In recent years there has been an enormous increase in the amount of text readily available in digital formats. In order to cope with this flood of information human analysts require automatic support. However, due to the semi-structured nature of text and the impressive flexibility and complexity of natural language, the development of automatic methods for text analysis is challenging. For many tasks fully automatic processing is not possible at present. Understanding a text properly often requires world knowledge and interpretation. Furthermore, in many analysis tasks creativity and the ability to detect previously unknown patterns is needed. On the other hand, to process large amounts of text, the computational speed and storage capabilities of modern computers are indispensable. Therefore, a tight cooperation between the human and the machine is mandatory to analyze large textual datasets.
The focus of our group’s work is to develop novel interactive visualizations that are tightly coupled with automatic natural language processing methods to enable human users to explore their data. The document analysis tasks that we support cover a broad range of topics.