Hien Nguyen, Yilin Li, Lyn Peterson
-
What is the chance of survival of a character from the well-known American TV show Game of Thrones based on what we know about the history of battles between the character's affiliated side and their enemies?
Goal: Regression. Predictive. We will train a model to predict a character's chance of survival from this Kaggle dataset and compare the results to the book events. -
Can GSS data, such as income, gender, education, party affiliation, race, etc., predict someone’s views on gender roles (ex. Role variable “girls should act like girls”)? Classification (group responses into agree, disagree, neutral) https://sda.berkeley.edu/sdaweb/analysis/;jsessionid=F77A62604D45529D21D8952C135B1CC5?dataset=gss16
-
The MNIST database of handwritten digits is a classic dataset for machine learning learners. Can we build up classification models to recognize what the digit(0-9) is in a given image based on 60,000 training images. The idea is mainly from Yann LeCun, NYU. http://yann.lecun.com/exdb/mnist/ They provided the results from numerous models with corresponding papers, which could be good references for us.