—Information about individuals on publicly available web sites stands as a valuable, yet unorganized, data source. Turning such an enormous data source into a “database” is highly desirable as it has the potential to lead to novel ways of using the available information to the largest extent. In this paper, we present PopulusLog, a novel web data mining system. PopulusLog is a pioneering example of next generation search engines which produces and provides access to non-intuitive knowledge on the web. It involves a framework for tools that collect, extract, mine, query, browse, and visualize information about anonymous people. Keywords— information extraction; search engines; machine learning; web databases; entity tagging.