Building SenticNet 7

• Main Author: Perh, Zhi Hao
• Other Authors: Erik Cambria
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Engineering::Computer science and engineering::Computing methodologies::Document and text processing
• Online Access: https://hdl.handle.net/10356/147936
• Institution: Nanyang Technological University

The evolution of Artificial Intelligence (AI) has brought many possibilities in using machines to solve real-world problems and to bring conveniences to our daily lives. One of the examples that use AI in our daily lives is smart assistants like Google Assistant or Siri. This evolution has also brought new possibilities in natural language processing (NLP). From the beginning of the symbolic approach to the statistical approach and currently the neural network approach. The neural network approach is implemented with the concept of how the human brain processes data and this is also described as the deep learning technique. The sentiment analysis is usually the combination of the NLP and machine or deep learning techniques. It is a text analysis technique that identifies and determines the sentiment in a text such as positive, negative, or neutral. In this project, we will examine and evaluate the different methods of sentiment analysis techniques such as the lexicon approach and deep learning approach. This project aims to implement an error checking program as well as a semi-automated tool for synonyms and antonyms to enhance the SenticNet knowledge base. The error checking program utilizes the keyword extractor, sentiment analysis and synonyms functions. This program is used to perform checks with the existing SenticNet knowledge base for any discrepancy. The semi-automated tool is used to generate the list of synonyms and antonyms for a given word.