INDE 577 — Data Science and Machine Learning

Instructor: Randy R. Davila

Models included in the course:

• Supervised Learning:
  • Model Building and Error Analysis
  • Linear Regression
  • Gradient Descent
  • Logistic Regression
  • Deep Neural Network
  • k-Nearest Neighbors
  • Decision/ Regression Trees
  • Ensemble methods
• Unsupervised Learning:
  • k-Means Clustering
  • Principle Component Analysis

Datasets included in the course:

• Wheat seed dataset:

  • Similar to iris dataset. The data set contains seed information belonging to three different wheat varieties: Kama, Rosa and Canadian, represented by 1, 2 and 3 respectively
  • Link:https://www.kaggle.com/datasets/jmcaro/wheat-seedsuci

• Fashion MNIST dataset:

  • It has the same structure as MNIST dataset, with a training set of 60000 samples and a test set of 10000 clothing images. The size of the image is also fixed to 28 × 28. In this way, the preprocessed image data is reduced to the minimum level.
  • Link:https://www.kaggle.com/datasets/zalando-research/fashionmnist

• Anime Recommendations Database:

  • The dataset contains the name and type of animation and the ratings of animation by more than 70000 users.
  • Link:https://www.kaggle.com/datasets/CooperUnion/anime-recommendations-database

• Boston Housing dataset:

  • The Boston housing dataset consists of 14 features and contains information collected by the U.S. Census Bureau about housing in Boston, Massachusetts.
  • Link:https://www.kaggle.com/competitions/boston-housing/overview/description