Agenda:
* Introduction to the hackathon competition
* Explanation of the problem statement and objectives
* Overview of the competition interface and rules
* Discussion on data analysis tasks and evaluation metrics
* Clarification on collaboration and plagiarism guidelines
* Instructions on team formation and submission process
* Demonstration of notebook creation and submission
* Q&A session
Summary:
The meeting began with Anirudh introducing the hackathon competition and thanking the
participants for joining. He explained that the competition is a Kaggle competition for the data
analytics course and its focus is on user retention in a streaming service. The objective is to
analyze the provided dataset and derive insights that can help the streaming service retain
users.
Anirudh then provided an overview of the competition interface, which includes sections for
overview, data, code, rules, and team. He emphasized the importance of exploring the data and
going beyond the given tasks to find interesting insights. Participants were encouraged to use
various sources and tools, such as ChatGPT and Claude, but to ensure their approach is
explainable.
The competition is set to begin at 10:00 AM the next day and participants are expected to
complete the tasks by 1:00 PM. However, they can continue working on their submissions until
the end of the day. The tasks are divided into five parts, with different marks assigned to each
task. The evaluation metric is R-squared, but participants are encouraged to explore other
metrics and provide their reasoning.
Anirudh explained the deliverables, which include a submission file and a notebook. The
submission file should contain the predicted user retention for each movie, while the notebook
should document the analysis process and include explanations, code, and visualizations.
Participants were advised to follow the naming convention for their team and submission files.
The meeting then shifted to the rules and guidelines. Plagiarism was addressed, with
participants advised to mention any collaboration or borrowed code in their notebooks. The
importance of interpretability and documentation in the notebooks was emphasized.
Anirudh and Vibhav demonstrated the process of team formation and notebook creation. They
explained how to add team members, share edit access, and submit the notebook. Participants
were also shown an alternative method of submitting the CSV file directly.
The meeting concluded with a Q&A session, where participants asked questions about
collaboration, test cases, language constraints, and submission limits. Anirudh and Vibhav
provided detailed responses and clarified any remaining doubts.
Overall, the meeting provided participants with a clear understanding of the hackathon
competition, its objectives, rules, and submission process. They were encouraged to explore the
data, use various tools, and focus on interpretability in their analysis. The meeting ended on a
positive note, with participants ready to begin the competition and showcase their data analytics
skills.
