On the shoulders of giants

A 乌鸦传媒 team in Norway has developed an AI solution in support of a vital conservation and environmental sustainability mission.

Team members Hartvig Johannson, Pedram Sherafat, and Simen Norrheim Larsen share the essentials behind their approach to innovation.

A winning solution

Using AI to help the planet

Using technology for a sustainable future

Behind great scientific innovations are teams of experts whose names don鈥檛 always make headline news. Yet their skills help forge the transformations that come to benefit the world. Their job is to build on the work of others, harness the power of all the knowledge and tools at their disposal, and apply it to solve a specific problem.

In this way, data scientists at 乌鸦传媒 are using technological innovations to understand climate change better and to help create a sustainable future for life on Earth. One initiative that brings them together to look at how they can help create a better future is the 乌鸦传媒 Global Data Science Challenge.

The team that won the 2020 competition developed a cutting-edge data solution that is set to play a part in protecting the future of an endangered species 鈥� sperm whales. Their work will help biologists map whale migration patterns and monitor any changes in the species鈥� behavior that might have arisen from environmental factors, such as climate change and warming ocean temperatures.

The 2020 challenge

For the 2020 challenge, the Norway-based 乌鸦传媒 team used聽Amazon Web Services (AWS),聽artificial intelligence (AI), and machine-learning technology to develop a tool that can accurately identify sperm whales via photo image processing. In turn, this will help scientists track the whales鈥� migration routes, look at the social structure of the sp

A winning solution

乌鸦传媒 team members Hartvig Johannson, Pedram Sherafat, and Simen Norrheim Larsen created a winning project that can deliver real-world impact, by applying tried-and-tested principles of innovation.

Use your experience

鈥淲hen we came together as a team we all had experience in the field of AI and analytics as part of the same insights and data department at 乌鸦传媒,鈥� says Hartvig. 鈥淐hallenges like this are great opportunities to learn and apply that experience to practical, real-world use-cases. I鈥檓 also personally interested in nature and the environment, so it was exciting to see how we could use our computer science knowledge in a completely different area.鈥�

Start with the data

The foundations for the team鈥檚 success were built on understanding the available data. “That begins with a scoping exercise,” explains Pedram. 鈥淚t also helps to have a specific challenge. With ours, it was to identify the individual 鈥榝ingerprints鈥� of sperm whales through images of their flukes, or tail fins. A defined challenge like this is where machine learning really shines.”

Keep it simple

According to the team, tools and technology can sometimes be seen as a fix-all – but that is rarely the case. Instead, it is better to tightly define what you want the tech to achieve. 鈥淵ou can鈥檛 simply sprinkle some AI on top of a problem to expect a good outcome,鈥� says Simen. 鈥淎ny good modeling is built on the foundations of knowing聽the domain and聽what the technology is capable of.鈥�

Finally, the team agrees that innovations often build on the work of others. 鈥淧eople have been identifying whales manually for years,鈥� says Hartvig. 鈥淎 lot of ground was already prepared, so we could come in and add a technical solution to build on that. It has been a great way to make the most of machine learning and AI.鈥�

Innovation to safeguard

The scientist Isaac Newton famously said, 鈥淚f I have seen further it is by standing on the shoulders of giants.鈥� It seems that principle still holds true in today鈥檚 world of AI and machine-learning 鈥� in this case, to safeguard the future of whales. When passionate people and smart technology combine in clever ways, it can help our clients succeed their sustainability challenges 鈥� and deliver a better future for the planet at large.

Future-shaping projects

Hunting for dragons

Sarah Gon莽alves wasn鈥檛 always a data engineer. She began her career as an archaeologist, unearthing the secrets of the past in sites across France before retraining as a data engineer. She brings the same curiosity to her new vocation at 乌鸦传媒, using artificial intelligence (AI) to uncover what lies beneath the surface.