Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/361228
Title: | Data Analytics for Improving Passenger Services in an Organized Sector |
Researcher: | Thakur Rashmi |
Guide(s): | Deshpande Manojkumar |
Keywords: | Data Analytics Engineering Engineering and Technology Engineering Multidisciplinary Passenger Services |
University: | Narsee Monjee Institute of Management Studies |
Completed Date: | 2019 |
Abstract: | Data analytics is an emerging trend in the digital era. Data analytics require processing of huge amount of data obtained from internet. Sentiment Classification is a subtype of data Analytics. It tries to gather the sentiments/opinions of the users provided as online reviews. Existing literature describes number of Sentiment classification methods. But they have reduced efficiency and require large memory for processing when used for classifying online reviews which are available at the social networking platform. newlineSentiment classification for the transportation can greatly help the service providers for provisioning a customer friendly environment in the public transport services. Presented research focuses on developing a sentiment classification framework for improving passenger services. newlineAn important contribution of this is development of Kernel Optimized-Support Vector Machine (KO-SVM) classifier for the sentiment classification. The KO-SVM classifier is implemented in MapReduce framework for improving the speed of the framework. newlineThe next contribution of this research is the development of the Online Kernel Optimized newlineSupport Vector Machine (OKO-SVM) classifier which processes the online reviews. The newlineOKO-SVM classifier uses the online incremental learner for adapting to the various newlineonline reviews. newlineThe final contribution of the presented work is the development of Affect classification newlinemodel which focuses towards the department category for the online Passenger reviews. newlineThe entire experimentation is done by considering the reviews from the train review newlinedatabase and Airlines database. The results of the proposed model are compared with newlinevarious Sentiment classification models, and are evaluated based on the metrics such as newlineSensitivity, Specificity, Accuracy and Precision. It can be concluded that the proposed newlineSentiment classifier achieved high classification performance than the existing classifiers newlineand thus can be used for the sentiment classification of Passenger reviews. newline |
Pagination: | i-ix;155 |
URI: | http://hdl.handle.net/10603/361228 |
Appears in Departments: | Department of Electronic Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
80_recommendation.pdf | Attached File | 21.63 kB | Adobe PDF | View/Open |
certificate.pdf | 138.38 kB | Adobe PDF | View/Open | |
chapter 1.pdf | 70.96 kB | Adobe PDF | View/Open | |
chapter 2.pdf | 135.74 kB | Adobe PDF | View/Open | |
chapter 3.pdf | 275.12 kB | Adobe PDF | View/Open | |
chapter 4.pdf | 318.85 kB | Adobe PDF | View/Open | |
chapter 5.pdf | 134.21 kB | Adobe PDF | View/Open | |
chapter 6.pdf | 27.15 kB | Adobe PDF | View/Open | |
table of contents.pdf | 1.72 MB | Adobe PDF | View/Open | |
title.pdf | 10.39 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: