How Data Science Can Help You Sense Your Brand’s Effect on Social Media

Featured guest blog from our friend, Zacharias Voulgaris, Ph.D., founder of


The pulse of comments on social media

Linguistic data, like the data in free-form text that makes up most of the social media databases, isn’t easy to process. Its diversity, large volume, and inherent complexity make it challenging to understand in an automated manner. However, it’s possible to get a pulse of it so that we can at least get something insightful out of it. Sometimes that’s all we need since the details of each comment are too specific to be of interest when it comes to making decisions about a brand.

Sensing the pulse of your brand with sentiment analysis

Sensing this pulse of the social media as it’s reflected in the comments various people make on your brand (and your products or services) is not only feasible but also commonplace nowadays. There are even companies that base their whole business model on this, offering all kinds of insights stemming from the social media feeds they find.

The technology behind it is quite simple though: a well-researched technique of natural language processing called sentiment analysis. This is basically an application of machine learning on the text data from the social media feeds after the relevant text is identified and all the unnecessary words are filtered out. In a nutshell, it involves figuring out if a piece of text and its metadata (any data linked to that text, such as a timestamp, the user’s location, etc.) has a certain sentiment type. The latter is usually “positive sentiment” or “negative sentiment,” but it could be “neutral sentiment” too.

Yet, sentiment analysis goes far beyond just finding how many “positive” and how many “negative” words exist in the text. It employs all kinds of text analytics methods to identify the words or phrases that are relevant for each sentiment type and build a model around them. This fairly simple classification problem gets quite complex when the comments are riddled with abbreviations and typos, but with enough work, it can be made robust enough to handle them.

Refining the analysis towards specific products or services of your company

The logical next step in all this is to refine the analysis so that it focuses on particular products or services your company offers and see how the corresponding sentiments relate to the overall sentiment about the brand. Sentiment analysis does that too. This naturally has a lot of benefits, beyond the basic understanding of how your customers feel about what you have to offer. This kind of refined sentiment analysis can help you evaluate the effect of your marketing campaigns and if coupled with some time-series analysis, you can get a quantified effect of your marketing over time and in specific geographical locations. Also, this kind of analysis can be automated to some extent, making the insights available at the push of a button. This level of insight was unimaginable before data science.


Data science can benefit the business world in many ways, yet the one that stands out when it comes to marketing and brand awareness is sentiment analysis. This relatively straightforward machine learning technique in the natural language processing domain can add a lot of value to your marketing research. This is accomplished by measuring how positive or negative people’s sentiment is towards your brand, through their social media comments and some other data related to these comments, called metadata. Moreover, sentiment analysis can be performed for specific products or services, making the corresponding insights more applicable and providing a way to measure your marketing campaigns’ effect on your audience.

About the author

Dr. Zacharias Voulgaris was born in Athens, Greece. He studied Production Engineering and Management at the Technical University of Crete, shifted to Computer Science through a Masters in Information Systems & Technology (City University of London), and then to Data Science through a PhD on Machine Learning (University of London). He has worked at Georgia Tech as a Research Fellow, at an e-marketing startup in Cyprus as an SEO manager, and as a Data Scientist in both Elavon (GA) and G2 (WA). He also was a Program Manager at Microsoft, on a data analytics pipeline for Bing. Currently, he works as a Data Science consultant, a content creator for Technics Publications, and as a Mentor at Thinkful LLC. You can read his blog at

Share your love

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: