An automated advisor system to suggest response after analyzing user writings in social network

In the field of deep learning persistent research is going on to train the system by applying various algorithms and techniques. With a view to developing a well trained system many language corpus are built and then let the system to recognize the data. In this paper, we have proposed and implemented an automated comment advisor system that suggest emotion for comments after extracting writings (status or comments) from a social networking site (SNS) i.e. Facebook. Our developed system analyzes the sentences in each comment, parses the sentence for tokenize words to match the corpus type and finally makes a decision whether the comment reflects positive, negative or neutral emotion of human thoughts. This system also learns from others comment sentences and finds more appropriate emotion for the neutral sentences with ambiguous words where it is hard to find any emotion from the sentence.