MIDS Spring 2016 Capstone Presentations
On May 2, the spring 2016 datascience@berkeley graduates presented their capstone projects in a public webinar. For the capstone project, the graduates were tasked to solve a real-world situation or problem utilizing their data science skills in communication, problem-solving, influence, and management to provide a fully realized solution. Some groups tackled modern safety concerns with projects like SmartCam, a surveillance camera optimization application, or SafeRoad, designed to prevent all fatalities due to traffic collisions. Others focused on streamlining the kitchen. The Naïve Baker helps plan meals using pantry items and personal preferences, while Predicting Demand of Beer Styles will revolutionize the brewing industry by providing insight on the demand for select beers in a given region. Below we’ve compiled the full list of the capstones, short descriptions of each, and links to learn more about the projects.
This year also saw the introduction of the Hal R. Varian Capstone Award. Named after Hal R. Varian, founding dean of UC Berkeley’s I School and current Chief Economist at Google, the award was given to project groups based on an exceptional demonstration of the skills and tools achieved at the culmination of the MIDS program.
The Winning Projects for the Spring 2016 Presentation
First Place
- Tomographica is an artificial intelligence tool for radiologists. With higher sensitivity and fewer false-positives than any offering currently on the market, we can automatically detect lung nodules in thoracic CT scans.
- Tom Kunicki, Ryan Chamberlain, Janak Mayer, Charles Maalouf
Second Place
Soybean Yield Prediction: Using Satellite Imagery to Predict Commodity Yields
- Assess the power of satellite images combined with weather data to predict yields and to do a comparative analysis of modeling strategies.
- Zhengyu (Taylor) Ma, Marguerite Oneto, Jasmine Tianjiao Qi
Capstone Project Presentations for the Spring 2016
Below is a summary of the Spring 2016 capstone project presentations:
ACCOUNTability: An analysis of consumer financial complaints database
- Our project website provides insights into the consumer finances complaint database. We aim to help to consumers, regulators, and others.
- Murat Aydogdu, Gopala Tumuluri
- Providing decision support in selecting open source tools and managing risk in the enterprise environment.
- Evan Kepner, Christopher Dailey, Andrea Soto
- Want to see if your BioTech startup is likely to succeed? Is that other one a sound VC investment? What will they IPO for?
- Denis Vrdoljak, Gunnar Kleemann, Danny Wudka
- Better dementia diagnoses through data.
- Nihar Patel, James Jasper, Amir Ziai, Ji Yan
- What is income inequality in America and why should it matter to you? Our goal is to help you understand what income inequality looks like and to identify the many components that make up what it takes to live and to thrive in the United States today. To do so, we’ve compiled data and developed visualizations from many different aspects of income.
- Steven Schaefer, Jared Maslin, Eric Freeman
- “Job Fiction”, a tool to help job seekers find Data Science jobs. It models creates the job seeker’s ideal fictional job and finds jobs that fit best both contextually and matching user preferences. Unlike other job search tools, this tool does not rely solely on key words and job titles. Job Fiction finds the perfect jobs without prejudice to job.
- Rajesh Thallam, Angela Gunn, Stephane Truong
Predicting Demand of Beer Styles
- Predicting the Demand of Beer Styles: A project to help brewers make the right beer in the right place at the right time.
- Raymond Buhr, Carson Forter, Satish Kumar Terala
- Using machine learning and historical data to predict H1B visa application wait times.
- Lucas Dan, Samuel Kabue, Sarah Neff
- Ratings can be very subjective. We use the written reviews from the Yelp Academic Dataset to produce a standardized star rating.
- Tim David, Jasen Jones, Malini Mittal
- All traffic fatalities are avoidable; so no traffic fatality is acceptable.
- Katrina Adams, Katherine Shelley, Kasane Utsumi
- Most travel recommendation engines focus on the body of the bell curve – the most likely duration of your trip. The best estimate. Our tool provides visibility into tail events – the “black swans” of the distribution. We combine data about special events with historical information on traffic congestion to predict which neighborhoods to avoid.
- Mandeepak Singh , Leanne Bradley, John Bocharov
- Surveillance cameras are common tools to monitor a home or business, but reviewing footage is tedious. We built a scalable system that harnesses ML to save users time.
- Juan Jose Carin, Vineet Gangwar, James Route, Sayantan Satpati
- Imagine a world where your kitchen helps you decide what to cook using your pantry and preferences. The Naive Baker is the beginning.
- Christopher Llop, Richard Johnson, Michael Demertzi, Miki Seltzer
- The tourist concierge is a data analytics platform to assist you on finding that perfect dream location destination.
- Albert Byun, Saru Mehta, Julian Phillips
Citation for this content: datascience@berkeley, the online Master of Information and Data Science from UC Berkeley