Catalog Description:

Overview of data structures, data lifecycle, statistical inference. Data management, queries, data cleaning, data wrangling. Classification and prediction methods to include linear regression, logistic regression, k-nearest neighbors, classification and regression trees. Association analysis. Clustering methods. Emphasis on analyzing data, use and development of software tools, and comparing methods.

Contact Hours:

• Lecture: 3 hours

Prerequisites: [MA305 or MA405] and [ST305 or ST312 or ST370 or ST372
Co-requisites: None
Restrictions: None
Coordinator: Dr. Rada Chirkova
Textbook: Data Mining with R

Course Outcomes:

Our goal is to help you gain skills in handling and analyzing data from ‘end to end.’ To mirror data science working environments, some of your assignments and in-class work will be done in multidisciplinary teams. By the end of this course, we want you to be able to:

• Use software tools to do data cleaning and data wrangling
• Use software tools to view and query stored data in a variety of formats
• Use software tools to implement k-nearest-neighbors, tree-based methods, multiple regression, and logistic regression for classification
• Develop association rules using the a priori algorithm
• Use software tools to implement k-means and hierarchical clustering
• Evaluate and compare the performance of classification algorithms

Topics:

• What is Data Science?, Syllabus Review, Course Project, Course Software
• Statistics and Data Structures Boot Camps
• SQL, Business Intelligence, NoSQL
• Geospatial Analytics
• Introduction to R
• Visualization
• Prediction and Classification: Performance Metrics, K-nearest neighbors, tree-based methods, multiple regression, logistic regression
• Using R with large data sets
• Large-scale data analysis
• Cluster Analysis: Distance metrics, k-means clustering, hierarchical clustering
• Association Analysis
• Time Series