Sentiment analysis has never been more comprehensive

It never ceases to amaze us how far sentiment analysis has come. We know it can achieve even more and we can’t help but get excited just thinking about it.

We expect great things to come from TheySay, with whom we’ve worked for the past 2 years. A leading company in the field of advanced text and sentiment analytics, it was founded by Stephen Pulman (Professor of Computational Linguistics at Oxford University) and Dr Karo Moilanen, and headed from a business development perspective by Jon Halestrap. You may remember Stephen and Jon from our Engage 2015 conference in April.

TheySay Analytics is a fundamental part of our Healthwatch Informatics System, giving consumers of health and social care services a voice and the chance to really make a difference.

Curious about the methodology used? TheySay applies an automated analysis algorithm that mimics human affective common sense reasoning. Monitoring tweets, news, blogs and more, they identify specific keywords and keyphrases in relation to certain topics.

These are then pulled into a real-time opinion streaming system where they undergo further analysis, creating sentiment scores (positive, neutral, negative), multi-dimensional emotion scores (the amount of sadness, anger or surprise), speculation and intent analysis (mentions of the future), as well as humour detection.

But that’s not all. Automated Natural Language Processing, Machine Learning and Deep Learning methods are then used to interpret text on various levels.

Have a look at the research TheySay conducted on Apple’s WWDC keynote and see Twitter’s reaction based on a collection of 94,528 tweets from 30 minutes before the event started until just after it ended.