Overview
At the heart of the Attensity Text Analytics suite is Attensity's industry-leading unstructured data extraction and transformation technology. This technology extracts information previously hidden in free-form text and turns it into valuable relational data that can then be analyzed with other structured data. Fusing unstructured and structured data provides government agencies and enterprises with a complete picture of the organization.

Attensity's Data Extraction and Transformation Process

Attensity uses patented information extraction technology that is the result of over a decade of computational linguistics research. A partial parser dissects the text quickly and robustly. It easily handles even the worst text: sentence fragments, grammatical errors, misspellings, and incorrect punctuation. This is essential when your text consists of emails, text messages, transcripts of customer calls, technician notes and shorthand. Attensity handles all of these with extremely high precision and accuracy.

After text is parsed, it is run through Attensity's patented extraction engines. Attensity provides the widest spectrum of extraction engines in the market to enable customers to leverage the best and most appropriate approach for extracting data from their text. At one end of the spectrum is targeted extraction. Targeted extraction is useful when you know what you are looking for. Entity and event definitions, forms of targeted extraction, can be created to precisely target and extract only the information desired.

Implementing Text Analytics solutions can be time-consuming due to the amount of predefinition required to teach the system what facts to extract. To enable customers to quickly leverage powerful, accurate fact extraction, Attensity engineers have developed a patented solution called Exhaustive Extraction™, which requires no predefined extraction logic. This leads to a unique text analysis capability, allowing users to explore the facts of a collection of text even if they don't know what they are looking for "but will know it when they see it." This capability lies at the other end of the Text Analytics spectrum and is unique to Attensity.

Attensity technology is available through the high-availability, enterprise ready, scalable Attensity Server. Attensity Server is ideal for handling large volumes of data and heavy user loads. The technology is also available for integration via the Attensity SDK which provides access to Attensity's patented extraction engines via Java APIs, a Web Services interface and through UIMA adapters.

• Widest range of extraction technology from entity, event and keyword extraction to extractors that focus on facts, modifiers, actions, objects, etc.
• Open architecture that allows users to plug in custom converters, preprocessors and output handlers
• Industry standard interfaces including SOAP and XML
• Integration access via Java APIs

• Best of breed technology that provides accurate, actionable output
• Quickly translates mountains of free-form text into structured information, enabling users to quickly analyze more of your data in less time
• Robust parsing works despite sentence fragments, grammatical errors, and misspellings allowing users to process even poorly formed text typical in call logs, claim forms or feedback surveys
• More extraction types allow users to get the type of information needed for specific analysis requirements