EC ENGR 219
Large-Scale Data Mining: Models and Algorithms
Description: Lecture, four hours; discussion, one hour; outside study, seven hours. Introduction of variety of scalable data modeling tools, both predictive and causal, from different disciplines. Topics include supervised and unsupervised data modeling tools from machine learning, such as support vector machines, different regression engines, different types of regularization and kernel techniques, deep learning, and Bayesian graphical models. Emphasis on techniques to evaluate relative performance of different methods and their applicability. Includes computer projects that explore entire data analysis and modeling cycle: collecting and cleaning large-scale data, deriving predictive and causal models, and evaluating performance of different models. Letter grading.
Units: 4.0
Units: 4.0
Most Helpful Review
Winter 2023 - This class is more or less a high level overview of machine learning. It covers common tools for analysis and feature extraction like dimensionality reduction and goes over common ML models. Some examples here include Naive Bayes, SVMs, decision trees, neural networks, etc. The coursework load is fairly light for an engineering class with 4 projects, which are long and reasonably well-guided assignments. These can be done in a group or alone, where I opted for the latter and would generally recommend that as it is quite doable and you learn more this way than by carving them up. Regarding the lectures and Prof. Roychowdhury, I generally found them a bit disorganised and did not engage with them much. Roughly 3 lectures in I pretty much focused only on the assignments and was fine. Said assignments I very much enjoyed though, as they were heavy on programming and analysis, which I wanted to practice. They were not particularly mathematically rigorous though, so I would recommend ECE 246 "Foundations of Statistical Machine Learning", by Prof. Diggavi for that.
Winter 2023 - This class is more or less a high level overview of machine learning. It covers common tools for analysis and feature extraction like dimensionality reduction and goes over common ML models. Some examples here include Naive Bayes, SVMs, decision trees, neural networks, etc. The coursework load is fairly light for an engineering class with 4 projects, which are long and reasonably well-guided assignments. These can be done in a group or alone, where I opted for the latter and would generally recommend that as it is quite doable and you learn more this way than by carving them up. Regarding the lectures and Prof. Roychowdhury, I generally found them a bit disorganised and did not engage with them much. Roughly 3 lectures in I pretty much focused only on the assignments and was fine. Said assignments I very much enjoyed though, as they were heavy on programming and analysis, which I wanted to practice. They were not particularly mathematically rigorous though, so I would recommend ECE 246 "Foundations of Statistical Machine Learning", by Prof. Diggavi for that.