We love working with data
The most rewarding data describe how things connect over time. For example, from the patterns of how people and products connect in dynamic consumer networks, we can predict how the system will evolve over time. Revealing the inner workings of interconnected systems is the core of network science and the heart of our passion for data.
For Martin, the love came true when he, after his Ph.D. in network science at the legendary Niels Bohr Institute in Copenhagen, moved to the University of Washington in Seattle to do his postdoc in 2006. There he started working on a grand challenge in network science: how to simplify and highlight essential regularities in networks into maps. Mapping networks is a holy grail of data science because in the myriad links and nodes of a network hide answers to how we can predict how the network will evolve. For example, who will buy what next in a consumer network?
Based on information theory, we took a novel and advantageous approach to mapping networks. But coming up with the underlying math was not enough. Mapping large networks requires efficient algorithms as well. And comprehending and communicating the results required compelling visualizations such as our innovative alluvial diagrams for mapping change over time. Most researchers focus on one of the three aspects, the math, the algoritms or the visualizations, but we decided to go for all of them. It has paid off. Because when the right math, algorithms, and visualizations came together and worked in synergy, we immediately started making exciting discoveries in complex interaction data.
After we had shown maps of science and how it changes over time in ways that nobody had seen before, we soon received requests to help other researchers make good maps of their interaction data. We realized that we needed powerful tools for automating the process of going from raw data to insightful maps and discoveries. From 2009 at Umeå University, Martin started working with Daniel, who built the first interactive map and alluvial diagram generators. At the same time, Andrea's benchmark tests showed that our algorithms outperform other approaches. As a result, thousands of researchers have used our software available on mapequation.org. Moreover, the broad interest in our tools has allowed us to collaborate with many different researchers in exciting projects, including mapping change in the overnight money market with researchers from the Federal Reserve Bank of New York.
When Andrea joined our research team in 2014 after his postdoc at Northwestern University, we reached a critical mass of skills. With his landmark work on data-driven algorithms and skills to build interactive visualizations of large data sets such as Wikipedia and IMDb, we decided it was time to let people outside of academia to benefit from our research. We were fortunate to team up with Jakob and Niklas, who, thanks to more than 20 years combined of experience in media, e-commerce, and international business, have established the essential link between our research solutions and the industry data challenges.
By recruiting Christian, John, Florian, and Robert, who complement our skills for maximum productivity, we can realize our goal to develop an automated and self-learning analytical tool for transactional data with intuitive, interactive visualizations that empower people to convert their data into actions and insights. For people who are struggling to make use of their data with multiple incompatible software and tedious manual work, our tools can revolutionize their work by making it simple and rewarding to turn complex data into meaningful actions.
We want to help people exceed their goals, generate new knowledge, and be more productive. We want people to love their data.