:::{updated} F22 :::
This week we introduce classification as a prediction task, with methods for evaluating classifiers.
:::{module} week10 :::
In this video, I introduce the week and what classification is.
:::{video} :length: 6m39s :name: 10-1 - What Is Classification :::
In this video, I introduce {term}log odds, along with the {term}logistic function and its inverse,
the {term}logit function.
Log odds are a useful concept in many situations!
:::{video} logistic :name: 10-2 - Log-Odds and Logistics :length: 10m4s :::
We're now ready for our first classification model: logistic regression.
:::{video} :name: 10-3 - Logistic Regression :length: 9m7s :::
The confusion matrix describes the outcomes of a classification model and is the basis for computing effectiveness metrics.
:::{video} confusion :name: 10-4 - The Confusion Matrix :length: 11m48s :::
- The Wikipedia article has a very good diagram of the confusion matrix and its derived metrics.
The demo notebook for our initial logistic regression videos.
:::{video} baselines :name: 10-5 - Baselines :length: 9m44s :::
This is provided for reference.
The following StatsModels page documents its logistic regression:
This is not an assigned reading - it is here for your reference.
This video describes the log likelihood that is the objective function used by logistic regression.
:::{video} :length: 16m54s :name: 10-6 - Log Likelihood :::
This video introduces SciKit-Learn, and using it for a logistic regression.
:::{video} :length: 6m42s :name: 10-7 - Scikit-Learn :::
The SciKit Logistic notebook demonstrates training and using
{py:class}sklearn.linear_model.LogisticRegression.
This video introduces the receiver operating characteristic (ROC) curve, and its use in evaluating classifiers and selecting tradeoffs.
:::{video} :length: 7m25s :name: 10-8 - Receiver Operating Characteristic :::
Load the Penguin data, and use a logistic regression to try to classify a penguin as Gentoo or Chinstrap using various measurements. Delete the Adelie penguins first, so you have a binary classification problem.
This video revisits sources of bias and discusses the assumptions underlying prediction.
:::{video} :length: 22m :name: 10-9 - Biases and Assumptions :::
:::{reading} pdb :title: Prediction-Based Decisions and Fairness :url: https://arxiv.org/abs/1811.07867 :length: 3650 words :::
Read Sections 1 and 2 of the following paper:
Shira Mitchell, Eric Potash, Solon Barocas, Alexander D'Amour, Kristian Lum. 2018. Prediction-Based Decisions and Fairness: A Catalogue of Choices, Assumptions, and Definitions. arXiv:1811.07867 [stat.AP].
We'll come back to ideas here, but sections 1 and 2 describe the assumptions underlying most classification problems. While the overall topic of the paper is fairness in making these decisions, I am not assigning it because it is a fairness paper; rather, those first two sections provide a succinct description of the assumptions that we make when we undertake most classification problems. They apply no matter what properties of a classification problem or model we care about.
If you would like to learn more, I recommend:
The Week 10 quiz will be posted to {{LMS}}.
:::{reading} tech-to-prison :length: 2000 words :url: https://medium.com/@CoalitionForCriticalTechnology/abolish-the-techtoprisonpipeline-9b5b14366b16 :title: "Abolish the #TechToPrison Pipeline" :::
Read Abolish the #TechToPrison Pipeline (the Medium reading time estimate includes the thorough — and valuable — footnotes and list of 2435 signatories). This article probes in more detail the assumptions underlying classes of criminal justice data science applications.
Assignment 5 is due {date}wk11 sun long.