By AJ Avañez on May 5, 2018
Aivin Solatorio, Kalibrr’s Senior Manager for Research & Development, came in 5th out of over 2,000 entrants to the World Bank’s machine learning competition called “Pover-T Tests: Predicting Poverty” which ended last February 28, 2018. He also won the bonus prize awarded to the top participant coming from a developing country.
The competition was hosted by data science platform, Driven Data and funded by World Bank’s Knowledge for Change Program, with the aim of engaging data scientists from developing countries and applying a cost-effective solution to testing approaches to poverty prediction.
Poverty measurement and prediction is an important starting point to determine if poverty reduction policies and strategies are effective. Currently, measuring and predicting poverty is very challenging, time consuming and costly— typically done through in-depth surveys on household consumption. The competition helps World Bank see how harnessing machine learning and data science can predict poverty status based on easy-to-collect information. These models would allow World Bank to more rapidly measure the effectiveness of poverty alleviation policies and interventions to maximize the impact of these strategies.
"I was really happy when I learned that I won the bonus prize because I know that some participants there are elite Kaggle competitors. Putting the prize aside; however, coming from a developing country myself, I immediately recognized the significance of the competition and its corresponding social impact." Aivin said. "While AI technology is frequently associated with loss of jobs; this World Bank initiative shows the positive aspect of AI - providing novel solutions to help directly address global problems such as poverty." he added.
At Kalibrr, Aivin and his team work on applying artificial intelligence to help solve the problem of job matching and recruitment.