1st International Workshop on Analytics for Service and Application Management (AnServApp 2019)

Monday, 21 October, 2019
09:00-10:30 Keynote: Traffic Mining – to AI or not to AI
Stefan Burschka, RUAG MRO, Switzerland

Bio: Physicist, formation in Quantum Optics, Microwave Technology, Optical Communication, Robotic and AI. After a long industrial career in HW and custom chip design, Data Mining, IT Troubleshooting and Network Security he is now the head of "Traffic Intelligence" at Ruag MRO Switzerland. His team works in the field of Encrypted Traffic Mining, IT Troubleshooting and Network Forensic. He is a lead developer of the open source SW Tranalyzer and teaches students in analog/digital HW design and traffic mining with and without AI.

Abstract: The IP network introduced a new quality of threats and leaks to infrastructures such as telecom, electrical, transport and banking systems. Unnecessary high complexity, instable behaviour and uncontrolled library usage spread vulnerabilities and thus induced a new kind of eavesdropping activity even in encrypted traffic. The talk will address the techniques to find features to answer questions about encrypted traffic nature and content based on the said vulnerabilities. Moreover the practice of Infrastructure Troubleshooting and TM is outlined, which motivated the development of the flow/packet based open source tool Tranalyzer (T2). The audience will learn about the industry KISS (Keep It Simple Stupid) approach, the importance of context awareness and proper feature selection, explaining several magic numbers in T2's output. Selected exercises will support the understanding of the dangers of blind trust in AI based tools in feature selection tasks.

11:00-12:30 Session 2
Technical and Short Paper Presentations 
Chair: Session Chair: Stefan Burschka

Predicting Distributions of Waiting Times in Customer Service Systems using Mixture Density Networks

Majid Raeis (University of Toronto, Canada)
Ali Tizghadam (TELUS & University of Toronto, Canada)
Alberto Leon-Garcia (University of Toronto, Canada)

Data-Driven Emulation of Mobile Access Networks

Ali Safari Khatouni (Dalhousie University, Canada)
Martino Trevisan (Politecnico di Torino, Italy)
Danilo Giordano (Politecnico di Torino, Italy)

Communication Challenges and Solutions between Heterogeneous Industrial IoT Systems

Daniel Kozma (Budapest University of Technology and Economics, Hungary)
Gabor Soos (Budapest University of Technology and Economics & Magyar Telekom Nyrt, Hungary)
Daniel Ficzere (Budapest University of Technology and Economics, Hungary)
Pál Varga (Budapest University of Technology and Economics, Hungary)

Detecting Factors Responsible for Diabetes Prevalence in Nigeria using Social Media and Machine Learning

Oladapo Oyebode (Dalhousie University, Canada)
Rita Orji (Dalhousie University, Canada)

13:30-15:00 Session 3
Technical and Short Paper Presentations 
Chair: Session Chair: Khurram Aziz

Can you hear me now? A call detail record based end-to-end diagnostics system for mobile networks

Anssi Tauriainen (Aalto University, Finland)

Machine Learning for Location and Orientation Fingerprinting in MIMO WLANs

Hui Xiong (Dalhousie University, Canada) Jacek Ilow (Dalhousie University, Canada)

NFV-VIPP: Catching Internal Figures of Packet Processing for Accelerating Development and Operations of NFV-nodes

Masahiro Dodare (Nagoya Institute of Technology, Japan)
Yuki Taguchi (Nagoya Institute of Technology, Japan)
Ryota Kawashima (Nagoya Institute of Technology, Japan)
Hiroki Nakayama (Bosco Technologies Inc., Japan)
Tsunemasa Hayashi (Bosco Technologies Inc., Japan)
Hiroshi Matsuo (Nagoya Institute of Technology, Japan)

Are There Bots even in FIFA World Cup 2018 Tweets?

Moath Bagarish (Dalhousie University, Canada)
Riyad Alshammari (King Saud bib Abdulaziz University for Health Science, Saudi Arabia)
Nur Zincir-Heywood (Dalhousie University, Canada)