On April 14, 2020, more than sixty students reported on their work on twenty semester-long data science projects advised by faculty across different USC schools as part of the CKIDSxGRIDS DataFest Spring 2020. Projects included participating in a data challenge for tracking the progression of Parkinson’s Disease, integrating data to analyze global trends in climate, understanding human behavior through online game data, characterizing cybersecurity tools, and analyzing diversity in authorship of scientific publications.
Due to the coronavirus pandemic, it was held as a virtual meeting with three parallel tracks and a final awards ceremony. The event started with a presentation of the GRIDS data science student association.
The projects of DataFest S20 were (descriptions available here):
1. A Data Challenge for Parkinson’s Disease
2. Using Biomedical Researcher Judgments to Predict Clinical Trial Outcomes
3. Towards Automated Understanding of Scientific Software
4. A Knowledge Graph for Cybersecurity Experiments
5. Connections within Contemporary Feminism Movements
6. Tracking health and nutrition signals from social media data
7. Predicting Effective Tax Rate of Publicly-Traded Firms
8. Text Analysis, Social Networks and Crowdsourcing
9. Modelling Spatiotemporal Relationships between Waste Water Injection and Induced Seismicity
10. Data scraping for salary benchmarking
11. Microtelcos and the Digital Divide in CA
12. Disparities in educational achievement
13. Gender inclusion in science
14. Annotating Paleoclimate Data
15. Modeling Uncertainty in Drought Products
16. Characterizing the counter-narratives of climate change
17. Digital Democracy: Using Social Media to Improve Political Discourse
18. Turning Library Collections into Data Science Challenges and Resources
19. Capturing the Provenance of Data Analysis Using the PROV Standard
20. Team Dynamics in Online Multiplayer Games
Awards
Co-Chairs Andres Abeliuk, Deborah Khider, and Fred Morstatter gave several awards to students for their outstanding achievements:
- Best Data Science Open and Sharing Practices: Priya Dharlama and Jiajing Wang
- “For working with the team to try different approaches, compared them, and integrated their work successfully”
- Project: SOMEF: A framework for automated Scientific Software Metadata Extraction
- Best Interdisciplinary Data Scientist: Erika Xia
- “For consistent and timely achievements in scraping, processing, and assimilating large amounts of data to curate multiple datasets from social media”
- Project: Tracking health and nutrition signals from social media data
- Best Data Science Developer: Ziping Hu
- Project: Disparities in educational achievement
- Best Project Achievement: Erica Xia, Spoorti Nidagundi, Iris Liu and Abhilash Karpurapu
- “For integrating different data sources and yielding new insights”
- Project: Tracking health and nutrition signals from social media data
- Best Data Science Collaboration Practices: Yash Shah, Swetha Thomas, Hongyu Li, Raveena Kshatriya, and Abhi Thadeshwar
- Project: Digital democracy: Creating tools to enhance political discourse
- Best Data Science Team Leader: Kevin Tram
- “For keeping the team on task and motivated beyond the original scope of the project, and presenting the resulting tool to data analysts in the USC Provost Office.”
- Project: Data scraping for university salary benchmarking
- Best Data Science Team Leader: Aditya Gupta
- “For undertaking the challenges in analyzing large and heterogeneous physics publications data spanning more than 100 years with a critical and constructive approach.”
- Project: Gender inclusion in science
- Best Data Science Teamwork: Abhilash Panduragen, Aditya Jojodia, Sushmitha Ravikumar, Vanshika Sridharan
- “For self-organizing on the project, dividing the tasks, and making sure that progress was being made each week with little to no supervision.”
- Project: Characterizing the counter-narratives of climate change
- Best Data Science Insight: Hardik Surana
- “For successful extraction and filtering of relevant objects for the knowledge graph and finding relationships among them in a novel domain.”
- Project: Creating a Knowledge Graph of Datasets, Papers and Software for Cyber-Security
- Best Interdisciplinary Data Science Team: Keerti Bhogaraju, Ian Myoungsu Choi, Negar Mokhberian, Nazanin Alipourfard
- Project: Connections within Contemporary Feminism Movements
- Best Interdisciplinary Data Science Team: Chaitra Mudradi, Feilong Wu, Hsing-Hsien Wang, Maria MacHarrie, Shubhankar Singh, Yangtao Hu
- “For working with USC libraries to understand the unique requirements of archival data and library users.”
- Project: Turning Library Collections into Data Science Challenges and Resources
- Best Data Collection: Kevin Tsang and Jiaqi Liu
- “For integrating and improving complex data to create a novel data science resource.”
- Project: Team Evolution and Dynamics in Online Multiplayer Games
- Best Data Science Mentor: Nazgol Tavavi
- Project: A Data Challenge for Parkinson’s Disease
- Best Data Science Mentor: Nina Mehrabi
- Project: Gender inclusion in science
- Best Project Presentation: Priya Dharmala, Jiajing Wang, Verdant Diwanji
- Project: Towards Automated Understanding of Scientific Software
The faculty had nominated specific students or groups of students for these awards. The Co-Chairs noted that the Best Data Science Mentor, given to students that take leadership in a project team and help mopre junior students learn and contribute, had many great nominees and was given twice this semester.
DataFest Spring 2020 is the third DataFest event held at USC. Previous events were DataFest Fall 2019 and DataFest Spring 2019.