Summary

Data science plays an important role in many industries. In facing massive amount of heterogeneous data, scalable machine learning and data mining algorithms and systems become extremely important for data scientists. The growth of volume, complexity and speed in data drives the need for scalable data analytic algorithms and systems. In this course, we study such algorithms and systems in the context of healthcare applications.

In healthcare, large amounts of heterogeneous medical data have become available in various healthcare organizations (payers, providers, pharmaceuticals). This data could be an enabling resource for deriving insights for improving care delivery and reducing waste. The enormity and complexity of these datasets present great challenges in analyses and subsequent applications to a practical clinical environment.

Expected Learning

In this course, we introduce the characteristics of medical data and associated data mining challenges on dealing with such data. We cover various algorithms and systems for big data analytics. We focus on studying those big data techniques in the context of concrete healthcare analytic applications such as predictive modeling, computational phenotyping and patient similarity. We also study big data analytic technology:

Scalable machine learning algorithms such as online learning and fast similarity search;

Big data analytic system such as Hadoop family (Hive, Pig, HBase), Spark and Graph DB

Syllabus

Big Data

• Predictive Modeling
• Dimensionality Reduction & Tensor Factorization
• Graph Analysis

Healthcare

• Computational Phenotyping
• Patient Similarity Metrics
• Medical Ontology

Technologies

• MapReduce
• Spark
• Hadoop

Required Knowledge

Basic machine learning and data mining concepts such as classification and clustering;

Proficient programming and system skills in Python, Java and Scala;

Proficient knowledge and experience in dealing with data (recommended skills include SQL, NoSQL such as MongoDB).