by PG Ramachandran and Will Scott
IBM has announced the IBM AbilityLab™ Content Clarifier, which is designed to help simplify, summarize, and enhance content in order to increase comprehension for people with cognitive disabilities, the aging population, or those learning English as a second language.
Content Clarifier, an Application Programming Interface (API) built using IBM Watson services and the IBM Bluemix cloud development platform, works by leveraging natural language processing and cognitive computing to analyze and condense content into a simplified form so people have an easier time consuming and comprehending the most important concepts. It replaces complex words with easier to understand alternatives, reorders or rephrases sentences, and provides additional context about pertinent concepts, such as phonetic pronunciations, images, maps, or links to web references. It is designed to be integrated into any mobile, web-based, or desktop application to allow users to increase the understanding of content more efficiently.
IBM is also working with University of Massachusetts Medical School, University of Massachusetts Boston and Worcester Polytechnic University to further research into how technology can help simplify content for people with cognitive disabilities to make better decisions and live more independently. The results of this collaboration will add additional functionality into future versions of Content Clarifier.
Content Clarifier represents an advancement in machine learning and natural language processing by instantly taking a long and complicated body of content, filtering out unnecessary information, replacing complex and recurring words, and converting it something more understandable. It helps simplify the lexical, grammatical, or structural complexity of content while retaining the semantic meaning.
It also takes accessibility to the next level by delivering highly customized and personalized content, especially for someone with a cognitive disability, such as autism that now affects 1 in 68 children in the United States. This tool could be a way to help connect those with autism with words and speech and help them communicate and relate to the world around them.
Content Clarifier can be used across many different domains and industries, including labor. For example, according to the U.S. Equal Employment Opportunity Commission, “the majority of adults with intellectual and developmental disabilities are either unemployed or underemployed, despite their ability, desire, and willingness to work in the community.”
Today, most online job boards make their websites accessible for assistive technology users, but in general, the content is not modified for those with cognitive disabilities leading to possible barriers of employment. By using Content Clarifier, the content can be enhanced and reformatted so users can better understand the posted job descriptions, responsibilities, and qualifications.
“This solution would be extremely useful for Easterseals, which has been helping individuals with disabilities and special needs for more than 90 years,” said Tod Marvin, Chief Executive Officer of Easterseals Central Texas. “We spend a lot of time helping interpret information, especially employment opportunities, for our members. By using Content Clarifier, it could help reduce the approximately 80 percent unemployment rates of people with disabilities, by simplifying content in real time to increase comprehension and confidence, and open up a new world of independence.”
Marvin continued, “The best part is that we don’t need to buy new technology to make our information more accessible, we can use Content Clarifier to augment and transform what we already have.”
Content Clarifier can be used in many different scenarios, including helping condense news articles or videos into a format that conveys the keys concepts; summarizing meeting transcripts to convey the most important details of a meeting; simplifying domain specific semantics, such as medical and legal terms; or, as an educational tool that supplements the original content with multi-modal data (images, links to web resources) to provide additional context and insight.
About the Authors
PG Ramachandran is the Program Director the Advanced Tech team at IBM Accessibility. He has extensive experience in applied research and software development, and he holds several patents. Ram leads the research of mobile, the Internet of Things, and accessibility-related technologies and incorporates them into IBM solutions and services.
Will Scott is a Software Architect in IBM Research working with the IBM Accessibility team. His area of focus is developing solutions related to cognitive computing. Will received his Ph.D. in Engineering Science from Louisiana State University in 2003, with a research focus in machine learning and neural computing.