Sentiment Analysis Using Machine Learning and Deep Learning

ABSTRACT: This research report presents and provides an overview of the methods or classifiers used for sentiment analysis. The purpose of this research is to investigate and compare different methods like Naïve Bayes, Lexicon based approach, Linear SVM and BERT for sentiment analysis, to find out the accuracy that could be obtained by using different methods on a similar dataset with Amazon customer reviews. The practical work is done using the python language and dataset in JSON format.

AIM AND OBJECTIVES: The main purpose of this project is to understand the different models or algorithms which could be used for sentiment analysis on a dataset of many customer reviews from Amazon. This understanding could also help in finding out how different models work and the accuracy of the model for better use.

TOOLS AND TECHNIQUES: • Jupyter Notebook: It is an open-source software or web environment with various services enabling interactive computing spanning various programming languages. • Libraries: There are numerous libraries used for all the classifiers worked on. nltk, pandas, matplotlib, NumPy, wordcloud, seaborn, re, sklearn, spacy, textacy, transformers, torch, tqdm. These are all the libraries used during the classifiers. Few libraries like matplotlib, wordcloud, and seaborn are used for the data visualization where tqdm library provides output in the form of a smart progress bar by showing the iterations. It also shows the elapsed time and running time. The torch and sklearn or scikit-learn are the libraries that provide support for machine learning and deep learning algorithms. The nltk, spacy, and textacy libraries help in providing support for the natural language processing tasks. The pandas, numpy, and re libraries are used for the mathematical calculations and use which will be done to find or evaluate the performance of the algorithms.

REFERENCES: [1] Morganclaypool.com. 2021. Sentiment Analysis and Opinion Mining | Synthesis Lectures on Human Language Technologies. [online] Available at: https://www.morganclaypool.com/doi/abs/10.2200/s00416ed1v01y201204hlt016 [Accessed 13 August 2021]. [2] Wolff, R., 2021. Quick Introduction to Sentiment Analysis. [online] Medium. Available at: https://towardsdatascience.com/quick-introduction-to-sentiment-analysis-74bd3dfb536c [Accessed 1 July 2021]. [3] Ghosh, S. and Gunning, D., 2019. Natural Language Processing Fundamentals. Packt Publishing. [4] Bakshi, R., Kaur, N., Kaur, R. and Kaur, G., 2021. Opinion mining and sentiment analysis. [online] Ieeexplore.ieee.org. Available at: https://ieeexplore.ieee.org/abstract/document/7724305/authors#authors [Accessed 1 July 2021]. [5] Gupte, A., Joshi, S., Gadgul, P., Kadam, A. and Gupte, A., 2014. Comparative study of classification algorithms used in sentiment analysis. International Journal of Computer Science and Information Technologies, 5(5), pp.6261-6264. [6] Medhat, W., Hassan, A. and Korashy, H., 2014. Sentiment analysis algorithms and applications: A survey. [online] ScienceDirect. Available at: https://www.sciencedirect.com/science/article/pii/S2090447914000550#s0010 [Accessed 15 August 2021]. [7] Suppala, K. and Rao, N., 2019. Sentiment analysis using naïve bayes classifier. Int. J. Innov. Technol. Explor. Eng, 8(8). Available at: http://www.zeynepaltan.info/3-SentimentAnalysiswithNAiveBAyes.pdf [Accessed 16 August 2021], pp.264-269. [8] Shwartz, S., 2021. 12 Twitter Sentiment Analysis Algorithms Compared. [online] AI Perspectives. Available at: https://www.aiperspectives.com/twitter-sentiment-analysis/#:~:text=The%20Winner,twitter%20sentiment%20analysis%20approaches%20tested. [Accessed 1 July 2021]. [9] Horakova, M., 2015. Sentiment analysis tool using machine learning. Global Journal on Technology. [10] Palanisamy, P., Yadav, V. and Elchuri, H., 2013. Serendio: Simple and Practical lexicon-based approach to Sentiment Analysis. [online] Aclanthology.org. Available at: https://aclanthology.org/S13-2091.pdf [Accessed 16 August 2021]. [11] Gupte, A., Joshi, S., Gadgul, P., Kadam, A. and Gupte, A., 2014. Comparative study of classification algorithms used in sentiment analysis. International Journal of Computer Science and Information Technologies, 5(5), pp.6261-6264. [12] Rathee, N., Joshi, N. and Kaur, J., 2018. Sentiment Analysis Using Machine Learning Techniques on Python. [online] Ieeexplore.ieee.org. Available at: https://ieeexplore.ieee.org/abstract/document/8663224 [Accessed 16 August 2021]. [13] Samal, B., Behera, A. and Panda, M., 2017. Performance analysis of supervised machine learning techniques for sentiment analysis. [online] Ieeexplore.ieee.org. Available at: https://ieeexplore.ieee.org/abstract/document/8071579 [Accessed 16 August 2021]. [14] Gao, Z., Feng, A., Song, X. and Wu, X., 2019. Target-Dependent Sentiment Classification With BERT. [online] Ieeexplore.ieee.org. Available at: https://ieeexplore.ieee.org/abstract/document/8864964 [Accessed 17 August 2021]. [15] Sun, C., Huang, L. and Qiu, X., 2019. Utilizing BERT for Aspect-Based Sentiment Analysis via Constructing Auxiliary Sentence. [online] Arxiv.org. Available at: https://arxiv.org/pdf/1903.09588.pdf [Accessed 17 August 2021]. [16] Ni, J., Li, J. and McAuley, J., 2019. Justifying Recommendations using Distantly-Labeled Reviews and Fine-Grained Aspects. [online] Cseweb.ucsd.edu. Available at: https://cseweb.ucsd.edu/~jmcauley/pdfs/emnlp19a.pdf [Accessed 17 August 2021]. [17] Ni, J., 2019. Amazon review data. [online] Nijianmo.github.io. Available at: https://nijianmo.github.io/amazon/index.html [Accessed 17 August 2021]. [18] GOV.UK, 1988. Copyright, Designs and Patents Act 1988. [Online] Available at: https://www.legislation.gov.uk/ukpga/1988/48/contents [Accessed 28 April 2021]. [19] GOV.UK, 2018. Guide to the General Data Protection Regulation. [Online] Available at: https://www.gov.uk/government/publications/guide-to-the-general-data-protection-regulation [Accessed 28 April 2021]. [20] Albrecht, J., Ramachandran, S. and Winkler, C., 2021. Blueprints for Text Analytics Using Python. Sebastopol: O'Reilly Media, Incorporated. [21] Deitel, P., n.d. Python Fundamentals. [22] Weyk Global. 2021. Questions in Sentiment Analysis. [online] Available at: https://weykglobal.com/top-questions-in-sentiment-analysis [Accessed 25 August 2021]. [23] Medium. 2021. Confusion Matrix for Your Multi-Class Machine Learning Model. [online] Available at: https://towardsdatascience.com/confusion-matrix-for-your-multi-class-machine-learning-model-ff9aa3bf7826 [Accessed 25 August 2021]. [24] M S, N. and R, R., 2013. Sentiment analysis in twitter using machine learning techniques. [online] Ieeexplore.ieee.org. Available at: https://ieeexplore.ieee.org/abstract/document/6726818 [Accessed 26 August 2021]. [25] Gautam, G. and Yadav, D., 2014. Sentiment analysis of twitter data using machine learning approaches and semantic analysis. [online] Ieeexplore.ieee.org. Available at: https://ieeexplore.ieee.org/abstract/document/6897213 [Accessed 26 August 2021].