Programme

The characteristic of big data- MapReduce approaches: Hadoop, Spark e google MapReduce - subsampling-based approaches - divide and conquer approaches - online updating approaches - analyzing big data with classical statistical models - machine learning algorithms.

Core Documentation

Slides provided by the teacher

Jared Dean. Big Data, Data Mining, and Machine Learning: Value Creation for Business Leaders and Practitioners, 2014, Wiley.
Readings and lecture notes provided by the teacher.

Reference Bibliography

Ankam, Venkat. Big data analytics. Packt Publishing Ltd, 2016. Pang-Ning Tan, Michael Steinbach, Anuj Karpatne, Vipin Kumar, Introduction to Data Mining, Addison-Wesley. Dietrich D.. Data Science and Big Data Analytics: Discovering, Analyzing, Visualizing and Presenting Data. Wiley, 2015.

Type of delivery of the course

The course consists of 36 hours divided into lectures and exercises. During the exercises some applications on real data are illustrated.

Type of evaluation

The satisfactory achievement of the aims of the course is assessed through an exam with marks out of thirty. The exam includes an oral interview. The mark is expressed out of thirty and the pass mark is 18. The oral interview, of length approximately equal to 25 minutes, consists in theoretical questions on the main methods, models and, in general, notions included in the course program. In particular, the focus will be on evaluating the ability to correctly apply the taught methods, the rigour and clarity of expression.