Informational queries are well addressed by web search engines only when there is some (updated) web document that aggregates all relevant information in a meaningful, structured way.
There are, however, a substantial proportion of information needs that cannot be answered with a few links, either because the relevant information has not previously been manually compiled, or because the relevant information has to be mined or aggregated from multiple sources (such as opinions about something).
Our long-term vision -which we call holopedia or encyclopedia of everything- is that of a web engine able to answer any information need with the equivalent of an automatic Wikipedia entry. In this project we focus on the most relevant named entities -people and organizations- as a first step towards the full realization of the holopedia vision.
The main goal of the project is, therefore, to develop algorithms, techniques and systems able to mine and aggregate information relative to people and organizations from unstructured and structured web sources, such as social networks, blogs, news, semantic web data, and websites in general.