Please use this identifier to cite or link to this item: http://dspace.dtu.ac.in:8080/jspui/handle/repository/15924
Full metadata record
DC FieldValueLanguage
dc.contributor.authorKUMAR, SACHIN-
dc.date.accessioned2017-08-28T12:12:37Z-
dc.date.available2017-08-28T12:12:37Z-
dc.date.issued2017-07-
dc.identifier.urihttp://dspace.dtu.ac.in:8080/jspui/handle/repository/15924-
dc.description.abstractEmotions are the main influence of human behavior. Emotions can be defined as an opinion or feeling, a neutral feeling or an attitude towards something. The sense test is a method of computationally recognizing and classifying the opinions outlined in a piece of text, especially to determine whether the authors are positively, negatively or neutral towards a careful topic, product etc.. Inspiration of the senses uses data mining techniques, extract and capture natural language resources and text classification data and analyze it over here. Our focus here is to make sense analysis on twitter data. Millions of Tweets are posted daily on microblogging, which basically reflects the thoughts and feelings of users of the world. These tweets can be present in different languages. Our goal is to make sense analysis in many languages. Therefore, we propose a system that uses the Google translation API so that tweets can be converted into English language in many languages and then be able to analyze the feelings about it. Before translation, we must prioritize the Tweets to filter useful information from raw data. Multilingual cement analysis has been done using different classification algorithms. These algorithms are basically performing well, but when we apply deep learning, we have seen all of these algorithms so far. Our proposed methodology is used to describe multi-lingual sense analysis, using multi-lingual sense analysis more efficiently and accurately, with the accuracy of 63.5% and 73.5% respectively, using the naive Bayes and Max Entropy algorithms, the multilingual signal analysis has been described. And our deep neural network provides the accuracy of 93.5%.en_US
dc.language.isoenen_US
dc.relation.ispartofseriesTD-2905;-
dc.subjectMULTILINGUAL SEMANTIC ANALYSISen_US
dc.subjectNEURAL NETWORKen_US
dc.subjectEMOTIONSen_US
dc.titleMULTILINGUAL SEMANTIC ANALYSIS USING NEURAL NETWORKen_US
dc.typeThesisen_US
Appears in Collections:M.E./M.Tech. Computer Engineering

Files in This Item:
File Description SizeFormat 
Multilingual Semantic analysis using Neural Network.pdf2.16 MBAdobe PDFView/Open


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