Identification of Uncertain Semantics on Social Media Contents Using LSTM

Abstract

Social media is becoming a major information source. Nowadays, texts are being written in informal manner. In social medias information that are being shared and facts are  major concern. Social media users are utilizing them to create or obtain facts and information based on their interpretation. Therefore uncertainty identification in Social media applications is very essential. But, the present methodologies for identifying uncertainty are very imperfect in social media. This paper proposes Long Short term memory (LSTM) networks to represent the meaning of a word, phrase, or text and Recurrent Neural Networks (RNNs) to identify the very significant semantics.

Keywords: LSTM, Natural Language Processing