“Treparel’s KMX’s visualization capabilities around its autocategorization and clustering offer immediate insight into unstructured data sets and appear to be adaptable and customizable to customer needs. Its approach to autocategorization utilizes statistical principles and machine learning that require significantly less training and tuning on the part of customers than other approaches.” IDC, David Schubmehl
According to IDC, search and text analytics make computers more accessible and interactive. As core technologies for processing the unstructured information that dominates the Internet, they are part of the next wave in computing.
The relevance of search continues to grow with the explosion in digital information and the needs of businesses, consumers, governments, and other stakeholders to find, organize, navigate, publish, and make sense of it.
Search and text analytics technologies continue to spread tentacles of functionality into any applications that require language understanding: in enterprise applications, in consumer Web businesses, and in online social environments. As a result of the growing importance of these technologies, we see more intention by businesses to invest in search-based software and also in devoting more resources to supporting it.
Search and text analytics make computers more accessible and interactive. As core technologies for processing the unstructured information that dominates the Internet, they are part of the next wave in computing. (Source: IDC 2011, Susan Feldman & Hadley Reynolds).
Text analytics is difficult because language is complex, and deriving meaning from text in an automated way is complicated. To build an optimal application portfolio for solving specific text-oriented business problems IT professionals must first understand the high-level technical architecture of a text analytics system.
(…) It is easy to be overwhelmed with the highly specialized jargon of text analytics and to be confused by the vendor marketing material that often mixes technical processing terminology with descriptions of functional capabilities and combines that with the specific application focus and value proposition they are promoting. However, underneath each text analytics system is a basic pipeline approach to acquiring text, processing and analyzing text and creating output for display and additional analysis. In “Text Analytics Guidance: Building a Text Analytics Program,” Gartner established a high-level overview of the input/processing/output model for a text analytics system. Gartner 2010, Jamie Popkin.
KMX (red components) in the High-Level Architecture of a Text Analytics
We can strongly recommend further reading:
