Hilbert Technology Inc. Releases QuantumText14 – A New Document Analysis and Ranking System Performing Orders of Magnitude Faster than Latent Semantic Indexing

Share Article

Advanced Indexing System and Software Architecture deliver New Capabilities for Intelligence, Publishing and eDiscovery Providers

News Image
QuantumText14 will make a significant contribution to enable organizations to truly understand and apply all the intellectual property in their possession to strengthen their Return on Information.

Hilbert Technology Inc., a privately held innovative software technology firm, headquartered in New York City, announced today the release of QuantumText14 in response to the growing worldwide need to find, extract and apply new knowledge from large volumes of textual data.

QuantumText14 utilizes Hilbert Technology’s unique and patented approach of storing and organizing data to enable the rapid identification of the most relevant documents in large volumes of textual data in minutes, rather than the days or weeks required by the closest alternative systems.

With QuantumText14, an analyst can quickly cull the most relevant documents from a massive corpus of information within a time-limited window of analysis and return the most relevant documents with precision and recall faster than any known alternative. The culling or winnowing down process does not rely on key word or phrase search. Descriptive paragraphs, or documents in their entirety, can be specified as criteria against which every other document in a specified corpus is ranked in terms of content similarity and closeness.

Here’s how it works: An analyst either creates or selects a document that he or she wants to use as an “exemplar” (e.g., the ranking criteria.) Then, the full text of that exemplar document — at the word level — is scrutinized and compared against every other document in the entire large corpus of documents — also at the word level, resulting in a prioritized ranking with scoring commensurate with each document’s content relevance and correlation to the exemplar. The process is driven by Hilbert Technology’s core processing engine, the Hilbert Engine, and an advanced indexing system that, working together, return results with precision and recall at very high speed without requiring any advanced data tuning.

Bruce J. Crowley, Hilbert Technology Chairman and Chief Executive, cited a metric for comparison: “Recently, the company applied QuantumText14’s advanced indexing system to 3.5 million patents from the United States Patent Office Database — all issued patents in digital format from January 2012 dating back to 1974. The entire content of one patent, including a summary of invention, body and all claims, was set as the exemplar document. In 15 minutes, QuantumText14 returned a prioritized and scored ranking of all 3.5 million patents in terms of content closeness and relevance to the selected exemplar. In addition, the analysis implemented used a very small processing footprint: three mid-level desktop computers with 8GB of RAM accessing 7200 RPM hard drives without a RAID option.” Crowley added, “The requirement to find and extract new knowledge from large volumes of text data remains one of the most significant challenges facing the corporate, legal, financial and intelligence communities today, particularly with the heightened sense of urgency for enhancing enterprise knowledge management. QuantumText14 will make a significant contribution to enable organizations to truly understand and apply all of the intellectual property in their possession to strengthen their Return on Information.”

QuantumText14 can be applied to text in all natural languages — alphabet and logogram based languages, such as East Asian. The system can operate as a standalone or be integrated as a subsystem in large enterprise eDiscovery offerings encompassing customer service, technical support and forensic analysis.

About Hilbert Technology Inc.
Hilbert Technology Inc., headquartered in New York, NY, with an office in Doylestown, PA, applies its patented software processes and technology to provide its customers with ultra-fast data processing while requiring a comparative fraction of the hardware required by traditional databases and data processing systems. Recently, the company announced a contract with the U.S. Army’s CERDEC Command for developing and implementing a Context Independent Anomaly Detection System as part of the Army’s Joint Unified Multi-Capable Protection System, JUMPS. In addition, the Hilbert system has been delivering unparalleled results for clients ranging from NASA, to the United Kingdom’s Department for Work and Pensions, to leaders in the oil and gas industries. http://www.hilberttechnology.com

Share article on social media or email:

View article via:

Pdf Print

Contact Author

Julie Gilbert
Visit website