Minnesota Patent of the Month – October 2023
Founded in 1999, Smart Information Flow Technologies, LLC (SIFT) has developed an extensive expertise in Natural Language Processing (NLP), cybersecurity, and automated planning technologies. The company works to make the information flow between humans and technology better on both sides. Recently, the company has been granted a patent for one of their machine learning approaches to structuring text which opens up new possibilities for extracting meaning and context from unstructured text.
At the heart of SIFT’s invention is a trained machine learning model that processes unstructured text and identifies entities within it. Named Entity Recognition (NER) engines, combined with cutting-edge machine learning, enable the system to recognize and embed tokens from the text into a vector space. This embedding forms the foundation for understanding and categorizing the text.
But what sets SIFT’s approach apart is its focus on determining assertions within the text. These assertions are essentially statements or judgments applied to at least one entity in the text, expressed as vectors with various dimensions. Each dimension represents a specific aspect of the assertion, often in the form of probabilities or log-odds, addressing elements such as dehumanization, moral condemnation/justification, ingroup/outgroup dynamics, violence, and harmed/responsible-for-harm relationships.
The power of this method lies in its span-level model, capable of identifying entities and simultaneously deriving assertions based on candidate spans of tokens. This seamless integration of entity recognition and assertion determination unlocks a deeper level of understanding within the text.
SIFT’s innovation is also bolstered by its contextual language model (CLM), often employing state-of-the-art models like Bidirectional Encoder Representations from Transformers (BERT). The CLM and post-CLM layers work together to decipher the contextual meaning of the unstructured text, further enhancing the system’s capabilities.
This technology has far-reaching implications. It can be applied to a wide range of domains, from information retrieval to information extraction, and beyond. Understanding the nuances of text, including moral and quality assertions, opens doors to a new level of comprehension and analysis.
Are you developing new technology for an existing application? Did you know your development work could be eligible for the R&D Tax Credit and you can receive up to 14% back on your expenses? Even if your development isn’t successful your work may still qualify for R&D credits (i.e. you don’t need to have a patent to qualify). To find out more, please contact a Swanson Reed R&D Specialist today or check out our free online eligibility test.
Who We Are:
Swanson Reed is one of the U.S.’ largest Specialist R&D tax advisory firms. We manage all facets of the R&D tax credit program, from claim preparation and audit compliance to claim disputes.
Swanson Reed regularly hosts free webinars and provides free IRS CE and CPE credits for CPAs. For more information please visit us at www.swansonreed.com/webinars or contact your usual Swanson Reed representative.