Data Science in Practice: Tools and Applications

(3 Credit Hours) · Undergraduate

Given the growing volume and complexities of real-world data, successful deployment of data science pipelines into practice often require intangible factors, beyond the modeling, including careful formulation of the substantive problem of interest, non-trivial data pre-processing, powerful computational software, intuition into when and why models work or fail to work, effective visualization and communication of the results, awareness of the ethical consequences, and close collaborative efforts. Data Science in Practice: Tools and Applications explores these computational and critical thinking skills necessary to solve data science problems in real-world applications. To this end, this course will guide students through a series of hands-on learning projects based on real scientific datasets. Through these real-data projects, students will gain experience with data pre-processing, advanced visualization tools, unsupervised and supervised learning tasks, interpretability tools, and advanced computing tools that are commonly used in industry such as git, distributed computing, reproducible documentation, and open-source software packaging.