Aarhus University Seal / Aarhus Universitets segl

Technology

Keynote Presentations

 ­

 

 

 

Lars Arge
Professor
Dept. of Computer Science, Aarhus University

Big Data - possibilities and challenges

The pervasive use of computers, as well as tremendous advances in the ability to acquire, store and process data, has resulted in a spectacular increase in the amount of data being collected. There are many exciting possibilities for use of the data in both science and industry - big data can potentially have a very big impact! However, there are obviously also many challenges in unlocking this potential. Often its not clear what information can be extracted from the data and even less clear how to extract it. 

In his talk, Lars Arge will give examples of big data and its impact, especially highlighting how university research has resulted in innovative applications and solutions through collaboration between researchers and industry. He will describe some of the computer science research being performed at Center for Massive Data Algorithmics at Aarhus University, along with its applications in relation to for example flood risk estimation.

Lars Arge is a Professor of Computer Science at Aarhus University and Director of the Danish National Research Foundation (Danmarks Grundforskningsfond), and Center for Massive Data Algorithmics (MADALGO). MADALGO pursues a broad basic research agenda within efficient algorithms for big data, but also works with industry partners and researchers from other fields on using algorithms research advances in practical applications. Thus Arge has for example both obtained significant theoretical and practical results in relation to big terrain data. Arge is an elected member of the Royal Danish Academy of Sciences and Letters and The Danish Academy of Technical Sciences, a Fellow of the Association of  Computing Machinery (ACM), the recipient of the Danish Minister of Research Elite Research Award, and a co-founder of the company SCALGO that markets software and services in connection with big terrain data processing.­

­

­

­

­

Case Presentations

­

Cleaning Ocean Floor Scanning Data

Freek van Walderveen, SCALGO ApS; Ole Kristensen, EIVA A/S; Lars Arge, MADALGO, AU

Collaboration between algorithms researchers at MADALGO, AU, developers at SCALGO, and subsea experts at EIVA has resulted in the creation of a S-CAN software that cleans up multi-beam sonar scanning data about the ocean floor. The data is used for inspection of the many pipelines that transport oil and gas across the ocean. The research behind the software will be presented as well as examples on how its use results in a simpler and better cleaning of the very big data resulting from multi-beam sonar scanning.   

        

­

                         

 


 

Flood Risk Screening

Jeppe Sikker Jensen, COWI A/S; Morten Revsbæk, SCALGO ApS; Lars Arge, MADALGO, AU

Extreme and hazardous weather events resulting in flooding are becoming an increasingly common threat to our society. Collaboration between AU computer science researchers and industry has led to new flood risk screening products that predict the areas vulnerable to flooding. The case present software products that use the very detailed – and very big – data amounts in the national terrain model recently released by the Danish government along with other big datasets such as soil types and property values to predict flood risk events.  

     
                         

 

Inferring and Analyzing Transportation Behavior based on Mobile Sensing Data

Rolf Sode-Carlsen, Rambøll A/S; Kaj Grønbæk, AU

By using smartphone sensing frameworks, people’s movement activities are generating a big data collection that enables advanced analysis of traffic and transportation behaviors. These methods have huge potentials for traffic management. In a collaboration between Rambøll A/S and researchers from AU a mobile sensing project “Herning bikes to the moon” has been launched  as part of a  larger biking campaign. It supports collection of data from a large number of inhabitants in a city, and machine learning algorithms on the server extracts the detected bike trips from the multitude of transportation trips and count individual and collectively biked kilometers. In addition mobile sensing data from such campaign projects can be used as input to long term traffic management, since it provides knowledge of people’s transportation behavior in a certain area. 

                             

 

Air Quality at Your Street

Christian Lange Fogh, Environmental Protection Agency; Matthias Ketzel, DCE, AU

The public is concerned with air pollution and it is estimated that about 3,300 premature deaths are caused by air pollution. The awareness has led to the development of an interactive website where citizens can view the geographic variation of air quality down to the 2.3 million addressed in Denmark. A number of air quality models are used that integrate emission data with traffic, meteorology and street geometry data. The general public, real estate agents and different research areas are expected to make use of the new website which is developed by researchers at AU with consultation from the Danish Environmental Protection Agency.

           

 

Human Sensory Perception Based Healthy Eating 

Leslie Jørgensen, Rynkeby Foods A/S; Derek V. Byrne, AU; Line Holler Mielby, AU

The case will present evidence of practical applications of sensory research in industrial food production in Denmark. The beverage industry has long revolved around sugar reduction as a response to heightened calorie awareness. In this case, the focus will be on the case of stevia, a calorie free sweetener and sugar substitute. Stevia in principal is the perfect replacement for sugar on a nutritional level contributing zero calories to the diet. However, from a human senses perspective stevia is very different in perception terms to common sugar.

           

  

                                          

Big Data in Farm Animal Genetics

Anders Fogh, Nordic Cattle Genetic Evaluation; Bernt Guldbrandtsen, Aarhus University

For decades farmers and farmers’ organizations have amassed data on the characteristics and performance of farm animals. Ever increasing amounts of data are collected in databases, where they form the basis for optimizing management and selection of the best breeding animals. The databases are the foundation for close research collaboration between Aarhus University and farmers’ organizations and coops into the genetic mechanisms underlying differences in characteristics. In recent years growth of data has accelerated on several fronts. Automatic collection of data is introduced everywhere. At the same time revolutionary advances in biotechnology allow inferences on animal genetics in unprecedented detail.

     
                               

 

                                      

­