Dr. Uyi Stewart, a US-based Nigerian leading data scientist and Chief Data Scientist, IBM, has urged Nigeria to invest more in data science rather than infrastructure in order to achieve development.
He stated this at a workshop for Data Science entrepreneurs in Lagos, according to him, ‘We have a lot of challenges as a nation whether is in public sector, healthcare, agriculture, financial inclusion or education across every walk of life.
Even in Lagos, look at mobility, our ability to travel from one place to the other, we need to star to make smart decisions.
Farmer need to make smart decisions in other to farm better, worker need to make smart decisions in other to be able to reach at work on time. Even parents need to make smart decisions about their finances so that they do not just stay at the same level.
“Across all these dimensions that I have talked about, the key is really the ability to have the right data, run the right kind of insight of the data, generate the right kind of decision support and then let the humans make the decision. Even in government, government also needs to be supported in order to make the right kind of decision for the population and that can only come from data. We can’t formulate policy on emotions; we have to formulate policies based on data. So across every work or life data is valuable but the key is our ability to mine and analyse the data,” he said.
On the challenge of access data, he said that he key is the time to value. “Let’s take a scenario, I have data, I am company X and you come to me and you say ‘Give me your data’. I am going to ask you, what’s in it for me? I believe people are willing to share their data. But, innovation has to be two sides; it has to be business model and technology. We always go to them and say “Give me you data, I want to build this App” without laying on the table what’s in it for them. If you can show what business model underlying what you are doing and you can show value for the person you are asking data from, you will get the data.”
Stewart further explained that the issue of data credibility borders on data quality. “Actually they are two sides – data quality and data provenance. Provenance means when I get the data, it can percolate all the way and be consistent without somebody changing it. These are issues that machine learning thankfully allows us to address. Even if you say, “I don’t want to go into machine learning”, there is something that is called crowdsourcing which is also an approach to do data quality. It’s inexpensive. What that does is that it brings people together to validate an imprint in a democratic way. All of us cannot agree to lie. So, we need to focus on instruments either through technology or simple innovative ideas like crowdsourcing to ensure we have the right kind of data quality. It is central otherwise garbage in, garbage out.”