Personal tools

SC²S Colloquium - Oct 11, 2018

From Sccswiki

Jump to: navigation, search
Date: Oct 11, 2018
Room: 00.08.053
Time: 15:00 - 16:00

Tobias Bernecker: Outlier Detection Techniques using Sparse Grids Density Estimation

This is a Bachelor's Thesis submission talk advised by Paul Sarbu

Outlier detection helps to improve results of a clustering process by identifying noisy, anomalous data points in a dataset. However, lots of techniques for outlier detection require a density estimation of the data points, which cannot be computed exactly. To deal with this problem, spatially adaptive sparse grids can be used to learn and approximate the underlying density function of a multi-dimensional dataset. After this learning process, also known as sparse grids density estimation, the obtained approximated function can be evaluated at every data point to receive a corresponding density value. In this thesis, several outlier detection techniques including the Local Outlier Factor and the Local Density Factor are presented. Furthermore, a new density- based approach to obtain a factor for the outlierness of a data point is introduced. The purpose of this thesis is to assess whether outlier detection is a suitable field of application for sparse grids density estimation. To this end, this approach is integrated into an outlier detection framework allowing comparison to other known methods. To validate the results of the presented outlier detection techniques, several artificial datasets with a certain percentage of outliers are tested. Additionally, real datasets are used for further expirements and analysis of the studied detection methods.

Keywords: Sparse Grids, Density Estimation

Philipp Zetterer: Investigation of Cluster Analysis Algorithms Using Radio Measurement Data of Public Mobile Networks

This is an external Master's Thesis submission talk advised by Michael Obersteiner and Paul Sarbu

This master thesis deals with cluster analysis of mobile network data. This data is normally broadcasted by the infrastructure of the mobile networks e.g. GSM, UMTS or LTE. The goal is to show that this data can be clustered in order to find commonalities inside those data. Therefore, several algorithms are implemented which are able to detect clusters as well as outliers. They are of particular interest for various companies since they can be from different origin e.g. test cells or unregulated ones. In order to achieve that goal, different algorithms will be tested including partitioning approaches, hierarchical ones and density-based ones.


Keywords: cluster analysis, mobile networks, real-world data