2018 Dragon Datathon

Drexel LeBow’s Business Analytics Solutions Center is proud to partner with SEI Investments Company to present the 2018 Dragon Datathon for graduate students.

What is the Dragon Datathon?

Drexel LeBow’s Dragon Datathon is a high energy, one-day event that brings together teams of graduate students to turn data into actionable insight. During this event, student teams work on analytics challenges using open-source datasets.

In preparation for the datathon, graduate students will form teams of 5-7 individuals. For the duration of the event, students have the opportunity to meticulously turn the business problem into a series of statistical questions, design a technical approach, and use data analysis and data technology skills to create solutions that solve the problem at hand.

At the conclusion of the event, teams present their projects to a panel of industry and faculty judges, outlining the data-driven insights for their business problem, the scope of their analysis, technical challenges, overall solutions, implementation plan, and expected impact.

Students use their own laptops and are given access to a computer lab with software applications such as SAS, R, and Python. Dr. Chuanren Liu, Assistant Professor, Decision Sciences and MIS, will serve as the faculty advisor overseeing the challenge.

Meals and refreshments will be provided to participants.

Participant Eligibility

To be eligible to participate, team members must be currently-enrolled graduate students proficient in core techniques for business analytics, such as data mining, text mining, machine learning, and statistical modeling.

Datathon Challenge

SEI receives hundreds of applications and inquiries regarding its summer internship program each year. The challenge is to build a chatbot that will answer the most frequently asked questions during the recruitment process for the summer internship program. The user will be able to inquire with the chatbot about topics such as how to apply to the program, information regarding housing, location and travel, or how a typical day looks at SEI.

Guidelines

• Participants may use any programming language.
• An Excel file with sample questions will be provided, as well as expected answers that the chatbot will offer.
• Participants need to enrich the data with sentences about greetings and farewells.
• The core of the chatbot is an engine that can classify user intent based on questions the user provides. Participants should use machine learning algorithms to build this engine.

Judging Criteria

Teams will be judged based on the criteria below, and deliverables will be weighted accordingly.

• 30% - Implementation of machine learning model
• 40% - Accuracy and completeness of the engine
• 20% - Quality of data processing
• 5% - Potential for further improvement
• 5% - Final presentation of solution

Prizes

Students in the winning team will receive technology-related prizes from SEI and the opportunity to present to SEI leadership on-site at the company, along with a tour of the SEI campus and data center.

Registration

To participate in the Datathon, all students must register by Friday, February 9 on the Drexel LeBow events calendar at lebow.drexel.edu/events.

Draft of Event Timeline

Timeline is tentative and subject to change.

Friday, February 16, 2018 8:30 am: Team check-in; breakfast available. All team members must be present and checked in prior to the event kick-off at 9 am. Individuals not present for check-in will forfeit their ability to participate. 9:00 am: Event begins; data provided to teams 12:00 pm: Working lunch 6:00 pm: Team presentations and dinner