Speakers: Zygimantas Jasiunas and Nuno Durão, LASIGE – DI/FCUL
When: April 21, 14h00
Talk 1: Exploratory work of appliance monitoring for early fault detection to reduce e-waste.
Abstract: As the global economy grows rapidly, global sales and consumption of appliances grow in parallel. Based on consumer reports, millions of appliances are sold each year, and these numbers keep growing as the year passes. The majority of large scale appliances consist of refrigeration, laundry, air-conditioners and other kitchen appliances. As a side effect, tons of e-waste is produced each year of appliances that are no longer useful, functioning or performant as they used to be. Large household appliances such as eclectic stoves, fridges and washing machines are making more than half of all collected e-waste. Unfortunately, the percentage of recycled e-waste is not satisfactory. Only 40% of all e-waste is being recycled in European Union (EU). EU is keen on working towards recyclability, reusability and promoting prolonged appliance useful life. We are working on enhanced appliance monitoring using multiple sensors and exploring possible ways to identify early anomalies and faults of appliances, which would increase repairability and reduce e-waste by prolonging appliance life. In this presentation, we will show the lab we are creating to monitor several types of appliances and will discuss the approach required to acquire process and store sensor data.
Short bio: Zygimantas Jasiunas currently is a PhD student, developing his thesis in the area of machine learning and IoT. He is advised by Professors José Cecílio and Pedro M. Ferreira and his main interests are engineering monitoring solutions based on IoT-enabled hardware, machine learning and cloud computing.
Talk 2: Discovery of Web Attacks by Inspecting HTTPS Network Traffic With Machine Learning and Similarity Search
Abstract: Web applications are the building blocks of many services and play a critical role in today’s world. Traditional dependence of Deep Package Inspection to discover web attacks suffers with the adoption of HTTPS. This work proposes a new reliable method and system to identify traffic that may include web application attacks by analyzing HTTPS network flows and discovering payload content similarities.
Short bio: Nuno Durão is a MSc student in Information Security, currently working on his thesis under the supervision of Prof. Ibéria Medeiros and Prof. Vinícius Cogo.