Data Mining: Classification - Badge

From Badge Wiki
Jump to navigation Jump to search
Badge image

Name

Data Mining: Classification.

Issuer

Università degli Studi di Milano-Bicocca.

Issued since 20 April 2016.

Description

This Badge is earned by learners participating in the course "Data Mining: Classification" offered by EduOpen. The Badge is to all intents and purposes the course’s certificate of attendance.

Badge Criteria

This BADGE has been issued to the student who attendend, on the EduOpen Mooc Platform, the course on  "Data Mining: Classification " teached by Prof. Fabio Stella of the Department of Informatics, Systems and Communication of the University of Milano-Bicocca. The student watched methodology and hands-on video lectures about the following topics; data types, data exploration, missing data replacement and pre-processing. The student also watched methodology and hands-on video lectures about; formulation and solution of binary and non binary classification problems with and without cost matrix, classification performance measures and related estimation techniques, ROC, Lift and Cumulative gain curves, and features selection algorithms. Furthermore, the student watched methodology and hands-on video lectures on how to train, test and validate the following classification models; decision trees, logistic regression, feed-forward neural networks, support vector machines, naive Bayes, tree augmented naive Bayes and Bayesian classifiers. The student used the KNIME open source software platform to perform practice sessions, and was asked to develop and to upload, to the EduOpen Mooc plarform, 16 KNIME workflows.

Skills

The owner of this Badge participated in and successfully completed the course "Data Mining: Classification". The owner of this BADGE has the following Competences: - How to pre-process different data types. - How to formulate binary and non binary classification problems. - How to develop classification models to solve binary and non binary classification problems. - How to compare different classification models, with and without cost matrix, to select which is the "optimal classifier". - How to discover the relevant attributes/features to solve a classification problem. - How to develop a KNIME workflow to formulate and solve binary and non binary classification problems.

Tags

ComputerandDataSciences, Classification, Datamining

Badge Platform

Bestr

External links

Bestr Badges