Explainable Artificial Intelligence (XAI) For Prediction By APU Team Wins AWS Special Award

Three Data Analytics Undergraduates Got Loads Out of an Industrial Hackathon  

While talent acquisition is crucial for the industry to gain sustainable growth, industry players are getting more creative in identifying the best minds. In recent times, competitions organised by the industry are one of the innovative channels to attract top tech talents. 

Conversely, tech competitions held beyond the campus often motivate students to perform and excel by adopting innovative techniques and developing their ideas and skills. These are opportunities to gain substantial experience, showcase skills, and discover personal aptitude.

Recently, Team Outlier which was made up of three APU final year undergraduates in Computer Science specialising in Data Analytics studies made a mark at the RHB Get Your Hack On: Data Edition 2022, which was held from 21st to 23rd & 29th January.

Heng Yi Sheng, Chiah Yi Her, and Hooi Jun Ming, who competed with 15 shortlisted finalists out of 101 teams in total, are walking examples of how an extended competition organised beyond the campus could offer a lot more reward than just winning a prize. 

“We have been looking for a data-related challenge whereby we will be given datasets to solve a defined problem. Moreover, we would like to put our skills to the test and see how far we can go with our current skills and knowledge,” explained Heng, the team leader. 

As the trio came across this ‘datathon’ – similar to hackathons, it is where participants gather to solve practical problems through the application of data science tools and techniques – via social media, they decided to give it a try. 

“We have always wanted to participate in hackathon challenges to showcase the technical skills and knowledge we have learned in class. Besides, we wanted to get out of our comfort zone and grow as young professionals by pitching our ideas to the industries,” added Heng. 

As it was an open competition, Team Outlier needed to compete with working professionals to create functional algorithms, meaningful insights or exciting applications using datasets provided by the organizer to solve different challenges.

Participating on their own accord, without any mentor but solely relying on the strong knowledge of data mining, machine learning and deep learning throughout three years of studies, the trio amazed their lecturers by bringing back the Amazon Web Services (AWS) Special Award!

Looking back, Heng confessed, “To be honest, it was not easy communicating on a virtual platform. We have been doing much research on the company background and the banking products. After that, we brainstorm ideas with the consultant mentor from the bank.”

The virtual ‘datathon’ gathered students and working professionals to compete on the same track to create functional algorithms, meaningful insights, or exciting applications using existing datasets, to solve different challenges. (Photo credit to JomHack)

Breaking through obstacles along the way, the team decided to present their solution through a code editor and made it into the final round which they had been given an additional 5 days to enhance their project by deploying the solution into a web application.

“The preliminary round was a 3-day non-stop process, as we needed to prepare for the live demo of our solution that gives better customer experience in digital banking. Our team had to burn the midnight oil,” recalled Heng. 

On their winning factors, the team leader analysed, “Our team incorporated Explainable Artificial Intelligence (XAI) to explain features that contribute to the prediction. We believed the ability of the model to explain the features contributing to the prediction is highly appreciated in the industry.” 

The team summarized that they had learned much throughout this competition as it provided an opportunity for them to observe other teams’ solution and their creative ideas in tackling the competition. 

Greatly influenced by the “strive for excellence” ethos at APU, Team Outlier is grateful that this exposure had boosted their confidence and allowed the team to connect with new people. Besides, the trio also expressed their appreciation to dedicated lecturers who had shaped them. 

“I would like to express our utmost gratitude to Dr. Preethi Subramanian for introducing us the Explainable Artificial Intelligence (XAI), a new and the most recent research in the field of artificial intelligence. 

“She was my Final Year Project supervisor who is extremely knowledgeable and experienced in machine learning (ML) and artificial intelligence (AI) discipline. Her dedication, involvement, and encouragement have allowed me to conduct profound research in the field of ML and AI,” Heng expressed his gratitude.