In the context of the Crossover project, I’m currently reviewing the state of the art in opinion mining and sentiment analysis.

In recent years, there’s been an explosion of new products, in particular in the field of social media analytics. I came across several of them, which I gathered in our Diigo group.

They all tend to provide information on What is being said; by Whom, where and when.

They all perform similar cycle: gather the raw data; transform it and process it; and deliver the final results.

However, most of them are applied in the private sectors, and therefore tend to focus on sentiment analysis, in order to detect how people feel about a product or service.

On the other hand, Discovertext.com is more oriented towards government; it is more complex and enables users to perform itereatively all sorts of analysis, but therefore require more analytic skills; it can be used not only to process sentiments but actually to analyze ideas. In fact, it comes out of a research programme specifically designed to help processing the public comments to regulations.

Do you know any great application of opinion mining in government?