The R&D Tax Credit Aspects of Natural Language Processing (NLP) Innovation



By , , and


Natural-Language-Processing
        What if Siri could be a more personal and a more personable digital assistant?

        Personal assistants that are more personable are just one target technology to be gained from advancements in natural language processing. In the future, applications like Siri will need to process a lot of data and communications in natural language, such as news articles, reports, academic papers, and speech. The issue of how to create machines that can process natural language has lead to innovations and the creation of an emerging research front that may qualify for federal and state R&D tax credits.      


The Research & Development Tax Credit

        Enacted in 1981, the Federal Research and Development (R&D) Tax Credit allows a credit of up to 13 percent of eligible spending for new and improved products and processes. Qualified research must meet the following four criteria:

  • New or improved products, processes, or software
  • Technological in nature
  • Elimination of uncertainty
  • Process of experimentation

        Eligible costs include employee wages, cost of supplies, cost of testing, contract research expenses, and costs associated with developing a patent. On December 19, 2014, President Obama signed the bill extending the R&D Tax Credit for the 2014 tax year.


Natural Language Processing (NLP)

        Natural language is the language we use as humans, our own systems of language. The term natural language is used when relating to machine language, which is structured in a way that makes it easy for modern computers to parse and compute.

        Despite the structured and functional nature of written and spoken languages, processing our own languages so a computer can derive data from them is not an easy task. Languages have evolving needs for meaning, context, and ambiguity that don't lend themselves to computation by a machine. We fashion our language to meet our needs, whereas machines require static, unchanging rules, sequences, and substance for fast and correct computation.

        In order to solve the problem of processing the written and spoken word, we would need to create algorithms and create bodies of lexical understanding in a language that allow us to take samples in natural language and translate it to machine language. The challenges of this task include functionally breaking a written or spoken sentence up, correctly identifying the meaning, function, and context of the words, and resolving any semantic, contextual, or ambiguous meanings that exist in any words or throughout the sentence - all in each language and in widely varying media styles and formats.

        Topics in natural language processing are sophisticated in nature and stride several disciplines in language and computational science. Semantic parsing and dropout learning, for example, are just two techniques for analyzing or sorting strings of symbols and understanding any ambiguity that exists. Feature noising is a computational technique involving processing specifically featured sets of data. Textual entailment involves evaluating the implication of a statement's meaning.


Commercial NLP Innovations


Digital Reasoning Systems, Inc. - Nashville, TN
        Digital Reasoning combines NLP with cognitive computing techniques to create a deep insight generating analytics platform. Digital Reasoning's systems take information in natural language and create insights through analysis.  Analytical engines like this can bring about insights that could result in savings by reducing inefficiencies or waste and revealing criminal activity and abnormalities.     
 

FuseMachines, Inc. - New York, NY
        FuseMachines creates automated solutions to enhance customer service quality and improve call time. FuseMachines makes use of machine learning and NLP to create learning automated CS platforms to get information, process it, and interact with customers. The company has over 60 people and has a body of research in NLP and machine learning techniques.  


Palantir Technologies, Inc. - Palo Alto, CA
        Palantir Technologies is a company that uses NLP for a deep analytics platform. Instead of automated analysis, Palantir creates platforms and tools to enable and strengthen human analysis. Palantir builds platforms for integrating, managing, and securing data for human-driven, machine-assisted analysis.   
  

Microsoft, Inc. - Redmond, WA
        Microsoft's natural language process group currently has several projects in NLP, including machine translation, ESL Assistants, textual entailment, paraphrasing, a Japanese NLP Project, and Mindnet - a map-building semantics parser. Translating and processing text and speech between English and Japanese is a major focus of Microsoft's NLP research efforts.  


IBM, Inc. - Armonk, NY
        IBM's NLP research group has a long history with NLP projects, and currently engages in several areas of research of the topic. IBM's recent research continues to improve its systems and methods for conducting NLP. IBM's Watson is a premier example of applications of their research - Watson makes use of NLP to gather, process, output, and receive feedback on data as part of its cognitive computing methods in oncology and scientific analysis.     


University NLP Research


Stony Brook University - Stony Brook, NY
        Stony Brook University has it’s own natural language processing lab. According to its website, Stony Brook’s NLP lab specializes in "Statistical approaches and computational models for Natural Language Processing and Computational Linguistics. Interdisciplinary research connecting Natural Language Processing with Computer Vision, Psychology, and Cognitive Science.".  


University of California Berkeley - Berkeley, CA
        UC Berkeley's NLP group focuses on machine translation, computational linguistics, grounded semantics, and unsupervised learning. The group has links to demonstrations of some of the products of its research. Berkeley also uses NLP to translate, decipher, and conduct linguistic analyses on ancient languages and texts. These avenues of research may help humans more fully reconstruct or save forgotten or incomplete samples of languages and transcribe ancient texts.
 

University of Maryland at College Park - College Park, MD
        University of Maryland at College Park’s NLP Natural lab is part of the University of Maryland Institute for Advanced Computer Studies (UMIACS). The lab's work currently includes speech recognition, handwriting and optical character recognition, machine translation as well as language data exploitation applications including document summarization, sense-making across structured data such as ontologies and thesauri, information retrieval, ranking and personalization, and computational social science.  


Columbia University - New York, NY
        Columbia University's NLP group works with Columbia's Center for Computational Learning Systems and the Biomedical Informatics Department.  It has seven senior researchers that serve as principal investigators on multiple projects. Columbia's department works in written and spoken language fields similar to other major research universities, but also has somewhat more unique projects in Arabic NLP.


Cornell University - Ithaca, NY
        Cornell University's natural language processing center works on a diverse set of projects in NLP, computational linguistics, machine learning, and information retrieval.  Cornell has over 12 courses at the university to support learning in this area. Like many other NLP research groups, Cornell's NLP group collaborates with the other labs and departments including computational linguistics, machine learning, artificial intelligence, the cognitive science program, the computer science department, information science department, linguistics department and psychology department.  


Stanford University - Stanford, CA
        The Stanford University NLP center consists of about 28 people and conducts research in a broad set of topics including information extraction, semantic parsing, text-to-3D scene generation, sentiment and social meaning, dropout learning and feature noising, deep learning and NLP, machine translation, dialog and speech processing and other topics. Stanford also supports languages other than English in its projects, including NLP projects in Chinese, German, and Arabic.


University of Texas at Austin - Austin, TX
        The University of Texas at Austin has a Natural Language Programming Lab that operates under the university's artificial intelligence group. It specializes in connecting language and perception, information extraction, natural language learning, and cognitive processing, text categorization and clustering and text data mining.  


Conclusion

        Natural language processing research is a continually developing field that is creating and improving applications that help our daily lives. As globalization continues on and telecommunications technologies bind us closer together, our systems and tools will need to assist us in overcoming language barriers. As NLP systems become more sophisticated and accurate, they will enable us to communicate more effectively with each other, as well as give us better understandings of our own languages and societies. R&D activities in the NLP field are greatly increasing and this rapidly growing technology area may qualify for R&D tax credits.


Article Citation List

   


Authors

Charles R Goulding Attorney/CPA, is the President of R&D Tax Savers.

Charles G Goulding is the Manager of R&D Tax Savers.

Adam Starsiak is a Tax Analyst with R&D Tax Savers.


Similar Articles
The R&D Tax Credit Aspects of Blockchain for Supply Chains
The R&D Tax Credit Aspects of Physical Security Technology
The R&D Tax Credit Aspects of Driverless Cars
The R&D Tax Credit Aspects of SaaS Start-Ups
The R&D Tax Credit Aspects of Emotion-Recognition Technology
The R&D Tax Credit Aspects of AI in the Insurance Industry
The R&D Tax Credit Aspects of Emerging AV Trends
Enhanced R&D Tax Credits for Specialized Co-Shared Spaces
Ethereum's Impact on Digital Contracting Creates R&D Tax Credit Opportunities
The R&D Tax Credit Aspects of Geofencing
The R&D Tax Credit Aspects of Distribution Center Automation
The R&D Tax Credit Aspects of Law Firm Artificial Intelligence
The R&D Tax Credit Aspects of Avionics
The R&D Tax Credit Aspects of Telemedicine
State & Local Tax Jurisdictions Address Media Streaming and Cloud Services
Federal Government Provides Faster Approvals and Tax Credits for Consumer FinTech Products
The R&D Tax Credit Aspects of Voice-Activated Software
The R&D Tax Credit Aspects of Artificially Intelligent Hedge Funds
The R&D Tax Credit Aspects of LiDAR
The R&D Tax Credit Aspects of Educational Technology (EdTech)
The R&D Tax Credit Aspects of Cyber Security Start-Ups
The R&D Tax Credit Aspects of Construction Industry IoT
R&D Tax Credits Provide New Opportunities for Artificial Intelligence Start-Ups
The R&D Tax Credit Aspects of NYC Start-Ups
The R&D Tax Credit Aspects of Virtual Reality Technology
The R&D Tax Credit Aspects of Water Analytics
The R&D Tax Aspects of Artificial Intelligence Robo-Advisors
The R&D Tax Credit Aspects of Video Compression Technology
The R&D Tax Credit Aspects of Automated Coding
The R&D Tax Credit Aspects of Payment Technology
The R&D Tax Credit Aspects of Restaurant Technology
R&D Tax Credits and the Second Wave of Cloud Adoption
The R&D Tax Aspects of Data Storage Startups
The R&D Tax Credit Aspects of Cyber Security
The R&D Tax Credit Aspects of Mobile Applications
R&D Tax Credits for the Modern Insurance Industry
The+R%26D+Tax+Credit+Aspects+of+the+Internet+of+DNA
The R&D Tax Credit Aspects of Modern Dental Labs
The R&D Tax Credit Aspects of IoT Communication
The R&D Tax Credit Aspects of Bitcoin and Blockchain Technology
The R&D Tax Aspects of Near Field Communication
The R&D Tax Aspects of the New FDA Mobile Apps Requirements
Tapping the Power of Big Data and R&D Tax Credits for Utility Companies
The R&D Tax Credit Aspects of the Medical Software Industry
The R&D Tax Aspects of Computer Enabled Human Identification
The R&D Tax Credit Aspects of New York City's Engineering Education and Googlization
The R&D Tax Credit Aspects of Software Modeling Analytics
The R&D Tax Credit Aspects of Logistics Software and Process Innovation
The R&D Tax Aspects of Cameras of the 21st Century
The R&D Tax Credit Aspects of Network Security
R&D Tax Aspects of DNA Identification
R&D Tax Credit Aspects of Cyber Security and Homeland Protection
Financial Product Trading Platform Artificial Intelligence R&D Tax Credits
Gig City Startups and R&D Tax Credits
The Internet of Things Creates R&D Tax Credit Opportunity
The R&D Tax Credit Aspects of Mobile Banking Applications
The R&D Tax Credit Aspects of In-Image Advertising
R&D Tax Credits for Hybrid Call Centers - Airline, Hotel, and Car Rental Industries
The R&D Tax Aspects of Advertising Science
The R&D Tax Aspects of Data Science
R&D Tax Aspects of Radio Frequency Identification
The R&D Tax Aspects of Advanced Driver Assist Systems
The R&D Tax Aspects of the Internet of Residential Things
The R&D Tax Aspects of Web Television
R&D Tax Credit Aspects of Medical Robotics
R&D Tax Credit Aspects of Industrial Robotics
R&D Tax Credit Aspects of Service Robotics
Yes Alice, Patents and R&D Tax Credits Remain Available for the Internet of Things!
How Salesmen Can Use R&D Tax Credits to Sell Today's Software Products Engagements
The R&D Tax Aspects of Cloud Computing
The R&D Tax Credit Aspects of Hybrid Call Centers for Health Insurers
Now Every Business is a Software Business
The R&D Tax Aspects of Robot Software
The R&D Tax Aspects of Machine-to-Machine (M2M) Innovation
The R&D Tax Aspects of Financial Technology Services
Beacons Create R&D Tax Credit Opportunity
Fast Growth of Sharing Economy Impacts Tax Reporting
The R&D Tax Credit Aspects of Retail Technology
The R&D Tax Credit Aspects of Improving Virtual Reality Technology
The R&D Tax Credit Opportunities for Mobile Devices
The R&D Tax Credit Aspects of Wearable Technology
The R&D Tax Aspects of Big Data
R&D Tax Credit Fundamentals
Los Angeles Tech Boom Creates Large R&D Tax Incentive Opportunities
The R&D Tax Aspects of Software Development