Are You a Pessimistic OR an Optimistic Person?
By Aditi Bhat
Are you the calming voice of reason in a boiling hot comments-thread about a touchy topic? Or are you the one fueling the fire? Do you usually leave only negative ratings when you’ve had bad experiences? Or are you the one who’s always gushing about your favorite new restaurant all over social media? If your online behavior is anything like your everyday personality, then these social media sites and restaurants have you figured out!
Your social media will give you an idea:
Language processing tools can now figure out overall tone in social media posts. Sentiment analysis measures positive and negative keywords in the sentences you type out. It deciphers if your review or comment has you feeling good or bad. It’s not hard to extrapolate from there. Tracking data about your behavior patterns across social media platforms, over time, can indicate if you’re a positive or negative person online. Now let’s say you’re positive when everyone else is being negative, consistently across platforms and over time. It could be pretty easy to conclude that you’re an optimist – the ray of sunshine on a cloudy day. And the same applies to someone who is the cloudy day.
Technology is tracking you, every step of the way:
At this point you’re probably feeling like you’ve had your privacy violated. Or that your seemingly harmless social media posts and comments shouldn’t be used to categorize you as a type. It’s understandable. To avoid being tracked like this you would need to get off all social media platforms, implement a strict regime of clearing browser-history, cookies and credit card information. But you don’t need to worry that much. For the most part, business analysts performing sentiment analysis are looking for trends and aggregated data to collect as feedback mechanisms. For example, restaurants can aggregate data across platforms for brand management. Sentiment analysis on Twitter has helped in tracking reactions during the US presidential elections. Moving beyond the analysis of text, call centers could use audio data to track and measure trends in customer service.
That’s not it, there’s more:
Marketing and feedback mechanisms are not the only activities sentiment tracking can do. With some additional demographic information, a data scientist could draw connections between optimism and pessimism across age, gender, economic status, education level. And the list goes on. For instance – Microsoft Labs that predicted postnatal depression just by tracking one user’s twitter feed. The potential for sentiment analysis is huge. With consent, mining the social media interaction can reveal behavioral patterns or sudden changes in personality. Doing so can potentially identifying disorders that people might not even know they have.
There is a lot of work to be done in refining algorithms and going beyond a linear positive to negative scale. The opportunities for data science applications are endless.