In English

Software Lifecycle Management Unsupervised Anomaly Detection

Ludwig Friborg ; Victor Christoffersson
Göteborg : Chalmers tekniska högskola, 2017. 42 s.
[Examensarbete på grundnivå]

The purpose of this thesis is to evaluate if unsupervised anomaly detection, the task of nding anomalies in unlabelled data, can be used as a supportive tool for software life cycle management in nding errors which are tedious to detect manually.

The goal is to apply the techniques of unsupervised machine learning on data-sets that are collected and analysed from a miniature-scaled research vehicle system that resembles the operation of a real automotive vehicles electrical architecture.

Using a stacked autoencoder implemented with TensorFlow, the nal application is able to detect anomalies within the collected data-sets from the research vehicle. This proves the concept of utilising machine learning for error detection as a viable method. Finally concluding whether the techniques of unsupervised anomaly detection is applicable on a larger scale for real automotive vehicles.

Nyckelord: Machine learning, Unspervised Anomaly Detection, Autoencoder, Software Lifecycle Management, Arti cial Nural Networks.

Publikationen registrerades 2017-06-21.

CPL ID: 250028

Detta är en tjänst från Chalmers bibliotek