Draft Programme Session: Machine learning (STS50)

December 6, 2016

Session Type: Lecture (sequence of presentations followed by questions)

Using machine learning to match assistive technology to people with disabilities

The Arc of the United States is the largest provider of services to people with disabilities in the United States.  The Arc serves 1 million people with disabilities per year through 120,000 staff in 660 locations around the USA.   In partnership with Google, The Arc is developing a machine learning-powered online tool that recommends assistive technology to people with disabilities based on their individual profiles.  The tool is fuelled by a large volume of data on: people with disabilities, the technology products they use, and the outcomes they achieve with those products.

In this session we will present on the early insights and patterns we have surfaced from our recommender system.  In addition, we will open a conversation about other ways to use machine learning to identify and deliver the other kinds of services to people with disabilities.  And, we will raise and explore the potential for replicating or expanding these kinds of systems to countries in Europe.

Lastly, in theory, a system for recommending assistive technology to people could be expanded to recommend many other interventions.  I hope to open a creative conversation about other ways to use machine learning to deliver higher quality services to people with disabilities.


Contact: Abe Rafi, Director of Digital Strategy & Online Services, The Arc, United States of America
Email: rafi@thearc.org


Feel free to use this image to promote the Special Thematic Session within your network:


The AAATE Conference will take place from Sept 13-14 2017 in Sheffield as part of the AAATE Congress. Below are some related Special Thematic Sessions that may be of interest:

Methodology and methods

Strategic and Strategies



We call upon @ansidotorg, @Standards4EU and @BSI_UK to endorse Recommendation2016 for #accessibility in #standards: https://t.co/7b3cpvfnWP


ISO TC37 already endorsed Recommendation2016 for accessibility + inclusion in standards: https://t.co/7b3cpvfnWP @EU_Consumer @IECStandards


To facilitate the uptake of #accessibility and #einclusion we need them anchored in #standards: Recommendation2016… https://t.co/JKy3QFbERw