Features spaces and a learning system for structural-temporal data, and their application on a use case of real-time communication network validation data

• Published in: PloS one 2020-02, Vol.15 (2), p.e0228434-e0228434
• Main Authors: Schwenk, Guido, Jochinke, Ben, Müller, Klaus-Robert
• Format: Article
• Language: English
• Subjects: Biology and Life Sciences; Calibration; Classification; Communication; Communication networks; Computer and Information Sciences; Computer Systems - standards; Cost analysis; Data Accuracy; Data Mining - methods; Data Mining - standards; Datasets as Topic - standards; Decision trees; Deep Learning; Engineering and Technology; Humans; Machine Learning; Mobile Applications - standards; Mobile Applications - statistics & numerical data; Mobile communication systems; Natural language processing; Neural Networks, Computer; Objectives; Performance evaluation; Physical Sciences; Properties (attributes); Quality of service architectures; Real time; Recall; Reproducibility of Results; Research and Analysis Methods; Social Sciences; Statistical analysis; Teaching methods; Time; Time Factors; Validation Studies as Topic; Wireless communications
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0228434

The service quality and system dependability of real-time communication networks strongly depends on the analysis of monitored data, to identify concrete problems and their causes. Many of these can be described by either their structural or temporal properties, or a combination of both. As current research is short of approaches sufficiently addressing both properties simultaneously, we propose a new feature space specifically suited for this task, which we analyze for its theoretical properties and its practical relevance. We evaluate its classification performance when used on real-world data sets of structural-temporal mobile communication data, and compare it to the performance achieved of feature representations used in related work. For this purpose we propose a system which allows the automatic detection and prediction of classes of pre-defined sequence behavior, greatly reducing costs caused by the otherwise required manual analysis. With our proposed feature spaces this system achieves a precision of more than 93% at recall values of 100%, with an up to 6.7% higher effective recall than otherwise similarly performing alternatives, notably outperforming alternative deep learning, kernel learning and ensemble learning approaches of related work. Furthermore the supported system calibration allows separating reliable from unreliable predictions more effectively, which is highly relevant for any practical application.