![]() Finally, an ideal curriculum includes authentic real-world project experience, for example through practica or capstone work requiring data collection or wrangling (Berman et al., 2018 Chen et al., 2019 Irizarry, 2020 Song & Zhu, 2016 Wing et al., 2018). ![]() Grounded in computational thinking and data-driven paradigms, a robust curriculum should address programming languages (high-level such as Python, R, and Spark, and low-level such as C++) and data science tools, big data infrastructures (e.g., Hadoop, MapReduce, NoSQL, NewSQL, in-memory and cloud computing, databases, big data warehousing, and data virtualization), the big data analytics lifecycle (e.g., business analysis, data understanding, data preparation and integration, machine learning, data mining, statistical analysis, model-building, evaluation, deployment, and monitoring), data management and governance (e.g., data modelling, relational database knowledge, and stewardship, curation, and preservation), research methods (develop research questions, employ the scientific method, and evaluate outcomes), domain knowledge, ethics, human-centered data science, and the behavioral disciplines (soft skills, e.g., collaboration, thinking critically and empathically, asking creative questions, project management, communicating with domain experts) (Cao, 2019 Chen et al., 2019 Demchenko et al., 2017 Irizarry, 2020 Song & Zhu, 2016Wing, 2019 Wing et al., 2018).
0 Comments
Leave a Reply. |