Text Mining Infrastructure
The purpose of TMI is to be a general purpose yet robust framework to support high-end highly scalable textual data mining. To make this possible we provide support for all stages of Textual Data Mining from the raw data retrieval to the advanced modeling of that data. To achieve this we have broken high level concepts down into abstract components. The article below explains how we achieved a level of abstraction sufficient to maintain extensibility, while still maintaining the performance necessary to handle the computational requirements of large text mining applications. Our framework is designed to be user friendly so that it can be targeted to researchers interested in evaluating hypotheses and confirming theories without the hassle of creating their own set of tools from scratch. It is also designed so that it can also be used by developers of high-end applications as a robust software 'core'.
We want TMI to be the new standard in Textual Data Mining and we want your support. To achieve this we have opened our source to the public. Please see the Source link on the menu bar for downloads.
The Need For a Standard In Textual Data Mining
There is a need in the Textual Data Mining Field to create a standard where both research and application can be developed. This will allow for great and faster advancements to take place as less time will be spent building tools and more time can be spent solving real world problems and discovering new techniques. It is our mission to provide this standard in an open form so that it can be shaped to fix the growing demands of high performance Textual Data Mining software.