MSBA students excel at NUS Johnson & Johnson Analytics Innovation Challenge 2019
6 May 2019
![20190506_MSBA_students_NUS_Johnson_and_Johnson_Analytics_Innovation_Challenge From left: MSBA students Shivam Bansal, Aditya Gupta, Harshil Parashar, and Wu Zheyu at the NUS Johnson & Johnson Innovation Challenge award ceremony.](https://msba.nus.edu.sg/wp-content/uploads/2019/05/20190506_MSBA_students_NUS_Johnson_and_Johnson_Analytics_Innovation_Challenge.jpg)
6 May 2019 – First year NUS Master of Science in Business Analytics students Shivam Bansal, Aditya Gupta, Harshil Parashar, and Wu Zheyu, won first place at the NUS Johnson & Johnson Analytics Innovation Challenge. The competition’s final presentation and award ceremony was held in NUS Computing on 30 March 2019.
A total of 28 teams consisting of 120 participants participated in the challenge to use Artificial Intelligence (AI) and predictive analytics to generate sales performance commentaries for Johnson & Johnson. The teams worked on their solutions from 31 January to 15 March, after which eight teams were chosen to pitch their ideas to a panel of judges from Johnson & Johnson and the NUS Business Analytics Centre. The winning team, Team Anno Domini, received a cash prize of $5,000 for their solution.
“At Johnson & Johnson, sales performance insights commentaries are handwritten,” said Shivam, on behalf of his team. “It is a manual task for financial analysts and so we were challenged to use natural language processing and generation to identify products that require these commentaries and automatically generate insights based on multiple data sources.”
The team developed an AI based cross-platform solution to significantly improve the productivity and efficiency of Johnson & Johnson’s financial analysis. Using deep learning and forecasting techniques, the team’s platform is able to generate commentary and useful insights from the company’s ledger and sales data, external news, sales events and seasons. “We developed a complete solution for the team. We implemented a very strong deep learning framework to generate meaningful and insightful output. More importantly, our solution focused on providing output that was relevant for Johnson & Johnson,” said Shivam.
“The competition itself was very challenging as it required teams to have very strong technical capabilities, as well as strong business understanding,” Shivam added. “It was a great honour to win the first prize among talented peers who have done some amazing work.”