ISSN (Online) : 2456 - 0774

Email : ijasret@gmail.com

ISSN (Online) 2456 - 0774

RESULT ANALYSIS FOR NON GEO TAGGED TWEETS IN USER TIMELINES USING LOCATION INTERFERENCE


Abstract

Abstract: Webbased life like Twitter have gotten all around well known in the previousdecade. Because of the high infiltration of cell phones, internet based lifeclients are progressively going portable. This pattern has added to cultivatedifferent area put together administrations sent with respect to internet basedlife, the achievement of which intensely relies upon the accessibility andexactness of clients' area data. In any case, just a very little part of tweetsin Twitter are geo-tagged. In this way, it is important to derive areas fortweets so as to accomplish the reason for those area based administrations. Inthis paper, we handle this issue by investigating Twitter client courses ofevents in a novel manner. Above all else, we split every client's tweet courseof events transiently into various groups, each having a tendency to infer aparticular area. Along these lines, we adjust two AI models to our setting andplan classifiers that characterize each tweet group into one of thepre-characterized area classes at the city level. The Bayes put together modelconcentrations with respect to the data increase of words with area suggestionsin the client created substance. The convolutional Long short-term memory(LSTM) model treats client created substance and their related areas assuccessions also, utilizes bidirectional LSTM and convolution activity to makearea inductions. The two models are assessed on an enormous arrangement ofgenuine Twitter information. The test results propose that our models are compellingat deducing areas for non-geotagged tweets and the models outflank the best inclass and elective methodologies altogether regarding surmising exactness. 

Keywords: Twitter,Location Inference, Bayes, LSTM.


Full Text PDF

IMPORTANT DATES 

Submit paper at ijasret@gmail.com

Paper Submission Open For October 2024
UGC indexed in (Old UGC) 2017
Last date for paper submission 30th October, 2024
Deadline Submit Paper any time
Publication of Paper Within 15-30 Days after completing all the formalities
Publication Fees  Rs.6000 (UG student)
Publication Fees  Rs.8000 (PG student)