Below is a list of the topics that we spent the most time on in class
this semester, and are thus likely to be on the final.


Dataframes
  - https://pandas.pydata.org/pandas-docs/stable/dsintro.html#dataframe
  - Indexing: loc, iloc, boolean selection
  - Basic functionality: head, tail, adding columns


Data preprocessing (book chapter)
  - Sec 3.1: Why preprocess and major stages
  - Sec 3.2: Ways of dealing with missing values
  - Sec 3.4.3: Principle components (what does it do)
  - Sec 3.4.4: Attribute selection
  - Sec 3.4.6: Histograms, best practices
  - Sec 3.5.2: Normalization, why and how


SQL/NoSQL
  - I won't ask questions on NoSQL
  - SQL: Will probably be asked to write some queries
      - SELECT, LIKE, DISTINCT, ORDER BY, GROUP BY, aggregation


Analytics
  - Decision trees: structure, learning algorithm
  - Logistic regression: form of classifier, role of weight vector,
    sigmoid
  - SVMs: properties, kernel trick, maximum margin


Experiments
  - Overfitting: what is it, when does it occur
  - Cross validation: train, test
  - Evaluation: parametric vs non-parametric tests


Clustering
  - Sec 10.1: Overview
  - Sec 10.2.1: k-means
  - Sec 10.3 up to 1.3.3
  - Sec 10.4.1: DBSCAN


Data visualization
  - Lie factor (https://infovis-wiki.net/wiki/Lie_Factor)
  - Parallel coordinates
  (https://en.wikipedia.org/wiki/Parallel_coordinates)


Map Reduce
  - Book through Sec 3.1
  - Focus on how to frame a problem
  - Why is MR useful?


Spark
  - RDD
  - Python only, not Scala
  - Know what various functions do
  - Probably asked to evaluate or write code (1 - 2 lines)


Bias/Variance
  - Know what bias, variance, and noise are
  - Explain what factors influence bias and variance
  - Bagging