Made by Students
AI, MACHINE LEARNING RESEARCH PROJECTS
Graph Alignment-Based Protein Comparison
Inspired by the question of identifying mechanisms of viral infection, we are interested in the problem of comparing pairs of proteins, given by amino acid sequences and traces of their 3-dimensional structure.
Basic Machine Learning and Classification
Flower Classification
Using Machine Learning to Combat Fake News
Art Genre Classification with Machine Learning
Identifying Melanoma With Deep Learning
Analyzing Trending Youtube Videos with Machine Learning
Predicting Medical Expenses with Machine Learning
Using AI to Predict Vulnerabilities by Age
Predicting Deaths from Overdose using Neural Networks
Stack Overflow Tags with Machine Learning
Neural Network on SHVN Dataset Paper
The problem I aimed to solve with deep learning will use this dataset to solve is training a model to successfully classify skin cancers, which is important for diagnosis.
Machine Learning Fires
This project attempts to analyze game data from Hearthstone, an online card game developed by Blizzard.
Study of the Impact an asteroid collision with earth can have and identified hazardous asteroids based on criteria provided by NASA.
Alvin Studied the impact an asteroid collision with earth can have and identified hazardous asteroids based on criteria provided by NASA. He got very promising results and understood the subject in depth:
Machine Learning Suicide Prediction
The best project by far was Jonathan's: He studied a rather taboo subject such as suicide rates and tried to pin-point factors that influence it within a country and how it relates to economic index of a country. He produced high
Canon vs Fanon? — Distinguishing Fanfiction from Canonical Writing Using a Neural Network
Predicting Air Pollution Levels from Satellite Images Using Deep Convolutional Neural Network
Novel method for improving deep-learning accuracy of diagnosing breast cancer using different image augmentations
Determining the impact of information loss across different image augmentations towards the performance of various convolutional neural networks on classifying mammograms