'Parcels of Venice' project wins SNSF support

Jacopo de Barbari, Venetie MD, detail, View of Venice, 1500 ab.

Jacopo de Barbari, Venetie MD, detail, View of Venice, 1500 ab.

College of Humanities (CDH) Professor Frederic Kaplan, head of the Venice Time Machine project (VTM), and VTM Project Coordination Manager Isabella di Lenardo have won CHF 700,000 in funding from the Swiss National Science Foundation (SNSF) for a project that aims to bring a new dimension to two centuries’ worth of spatial and population data from the city of Venice.

The four-year project, entitled “Parcels of Venice: a computational approach to the evolution of Venice in the 19th century”, will provide support for PhD students to work together on three separate but intertwined challenges.

First, the co-principal investigators want to improve an existing machine learning algorithm for linking the cadasters, or real estate maps, of Venice established during Napoleon’s rule with population records for each map section, or parcel. These records, called sommarioni, were used to keep track of who inhabited and owned which properties, and how much they were taxed. Kaplan, who leads the CDH Digital Humanities Lab, says that when put together, the two types of data will form a “full snapshot” of the Venice population over the last 200 years.

Through the VTM project, researchers had already created an algorithm for automatically segmenting and “reading” these Napoleonic cadasters – which were the first standardized geometric representations of the city – and linking them with corresponding sommarioni. Now, Kaplan and di Lenardo want to further develop the algorithm so it can be applied to other types of records.

“We are confident that we will be able to extract generically all these different full versions of the populations at different moments in Venice’s history,” Kaplan says.

Another goal of the “Parcels of Venice” project is to maintain the connection between the algorithm and the extracted data, so the system can be continually updated as the algorithm “learns”.

Parcels of Europe

In parallel, researchers will work on generalizing and extending the system to other locations. Kaplan and di Lenardo hope that ultimately, it will be applicable to cities across Europe as part of the Time Machine Europe project, which they also coordinate. In March, the project announced that it had been selected to receive €1 million in funding from the European Commission to prepare a detailed roadmap for extracting and utilizing “Big Data of the past”.

“There are a lot of questions to be answered about the standardization of data: how we can generalize the Venice case to other Napoleonic cadasters in Europe, and how we can visualize and present all this information,” di Lenardo explains.

Finally, the grant will provide support for researchers to compare the extracted Venice population data with historical literature based on small-scale case studies of the city’s social and urban structures.

“For for the first time we will be able to see the whole population of Venice, so the question now is: does that change the theories, hypotheses, and results that have been published so far?” Kaplan says. “It’s a great moment to do digital history now. We’ve been preparing for this research over the last six years, and now we have the possibility of taking the first fruits of these efforts.”