Omdena Chapter Page: Bulgaria

Chapter Lead – Monika Manolova

 

THE BACKGROUND

This challenge will be a collaboration between Omdena Bulgaria Local Chapter and LiberAid.

LiberAid is a community of people with disabilities that is focused on the mission of helping people with physical disabilities travel more freely and enjoy their experience. 

According to the Wheelchairfoundation, WHO and Euobserver:

• About 1% of people worldwide have rheumatoid arthritis
• In the 34 developed countries it is estimated that 1% or 10,000,000 people require a wheelchair
• In the 156 developing countries it is estimated that at least 2% or 121,800,000 people require a wheelchair
• Overall, of the 7,091,500,000 people in the world, approximately 131,800,000 or 1.85% require a wheelchair
• An estimated further 110 million people worldwide need wheelchairs but don’t have them – 500,000 in developed countries and 109.5 million in developing countries.
• In the European Union, wheelchair users comprise around 1% of the population, or 5 million people.
• Every year, about 500,000 people worldwide suffer a spinal cord injury. The main causes of injuries are traffic accidents, violence, and falls.

 

People with disabilities, require constant support to visit sites, access public amenities and enjoy tourism. There are everyday challenges that concern transport modes, connectivity, origin, and destination and building accessibility, that are hard to manage regardless of the regional context.  Rather than a source of agitation travel should be an enjoyable experience even for people at a physical disadvantage.  

 

THE PROBLEM

Through machine learning, we can predict and map accessibility for people with disabilities. We can help enable and enhance the experience of travellers that are faced with physical challenges, and we can improve their day-to-day experience and quality of life. 

There are various data resources, which may be applied to an accessibility ML solution aimed accessibility for the disabled. There is mapping of the safest routes and route modelling, but there is also accessibility valuation of sites, sidewalks assessment and localisation of support. The ML solution should consider the specific needs of the end user and provide user-friendly analytics. 

Various modes of transport and public sites should be analysed from the perspective of people with physical disabilities. The ML algorithm should consider the various valuation elements and index accessibility. 

 

The outcome of the challenge should be an API of an ML model which predicts and accessibility for people with disabilities within an urban context. It can be initially modelled around the city of Sofia Bulgaria, but adjustable to alternative urban contexts.   

 

THE PROJECT GOALS

1. Gather information about accessibility to buildings and public sites in Sofia.
2. Create an ML model which can predict accessibility difficulties for the end user.
3. Develop a user-friendly dashboard which can showcase the solution.

 

THE LEARNING OUTCOMES

1. Data Collection
2. Data Pre-processing
3. Exploratory-Data Analysis
4. Modelling
5. Model Deployment into possible API
6. Visualisation and Publication

 

THE TASKS & TIMELINE

 

Week 1	Week 2	Week 3	Week 4	Week 5
Data Collection (pre-week 1 even) Data Pre-Processing	Data Pre-Processing	Exploratory Data Analysis Modelling	Modelling (cont)	Possible deployment into API
Week 6 Visualisation and publication	Week 7 Visualisation  and publication (cont)	Week 8 Visualisation and publication (cont)	Week 9 Wrap up	Week 10 Wrap Up