Research Project Management, part 1: Introduction to Sentiment Analysis

This is the first article in a series of four articles about postgraduate research project management.

Published on 19 June 2022

Research Project Management, part 1: Introduction to Sentiment Analysis

Sentiment Analysis, also known as Opinion Mining, is a branch of Natural Language Processing, that aims to extract people’s opinions, sentiments, or emotions through computational linguistics and data mining techniques (Wongkar and Angdresey, 2019).

Usually, the main objective is to extract meaning and to understand people’s opinions on any topic, event, individual, and so on (Mandloi and Patel, 2020). In general, this method’s key objective is to classify each feature’s polarity within a text into one of the three (or more) sentiment scores: positive, negative, or neutral (Wongkar and Angdresey, 2019). There are numerous sources of information that can be used to perform sentiment analysis, such as review websites (Birjali, Kasri, and Beni-Hssane, 2021), social media platforms (Chinnasamy et al., 2022), or any other publicly accessible sources of information (Yue et al., 2019). However, in terms of this project, social media platforms, especially Twitter, are particularly interesting.

In a digitally connected world, social media platforms like Twitter are often peoples’ preferred choice of medium to express and share information, opinions or ideas (Chinnasamy et al., 2022). Consequently, Twitter and other social media platforms provide a wealth of publicly available information that can be used by researchers to analyse human behaviour (Fiesler, Beard, and Keegan, 2020), beliefs, and perceptions (Chinnasamy et al., 2022) and attitudes (Liu and Liu, 2021). However, sentiment analysis can be multi-dimensional. Due to tweets’ geolocation data, it is possible to conduct geospatial analysis, for example, to create heat maps that can be used to visually examine the differences in the intensity and spread of public opinion across a certain geographical area (Forati and Ghose, 2021).

The geospatial analysis can be especially useful in determining the effect of geography on people’s perceptions about a topic and how different geographical features, e.g., cities and rural areas affect people’s opinions. There are numerous ways to manage a sentiment analysis research project, however, there is no general consensus about what methods should be followed. According to Wongkar and Angdresey (2019), sentiment analysis generally follows a six steps workflow that encompasses keyword selection, tweets retrieval, data pre-processing, sentiment detection, and classification followed by the analysis of the output. Therefore, the project management tools and techniques will be dependent on those methods. For example, keyword selection activities have to be planned accordingly to ensure that the following activities can be carried out using a dataset relevant to the problem that must be solved. On the other hand, what risks are associated with managing a project of this type and scope? What preventative measures can be used to ensure the successful completion of the project? What are the legal, social, ethical, and professional considerations when collecting and analysing user-generated content?

Conclusion

This article provided an introduction to the sentiment analysis research project. The following article will look at the tools and techniques to manage a project of this type and scope, and explore how those methods could be used to manage sentiment analysis project and the benefits they may provide.

References

Birjali, M., Kasri, M. and Beni-Hssane, A. (2021). A comprehensive survey on sentiment analysis: Approaches, challenges and trends. Knowledge-Based Systems, [online] 226, p.107134. doi:10.1016/j.knosys.2021.107134.

Chinnasamy, P., Suresh, V., Ramprathap, K., Jebamani, B.J.A., Srinivas Rao, K. and Shiva Kranthi, M. (2022). COVID-19 vaccine sentiment analysis using public opinions on Twitter. Materials Today: Proceedings. [online] doi:10.1016/j.matpr.2022.04.809.

Forati, A.M. and Ghose, R. (2021). Geospatial analysis of misinformation in COVID-19 related tweets. Applied Geography, [online] 133, p.102473. doi:10.1016/j.apgeog.2021.102473.

Mandloi, L. and Patel, R. (2020). Twitter Sentiments Analysis Using Machine Learninig Methods. 2020 International Conference for Emerging Technology (INCET). [online] doi:10.1109/incet49848.2020.9154183.

Wongkar, M. and Angdresey, A. (2019). Sentiment Analysis Using Naive Bayes Algorithm Of The Data Crawler: Twitter. 2019 Fourth International Conference on Informatics and Computing (ICIC). [online] doi:10.1109/icic47613.2019.8985884.

Yue, L., Chen, W., Li, X., Zuo, W. and Yin, M. (2018). A survey of sentiment analysis in social media. Knowledge and Information Systems, 60(2), pp.617–663. doi:10.1007/s10115–018–1236–4.

A view of the Newcastle-Gateshead Quaside from the Tyne Bridge

Let's work together to bring your digital dream to life.

Get in touch to book a free consultation