The EUSPN-2017 took place in Sweden, Lund between 18 and 20 of September 2017.

The 8th International Conference on Emerging Ubiquitous Systems and Pervasive Networks (EUSPN-2017) is a leading international conference for researchers and industry practitioners to share their new ideas, original research results and practical development experiences from all Ubiquitous Systems and Pervasive Networks related areas.

The paper presented in the conference is a part of our research in Project ASPires. It presents a performance analysis of the end-to-end sensor-to-cloud personal living platform. ( download )

The analysis is based on a typical architecture starting from a single sensor and actuator and continuing to the virtualized services at smart dust, dew, fog and cloud level.

The system is diverse and allows interconnection of different sensors/ actuators technologies directly or throughout gateways.

The experiments in a living lab presented use energy harvesters and ZigBee PRO sensors. The results from sensor network are applicable for non-real-time and non-critical data connection.

For critical and non-critical measurements that need to be supported in a typical living environment there is a necessity to use different priorities of the services and different sensors as well. 

For more information, click here.

SQL Saturday is a free training event for Microsoft Data Platform professionals and those wanting to learn about SQL Server, Business Intelligence and Data Analytics.

SQL Saturday 642 Sofia is the 9th such event organized in Bulgaria and this year it set a national record of 113 pre-conference session submissions and more than 30 lecturers at 5 parallel tracks. The admittance for all 400+ participants was free with most costs covered by sponsorships from well established companies in the IT industry on national and European level.

The ASPires project was presented by Ivelin Andreev, software architect at Interconsult Bulgaria, in the context of an area where the ICB company expertise in Machine Learning could be applied for detection and early prevention of forest fires. The key session topic Machine Learning- Unpacking the black box is intended to break some common myths for machine learning and provide a practical approach towards evaluating and improving model performance. In addition, it also provided in-depth comparison of major ML platform vendors and touches upon some more advanced topics like artificial neural networks and in particular feature extraction with convolutional neural networks.

The presentation is made publicly available at SlideShare where it got 12700+ views for just 3 days.

The slides could be downloaded from here: https://www.slideshare.net/ivoandreev/machine-learning-for-iot-unpacking-the-blackbox or as pdf

The conference website and schedule are available here: http://www.sqlsaturday.com/642/Sessions/Schedule.aspx