NUS Champions Group Project Team 7UP

Inspiration

From watching Sci-fi movies like Terminators to YouTube videos of intelligent people building a machine-learning model from scratch to defeat a game, our team was initially fascinated by the brilliance of such engineering and eventually took the leap of faith to embark on this route of hoping to become a machine learning engineer one day. This competition provides the perfect platform to hone and test our skills, as well as give us a taste of what the real-world applications of machine learning are like.

What it does

The machine-learning model we built aims to predict the future domestic sales of a company based on its meticulous analysis of historical data and advanced predictive algorithms to empower businesses to forecast sales trends accurately.

How we built it

We started by first cleaning the data by removing rows with missing values and removing columns that we deemed unnecessary in building the model through a series of EDA. We then use the clean dataset to select and train the model via XGBoost , a scalable, distributed gradient-boosted decision tree (GBDT) machine learning library, to predict the domestic sales of the company.

Challenges we ran into

As our group has almost no professional experience in machine learning and the cleaning of dataset, we have to start with very little to no prior knowledge in data analysis. However, in the 3 days of intensive group collaboration, we managed to self-learn the basics of the machine learning algorithm and put together a model.

Accomplishments that we're proud of

We managed to meet the deadline of the competition despite the heavy workload of the university modules we are currently taking. As machine learning is fairly new to us, especially when working with a real-life and complex dataset, we persevered and managed to build a machine-learning model that can accurately predict the future sales of the company with no guidance.

What we learned

Through this project, we learned the importance of teamwork, especially in a professional setting. The workload of the project is too large to be completed solo and hence justifies the need for a team with good chemistry. Additionally, we gained a handful of experience with libraries which were not yet taught in NUS, which enabled us to gain greater insight into the role of a data analyst and further deepen our interest in computing,

What's next for NUS Champions Group Project Team 7UP

If admitted to the finals, we will definitely give our best in the pitching of our project to the judges. In the meantime, our group will continue to deepen our expertise in data analysis outside the school curriculum.