Named entity recognition and tweet sentiment derived from tweet segmentation using hadoop

Twitter is well known website famous for micro blogging where millions of users exchanging their opinions and thoughts. The tweets users are sharing has a error sum nature. The information available in tweets is insufficient. Because of character limitation tweets are short in nature many applicatio...

Full description

Saved in:
Bibliographic Details
Published in2017 1st International Conference on Intelligent Systems and Information Management (ICISIM) pp. 194 - 198
Main Authors Powar, Swati, Shinde, Subhash
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.10.2017
Subjects
Algorithm design and analysis
Batch of tweets
Electronic publishing
Encyclopedias
global context
hadoop framework
Internet
local context
Sentiment analysis
tweet segmentation
tweet sentiment
Twitter
Online AccessGet full text

Cover

More Information
Summary:Twitter is well known website famous for micro blogging where millions of users exchanging their opinions and thoughts. The tweets users are sharing has a error sum nature. The information available in tweets is insufficient. Because of character limitation tweets are short in nature many applications like Information Retrieval has problems in information retrieval. Here we are proposing a batch processing framework for tweet fragmentation called TweetSeg. TweetSeg combines information from Confined context with information from Universal context for achieving better results for Named Entity identification. Tweeter is used largely so we want to find public sentiment of tweet by segmenting the tweet into fragments where each fragment can be a named entity, we can find meaningful information from the part and analyzing the sentiments expressed in the tweets by using these fragments in Hadoop framework.
DOI:10.1109/ICISIM.2017.8122173