Please use this identifier to cite or link to this item: http://ir.lib.seu.ac.lk/handle/123456789/3012
Full metadata record
DC FieldValueLanguage
dc.contributor.authorSafeek, Ilham
dc.contributor.authorKalideen, Muhammad Rifthy
dc.date.accessioned2018-02-08T09:19:25Z
dc.date.available2018-02-08T09:19:25Z
dc.date.issued2017-12-07
dc.identifier.citation7th International Symposium 2017 on “Multidisciplinary Research for Sustainable Development”. 7th - 8th December, 2017. South Eastern University of Sri Lanka, University Park, Oluvil, Sri Lanka. pp. 69-78.en_US
dc.identifier.isbn978-955-627-120-1
dc.identifier.urihttp://ir.lib.seu.ac.lk/handle/123456789/3012
dc.description.abstractIn the growing digital world, people use social media to communicate each other and to express their mindset. Facebook known as the most popular social media among the others. People express their life events, day to day life, and knowledge that they have, their feelings, past, present, future and so on. Our aim is to suggest a suitable career path based on what they are sharing in the public facebook profiles. The main objective of this paper is to retrieve and pre-process the facebook data from a unique facebook profile. 200 facebook profiles were used for this survey. Most of the researches done preprocessing using the text that the facebook users shared. But in this paper especially we use spell correction and emoticon analysis. In most of the research papers uses some tools to pre-process the facebook or social media data. But I am translating and pre-processing from more than 70 languages with maximum possible accuracy. The final outcome of facebook data pre-processing will help for the better sentiment analysis.en_US
dc.language.isoen_USen_US
dc.publisherSouth Eastern University of Sri Lanka, University Park, Oluvil, Sri Lankaen_US
dc.subjectSentiment analysisen_US
dc.subjectLemmatizationen_US
dc.subjectGraph APIen_US
dc.subjectFacebooken_US
dc.subjectNLPen_US
dc.titlePreprocessing on Facebook data for sentiment analysisen_US
dc.typeArticleen_US
Appears in Collections:7th International Symposium - 2017

Files in This Item:
File Description SizeFormat 
ID 010.pdf253.76 kBAdobe PDFThumbnail
View/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.