Feature Extraction Enchancement in Users’ Attitude Detection

Abstract

The social network are the trendiest applications which are developed for sharing opinions about different topics or events e.g. Twitter. As a result, this kind of applications becomes abundant data source for NLP researchers to innovate and enhance techniques that can track users' attitudes towards target event, topic or even another person. These users' attitudes are playing a pivotal role for decision makers, so they can take an appropriate action towards users' negative or positive reactions either. This paper focuses on users' attitude detection based on new feature set and applies on different machine learning models that can monitor and enhance users' attitude identification system. Annotated emotion tweets dataset and word emotion lexicon are used in training, building, and testing classification models.

Nested Biomedical Named Entity Recognition

“…ML approaches have different types of algorithms such as Conditional Random Fields (CRF) [5], Hidden Markov Model [6] and Support Vector Machines (SVMs) [7]. ML approaches are used in many fields such as predicting behavior of users [8], Emotion detection [9], and Disease Detection [10].…”