to cite or link to this item:
http://dspace.dtu.ac.in:8080/jspui/handle/repository/18933
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | MALPANI, RISHABH | - |
dc.date.accessioned | 2022-02-21T08:47:50Z | - |
dc.date.available | 2022-02-21T08:47:50Z | - |
dc.date.issued | 2021-06 | - |
dc.identifier.uri | http://dspace.dtu.ac.in:8080/jspui/handle/repository/18933 | - |
dc.description.abstract | We are probably living in the clearest era of human history. An era in which computing has moved from large-scale mainframes to PCs and the cloud. But that's not what happened, it's something we can think of over the years to come. Trip Distance Prediction is important in the development of mobility-on-demand and travel information systems. Accurate estimates of travel distance support the decision-making process for riders and drivers using such systems. In this project, the static trip distance of a taxi trip trajectory is predicted by applying some regression model to a highly conditioned set of trips. We are using the NYC Taxi data set which is available on Kaggle in which, so many rich features are present like Locations, duration Distance etc. Also, we are going to use classification on the datasets and predict the Trip type. It is important to compare the results for all the different Algorithms so that we can analyse the best algorithm. | en_US |
dc.language.iso | en | en_US |
dc.relation.ispartofseries | TD-5513; | - |
dc.subject | TRIP DISTANCE PREDICTION | en_US |
dc.subject | MACHINE LEARNING | en_US |
dc.subject | MOBILITY ON DEMAND | en_US |
dc.title | TRIP DISTANCE PREDICTION & RESULT COMPARISON USING MACHINE LEARNING | en_US |
dc.type | Thesis | en_US |
Appears in Collections: | M.E./M.Tech. Computer Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Rishabh Malpani M.tech..pdf | 3.27 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.