This web-page contains homework assignments for the course STK-INF4000 'Selected Topics in Data Science' taught at the University of Oslo in the spring term 2017.

Weekly Assignments

• Week 1
• Week 2
• Week 3
• Week 4
• Week 5
• Week 6
• Week 7
• Week 8
• Week 9
• Week 10
• Week 11
• Week 13
• Week 14
• Week 15

Syllabus

The course is meant to be very interactive and I'll react to some degree to student's requests concerning the syllabus. As soon as I can foresee the content of the following lectures, it will be published here and at UiO's course page.

• Week 1 (Jan. 16th):
  • Welcome.
    • What is data science?
    • Why should you take this course?
    • What you'll learn.
  • Practical information.
  • Python, part 1
    • Basics.
    • Data types.
    • Lists and tuples.
    • Functions.
    • Plotting.
• Week 2 (Jan. 23rd):
  • Python, part 2
    • Dictionaries
    • Objects
    • Generators
  • Web Scraping 101
    • Scraping etiquette.
    • Extracting information from HTML pages.
    • Spiders and Crawlers in Python.
• Week 3 (Jan. 30th):
  • Python, part 3
    • Scipy/Numpy
    • Unit testing
  • Databases 101
    • MongoDB
    • SQL
  • RESTful APIs
    • Twitter API
    • Facebook API
• Week 4 (Feb. 6th):
  • Introduction to Machine Learning.
    • Aims of ML.
      • Interpretability vs. accuracy.
    • Just enough theory.
    • K-Nearest Neighbors.
  • scikit-learn: Machine Learning in Python.
    • K-Nearest Neighbors on real-world data.
• Week 5 (Feb. 13th):
  • Project discussion.
  • Small data EDA in Python: The pandas package.
    • What is EDA?
    • What should I look for?
• Week 6 (Feb. 20th)
  • Linear regression.
    • Introduction to linear methods.
    • Linear regression in scikit-learn.
    • Linear regression in scipy.
    • Regularized linear regression.
  • Python leftovers.
    • scipy and numpy
    • Generators.
    • Error handling and logging.
• Week 7 (Feb. 27th)
  • Big data strategies.
  • MapReduce.
  • Apache Spark.
• Week 8 (Mar. 6th)
  • Linear classification.
    • The classification problem.
    • Discriminant Analysis.
    • Logistic regression.
  • Applied classification
    • Classifying by hand.
    • Scikit-lean and Statsmodels.
• Week 9 (Mar. 13th)
  • Applied classification in Spark
    • Spark data frames
    • Logistic regression in Spark.
  • Unsupervised Learning.
    • Anomaly detection.
    • Clustering.
      • K-Means in Spark.
      • Hierarchical clustering.
• Week 10 (Mar. 20th)
  • Tree Methods.
    • Decision trees, pros and cons.
    • Trees for classification.
    • Trees for regression.
  • Trees in scikit-learn.
• Week 11 (Mar. 27th)
  • Project delivery discussion (projects due Apr. 3rd).
  • Regression methods for anomaly detection.
    • Using trees in Spark.
  • Ensemble Methods.
    • Boosted decision trees in Spark.
    • Random forests in Spark.
• Week 12 (Apr. 3rd)
  • Project presentations.
• Week 13 (Apr. 24th)
  • Model Evaluation.
    • Train error, test error, and all that.
    • Optimism and AIC.
    • Bootstrapping and bagging.
  • Random Forests.
  • Clean Modeling: Data pipelines.
• Week 14 (May 8th)
  • Data pipelines in Apache Spark.
  • Natural Language Processing 101.
    • What is NLP?
    • Basic methods of NLP.
    • Building a simple sentiment analysis tool.
• Week 15 (May 15th)
  • Presenting results - the open source way.
    • Javascript 101
    • JQuery, Vue.js and Google Charts
    • Serving data with Flask.
• Week 16 (May 22nd)
  • What we missed.
    • Neural Networks and deep learning.
    • Support vector machines.
    • Heuristics.
  • Review and preparation for the exam.
  • Q&A.