TheySay Reads Between The Lines From Online Content

Isn't it always fun to know what 'they' are saying? Well fun or not, it is important for an organisation to have feedback from their customers and with social networks flourishing, internet provides everyone a platform to express their opinions. A new start-up company, TheySay, works towards the goal by accurately studying what people say online using 'sentiment analysis.' TheySay was originally a software designed by Professor Stephen Pulman and Karo Moilanen from Oxford University’s Department of Computer Science and was commercialized by Isis Innovation to exploit the potential of the software.



Sentiment analysis or opinion mining is an the application of natural language processing, computational linguistics, and text analytics primarily used to identify and extract subjective information in source materials. In the context of online reading, TheySay aims to bring  linguistic intelligence along with machine learning techniques to get past the barriers of existing sentiment analysis tools. TheySay with its sentiment analysis tools will use a scoring scheme to classify linguistic structures for sentiment in a sentence and score all individual words in a text.

To carry out the process of scoring, a very large database has been annotated by hand by Professor Stephen Pulman assigning them the emotional meaning they carry. Along with that, the grammatical context in which these words appear are also taken into account to recognize if the word has a positive or negative meaning in a sentence. Using this technique, TheySay can determine emotional attitudes towards a object mentioned in the sentence. This would eventually help give businesses, public sector bodies, political organisations, and individuals an insight into the reputation they are building and monitor feedback from their customers.

Source: #-Link-Snipped-# Image Credit: #-Link-Snipped-#