Course materials

Introductory module

Objectives: set expectations; explore data science raison d’etre; introduce systems and design thinking; introduce software tools and collaborative coding; conduct exploratory/descriptive analysis of class background and interests.

Week 0

• Thursday meeting: Course orientation [slides]

• Assignments:

  • install course software and create GitHub account

  • submit GitHub username in Canvas assignment (due by 11:59 PM, Monday, September 28th)

  • fill out intake form (due by 11:59 PM, Sunday, October 4th)

  • read Peng and Parker (2022);

  • prepare a reading response (due by 11:59 PM, Monday, September 28th)

Week 1

• Tuesday meeting: On projects in(volving) data science [slides]

• Section meeting: software and technology overview [activity]

• Thursday meeting: basic GitHub actions [activity] [slides]

• Assignments due by next class meeting:

  • read MDSR 9.1 – 9.2

  • prepare a reading response (due by 11:59 PM Monday, October 5th)

  • read (and follow along with) the [tutorial for connecting Git to RStudio]

Week 2

• Tuesday meeting: Introducing class intake survey data [slides]

• Section meeting: tidyverse basics [activity]

• Thursday meeting: planning group work for analysis of survey data [slides]

• Assignments:

  • first team assignment due Friday, October 16, 11:59 PM PST. [Click here to see group memberships.] The [assignment module is available here.].

Module 1: biomarker identification

Objectives: introduce variable selection, classification, and multiple testing problems; discuss classification accuracy metrics and data partitioning; fit logistic regression and random forest classifiers in R; learn to implement multiple testing corrections for FDR control (Benjamini-Hochberg and Benjamini-Yekutieli); discuss selection via penalized estimation. Data from Hewitson et al. (2021) .

Week 3

• Tuesday meeting: introducing biomarker data; multiple testing [slides]

• Section meeting: iteration strategies [activity]

• Thursday meeting: correlation analysis; random forests [slides] [activity]

• Assignments due by next class meeting:

  • read MDSR 10.1 - 10.2

  • read Hewitson et al. (2021)

  • fill out [self/peer evaluation form about group work on Module 0], due Tuesday, October 20th, 11:59 PM PST.

  • second team assignment will be due Friday, October 30th, 11:59 PM PST. [Click here to see group memberships.]

Week 4

• Tuesday meeting: review of random forests, logistic regression [slides]; cloud computing and virtual machines [slides]

• Section meeting: logistic regression and classification metrics [activity]

• Thursday meeting: development containers in VS code [slides]

• [[Video walkthrough]](https://gauchocast.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=94310a09-2206-4ad1-bcd8-b4ec017d9576)] for setting up development containers

Module 2: fraud claims

Objectives: introduce NLP techniques for converting text to data and web scraping tools in R; discuss dimension reduction techniques; introduce multiclass classification; learn to process text, fit multinomial logistic regression models, and train neural networks in R.

Week 5

• Tuesday meeting: data introduction and basic NLP techniques [slides]

• Section meeting: string manipulation and text processing in R [activity]

• Thursday meeting: dimension reduction; multinomial logistic regression [slides] [activity]

• Optional further reading:

  • MDSR Ch. 19

  • Cambria and White (2014)

  • Khan et al. (2010)

• Assignments:

  • fill out [self/peer evaluation form about group work on Module 1], due by Tuesday, November 3rd, 11:59 PM PST.

Week 6

• Tuesday meeting: feedforward neural networks [slides]

• Section meeting: fitting neural nets with keras [activity]

• Thursday meeting: finishing neural networks [slides]

• Assignments:

  • Read Emmert-Streib et al. (2020) (§1-5, §9) and prepare a reading response due Tuesday, November 10, 11:59 PM PST.

  • third team assignment (Module 2) due Friday, November 13, 11:59 PM PST. [Click here to see group memberships.] The [assignment module is available here.].

• Optional further reading:

  • Alzubaidi et al. (2021)

  • Goodfellow, Bengio, and Courville (2016) Ch. 6 (advanced)

Module 3: soil temperatures

Objectives: build a forecasting model; introduce concepts of spatial and temporal correlation; discuss function approximation and curve fitting with regression techniques; fit elementary time series models and regression with AR errors; spatial interpolation.

Week 7

• Tuesday meeting: data introduction; function approximation using basis expansions [slides]

References

Alzubaidi, Laith, Jinglan Zhang, Amjad J Humaidi, Ayad Al-Dujaili, Ye Duan, Omran Al-Shamma, José Santamarı́a, Mohammed A Fadhel, Muthana Al-Amidie, and Laith Farhan. 2021. “Review of Deep Learning: Concepts, CNN Architectures, Challenges, Applications, Future Directions.” Journal of Big Data 8 (1): 1–74.

Cambria, Erik, and Bebo White. 2014. “Jumping NLP Curves: A Review of Natural Language Processing Research.” IEEE Computational Intelligence Magazine 9 (2): 48–57.

Emmert-Streib, Frank, Zhen Yang, Han Feng, Shailesh Tripathi, and Matthias Dehmer. 2020. “An Introductory Review of Deep Learning for Prediction Models with Big Data.” Frontiers in Artificial Intelligence 3: 4.

Goodfellow, Ian, Yoshua Bengio, and Aaron Courville. 2016. Deep Learning. MIT Press.

Hewitson, Laura, Jeremy A Mathews, Morgan Devlin, Claire Schutte, Jeon Lee, and Dwight C German. 2021. “Blood Biomarker Discovery for Autism Spectrum Disorder: A Proteomic Analysis.” PLoS One 16 (2): e0246581.

Khan, Aurangzeb, Baharum Baharudin, Lam Hong Lee, and Khairullah Khan. 2010. “A Review of Machine Learning Algorithms for Text-Documents Classification.” Journal of Advances in Information Technology 1 (1): 4–20.

Peng, Roger D, and Hilary S Parker. 2022. “Perspective on Data Science.” Annual Review of Statistics and Its Application 9: 1–20.