CMPSC 445Applied Machine Learning in Data Science

3 credits · Computer Science · Penn State

Applied machine learning techniques are used in many different areas, such as the classification, visualization and analysis of data, clustering, and understanding of natural languages for human-computer interactions. These applications are crossing the boundaries of computer science and data science. Big technology firms have all started offering their own cloud machine learning platforms. This course will start with an overview of supervised and unsupervised learning, and introduce the associated libraries. It covers basic machine learning concepts, tasks, and workflow using an example classification problem based on K-nearest neighbors, Naïve Bayes, Support Vector Machine (SVM), K-means, and implementation using Python libraries. Natural language processing (NLP) techniques including n-gram models, grammar, parse trees, and part-of-speech tagging will be discussed. The issue of dimensionality of data will be discussed, and the task of clustering data, as well as supervised approaches for creating predictive models will be described, and learners will be able to apply Python predictive modeling methods while understanding process issues related to data generalizability (e.g. cross validation, overfitting). The course will also look at more advanced techniques, such as neural network, feed-forward network, back-propagation and deep learning with cloud AI services. Technological differences between using cloud services at a higher level of abstraction and coding locally will be discussed. Students will be able to identify the difference between a supervised (classification) and unsupervised (clustering) technique, identify which technique they need to apply for a particular dataset, manage and understand data, and engineer features to meet that need. Students will work in teams to develop web applications that use industry standard cloud services provided by one of the AI cloud service providers.