Bin Yang (杨彬)

Projekttitel

Data analytics and machine learning on massive trajectory data for greener and more efficient transportation 

Hvad handler dit projekt om?

Continued digitization of societal and industrial processes produces vast volumes of data. Meanwhile, the transportation sector is facing profound disruption due to developments such as electric cars, autonomous cars, and sharing economy. My project will study how to best utilize massive trajectory data that describes the driving of cars to support the increasing needs for greener, cheaper, more efficient, and more predictable transportation. More specifically, my project will establish a dynamic and uncertain network model by analyzing massive trajectory data, which enables highly accurate, efficient, and scalable travel cost (e.g., travel time and greenhouse gas emission) modeling and path finding.

Hvordan opstod din interesse for dit forskningsfelt?

My grandfather bought a computer for me and my cousin in the early 90s. I played many computer games and also made some simple programs such as drawing triangles on the computer, which I really enjoyed. During my university education on computer science, I started to realize that computers can help people solve very complex and important problems, which motivated me to pursue a Ph.D. in computer science. Now, in the era of big data, my research focuses on computer algorithms that are able to learn knowledge from large amounts of data to help people solve problems that can benefit society.   

Hvad er de forskningsmæssige udfordringer og perspektiver ved dit projekt?

Many transportation applications rely on path finding algorithms on weighted graphs, where weights are in the form of static values, which represent, e.g., distances or average travel times of roads. However, traffic is often dynamic and uncertain, rendering the existing weighted graph model unrealistic. The challenges are (1) how to build a new model that is able to capture dynamic and uncertain traffic conditions from massive trajectory data; (2) how to utilize the new model to enhance the efficiency and quality of path finding algorithms. From an application perspective, my project has potential to profoundly change the foundation for vehicular path finding. From a scientific perspective, my project will extend graph theory by opening a new direction on stochastic path finding.

Hvilke perspektiver vurderer du selv, at din forskning på sigt kan have for det omgivende samfund?

Both the EU and Denmark are very ambitious on making transportation green. The application aspect of my project is well aligned with that ambition. My project will also bring solutions relating to dynamic, uncertain networks into the open domain and thus will facilitate open innovation in enabling greener transportation. Due to the data-driven nature of my project, it will also support emerging data science education activities. Further, the project will improve the internationalization of Danish research by collaborating with researchers from U.S., U.K., Australia, and Singapore.   

Hvad vil det betyde for din forskerkarriere, at du indgår i Sapere Aude-programmet?

The Sapere Aude programme will greatly boost my research career as an independent researcher. The grant enables me to establish and consolidate a research group on data analytics and machine learning, which offers me a very good foundation to achieve ground-breaking research outcomes with substantial societal benefits in the years to come. In addition, it offers good opportunities to enhance my research network through collaborations with leading international research groups. 

Lidt om mennesket bag forskeren

Currently, I live in Aalborg.

I like soccer, basketball, and rock music. Now both my wife and I spend most of our spare time with our daughter and son.