At least 30% of people seeking support at a Citizens Advice office are repeat visitors. If we better understand the relationships between these problems, would it be possible to offer preventive advice, and thereby reduce cost and provide a better service? DataKind UK worked with Citizens Advice to find out.
DataKind UK works with charities across the country to help them harness the power of data science. We recently ran a year-long project with Citizens Advice. As you probably know, Citizens Advice offers confidential, impartial advice and support on a number of day-to-day issues ranging from claiming benefits to faulty goods. DataKind UK helped the organisation pull together and understand their disparate data sets to get a better picture of the emerging social problems that people in the UK face. Here is one small part of that project...
On average, 5,700 people walk into their local Citizens Advice Bureau every single day. Of these people, at least 30% are repeat visitors. They come in seeking advice with one problem, and return months later with a different problem. If we could better understand the relationships between these problems, could Citizens Advice offer preventive advice, and thereby reduce cost and provide a better service? That's what we sought to find out. Using Python, Scipy, Networkx, Spyre and D3.js, Billy Wong, a DataKind UK volunteer worked to tackle this problem. Come to our talk to see the result!