meta data for this page
  •  

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
perccom_summer_school_2018:group2:start [2018/06/08 15:16] – farnibakhan07perccom_summer_school_2018:group2:start [2018/06/13 16:02] (current) – tawseef1410
Line 14: Line 14:
 **Supervisor(s):** Professor Mohammad Shahadat Hossain; Professor Karl Andersson\\ **Supervisor(s):** Professor Mohammad Shahadat Hossain; Professor Karl Andersson\\
 **Hosting Institution:** Luleå University of Technology, University of Chittagong\\ **Hosting Institution:** Luleå University of Technology, University of Chittagong\\
-**Abstract:** Among the various natural calamities, flood is considered one of the most catastrophic natural hazards, which has disastrous impact on the socio-economic lifeline of a country. Assessing flood risks will allow to take proper precautions before it occurs to save human lives along with their residential and commercial properties. However, the assessment of risks involves severe uncertainty as crisis response teams need to take decisions based on the huge amount of incomplete and inaccurate information, which are mainly coming from sensor streaming in real time. Therefore, the present methods of flood risk assessment cannot assess the risk of flooding accurately. This paper rigorously investigates various types of uncertainties associated with different factors and voluminous dataset along with BRBES’s approach to develop an expert system which will be integrated with Big Data platform to assess flood risk in real time. In addition, a comprehensive study of the data-driven approach in comparison with the knowledge-driven approach regarding flood risk assessment has been conducted. The system processes all the complex mathematical computation on extremely large dataset through Spark machine learning library at the backend and through a web-based interface, the users can monitor the risk of flood in real time. In order to check the reliability of the system, a comparison has been performed among various approaches such as linear regression, decision tree, random forest, artificial neural network and experts’ opinion while the BRBES produces reliable results than from the other approaches. This research explores the challenges, opportunities and methods, required to leverage the potentiality of Big Data and BRBES to assess the risk of flooding.\\+**Abstract:** Among the various natural calamities, flood is considered one of the most catastrophic natural hazards, which has a significant impact on the socio-economic lifeline of a country. The Assessment of flood risks facilitates taking appropriate measures to reduce the consequences of flooding. The flood risk assessment requires Big data which are coming from different sources, such as sensors, social media, and organizations. However, these data sources contain various types of uncertainties because of the presence of incomplete and inaccurate information. This paper presents a Belief rule-based expert system (BRBES) which is developed in Big data platform to assess flood risk in real time. The system processes extremely large dataset by integrating BRBES with Apache Spark while a web-based interface has developed allowing the visualization of flood risk in real time. Since the integrated BRBES employs knowledge driven learning mechanism, it has been compared with other data-driven learning mechanisms to determine the reliability in assessing flood risk. Integrated BRBES produces reliable results comparing from the other data-driven approaches. Data for the expert system has been collected targeting different case study areas from Bangladesh to validate the integrated system.\\
 **Paper:** [[https://www.doria.fi/|Insert link to thesis paper]].\\ **Paper:** [[https://www.doria.fi/|Insert link to thesis paper]].\\
  
Line 30: Line 30:
 **Title:**  Architecting and Designing Sustainable Smart City Services in Living Lab Environment\\ **Title:**  Architecting and Designing Sustainable Smart City Services in Living Lab Environment\\
 **Author:** Alam, Md Tawseef\\  **Author:** Alam, Md Tawseef\\ 
-**Supervisor(s):** Professor Jari Porras; Professor Ahmed Seffah\\+**Supervisor(s):** Professor Jari Porras\\
 **Hosting Institution:** Lappeenranta University of Technology\\ **Hosting Institution:** Lappeenranta University of Technology\\
-**Abstract:** Insert thesis abstract(200 words)\\+**Abstract:** Smart Cities have become a popular trend in the recent years. In terms of sustainability, cities become smart when it provides intelligent services to the inhabitants using information and communication technologies without threating the future of environment, economy and the society. However, the process of developing such sustainable smart services has certain challenges; challenges to understand the needs of the people living in the city. Inhabitants of the city or the citizens are the key stakeholder in case of applications developed to provide services in a smart city. It has been found that, active involvement of the people throughout the process is a way to design such services. On the other hand, integrating sustainability, specially including environmental data to the smart city services has been found challenging. Therefore, this research discusses an approach on combining environmental data with regular smart city services and to include city inhabitants in the process, this approach is adapted from the concept of living lab methodology. Finally, an application has been developed to represent a smart city service following this method and people from various background from Helsinki City has evaluated the application, as well as evaluation of the method was done by a small number of software developers, which produced promising results. \\
 **Paper:** [[https://www.doria.fi/|Insert link to thesis paper]].\\ **Paper:** [[https://www.doria.fi/|Insert link to thesis paper]].\\