Big Data Analysis - Week 3

Monday the 4th - Friday the 8th of May 2020

Monday 4th of May (afternoon):
Lectures: Computational scaling and complexity, k-nearest neighbour algorithm (kNN), and k-means clustering algorithm (BV).
     You are expected to see these three short lectures before the lecture, which will then be an interactive discussion.
     The slides can be found her: Computational scaling, kNN algorithm, and Infrastructure.

Zoom:Link to lecture (Recorded!).
     Recording in Lecture video (109 MB) and Lecture audio (26 MB) along with Lecture chat (3 kB).
     Recording in Exercise video (116 MB) and Exercise audio (27 MB).

Exercise: The first exercises are described in the videos, and they consist of applying the k-NN and k-CM algorithms to toy data.
     Following this, you should try to apply the same (and other) algorithms to e.g. breast cancer and/or the Aleph b-jet data.
     KNN toy example: KNN.py or KNN.ipynb.
     Clustering toy example: Clustering.py or Clustering.ipynb.
     Cancer example: Cancer.py or Cancer.ipynb along with cancer data.
     Wine example: Wine.py or Wine.ipynb, along with Wine data and Wine names.

Wednesday 6th of May (morning):
Lectures: Recurrent Neural Networks (RNN) and Long Short Term Memory (LSTM) (TP) and Echo State Networks (ESN) and anomaly detection (James Avery).

Zoom:Link to lecture (Recorded!).
     Recording in Lecture video (217 MB) and Lecture audio (31 MB) along with Lecture chat (9 kB).
     Recording in Exercise video (155 MB) and Exercise audio (24 MB).

Exercise: Try to predict the next value in a sequence from Sequence training.
     Also try play around with: Echo State Network (ESN) on github, with the examples mini ESN and ESN sequence training.
     When you feel, that you understand LSTMs and the exercise, feel free to start/work on the Small project.

Wednesday 6th of May (afternoon):
Lectures: Population Mixture Models (AA).

Zoom:Link to lecture (Recorded!).

Exercise: The exercises are contained in the presentation. For example apply the Expectation-Maximization algorithm to data of your choice.
     Also, towards the end of the session, you're working to work on the small project.


Last updated: 5th of May 2020 by Troels Petersen.