EXTRACONN: Extraction and Analysis of Company Networks from News

Companies in nowaday’s business world have started to recognize the value of Social Network Analysis not only for the discovery of interhuman relations, but also for the examination of e.g. interorganizational dependencies and collaboration patterns. However, the corresponding IT-supported analyses require a large amount of explicitly structured data, which is rarely available and costly to create manually. The contribution of our presented approach to the above described problem is therefore two-fold. The first challenge is to bridge the gap between unstructered text data and the required network structures for the specific domain of business news. The aim is to obtain a network of companies that represents relevant facts (like industry relatedness) as close to reality as possible. We will then apply a chosen catalogue of social network analysis techniques in order to validate their suitability and to generate immediate feedback for the network generation step.