TEXTREND: Development of a business and governmental decision support toolbox using trend- and text-analysis tools

The two interconnected objectives of the TEXTrend project are (1) the creation of an integrated TEXTrend toolkit and service basis, and (2) the elaboration of demonstrative applications in varied fields of business and governmental decisions, exemplified by use cases.

The goal of TEXTrend platform is the in-depth analysis of dynamic information on the web for science or business. In contrast to the usual extracting or indexing services, this toolkit aims at retrieving an essentially new type of information from the data volumes accessible on the web. An example is trend analysis, the aim of which is to determine or predict the characteristics of public opinion and its changes with time on the basis of massive amounts of documents (blogs, articles, etc.). Another example is the exploration of latent social structures by automatic text processing, network algorithms and visualisation, such as the extraction of a power-and-impact network of a field.

As of December 1. 2009, the TEXTrend toolkit exists in version v0.6, free to download and use. TEXTrend integrates UIMA (for preprocessing and NLP), WEKA(classification), R (statistical analysis), Cytoscape (network visualization), Cfinder(clique analysis) and several other components. For a documentation chech back later or consult us.

TEXTrend is based on an OSGi approach, using the CIShell framework developed by Borner's team at the Cyberinfrastructure for Network Science Center, Indiana
University, USA.

The current version of the TEXTrend toolkit (TTv.0.6) as well as its separate (OSGi compatible) modules (bundles) can be downloaded form the Downloads Section of this home page.