Research and Technology Planning with Textmining

Fraunhofer INT

Here we demonstrate some text mining and web mining applications. The Jaccard's Coefficient Comparator is a text mining tool that compares two texts using the well-known Jaccard‘s Coefficient, a standard measure in information retrieval. It is defined as the size of the intersection divided by the size of the union of the representative sets of terms of both texts. The Jaccard's Coefficient Comparator is optimized to compare texts from taxonomies e.g. to compare technologies, scientific projects etc. Therefore texts can be inserted in XML format (see example Science Citation Index - Scope Notes). All inserted texts are compared among each other.

The Context based Internet Search presents a new kind of internet searching. Firstly queries are executed by use of the web search engine Google. The results are title, abstract and link. Beside this, all further terms which occur together with the terms from the query in the abstract are presented. By choosing one of these terms the query will be modified (the term will be added to the query or an existing term in the query will be deleted).

The Web Context based Text Analyzer helps users to analyse textual information. It shows context based information from google for each text phrase selected by the user.