Trendswatch: Information age

Every 10 minutes, people generate more data online than they created in the whole of human history up until 2003. It’s an incredible statistic – but how is the world to make use of this torrent of information?

The term “big data” refers to data sets so large or complex that up until recently they defied meaningful analysis by traditional software. However, technological developments and advanced algorithms have improved our capacity to collate, share and extrapolate statistics from such data sets on an unprecedented scale. 

In the US, researchers have used big data analytics to track disease epidemics through Twitter posts, and create predictive maps that forecast where different types of crime are likely to occur. Exponents believe that this fast-developing form of analytics could eventually help predict economic fluctuations, long-term health trends and even global conflict. 

For museums, tracking this kind of big data over time could provide valuable evidence of the long-term impact that cultural engagement has on health, wellbeing and education. At a more practical level, it could enable institutions to fine-tune their programming and marketing retail strategies to boost numbers and improve the visitor experience. 

It is also useful on a smaller scale; one US institution, History Colorado in Denver, is using big data IT software to collect real-time visitor information from its cash registers, allowing it to make up-to-the-minute decisions on staffing and use of space.

Big data analytics also has the potential to transform research in fields like cultural heritage. 

Are UK collections ready for big data? Most institutions have digitised at least some of their collections and archives. London’s Natural History Museum has gone a step further and is launching a web portal later this year specifically aimed at storing and analysing major data sets generated by the millions of specimens it holds.

But on a wider scale, big data analytics relies on an environment of openness, shared resources and cross-sector cooperation. It draws on layers of information from a variety of different sources. This requires, for example, more museums to make their digital content available via “linked open data”, that is, agreeing to drop licence restrictions and enabling it to be linked to other content within a shared space. 

Many museums are understandably concerned about losing control over how their collections data is used and aggregated. But the sector has already benefited from open data through projects such as:

All of these are valuable repositories that can be mined for big data. The next step might be to harmonise these disparate sources and develop some form of shared online infrastructure that would enable the cultural heritage sector and academics to search, curate and analyse the data they hold. 

Thirteen partners from across the European Union, including the Collections Trust in the UK, are participating in a project called the Digital Cultural Heritage Roadmap for Preservation, to explore how it might be done and what resources might be needed to facilitate this flow of knowledge. 

Relatively speaking, museum data is not all that “big”, but it is complicated. We are only beginning to learn how to model the complex web of historical relationships and events in a database. Cultural heritage data is widely recognised in the technology community as a significant challenge that can help drive innovation in the field. 

Techniques such as data mining, visualisation and statistical analysis offer opportunities for museums to gain new insights into their collections, their audiences and even their buildings. 

New developments like the Natural History Museum’s Data Portal and the British Museum’s Collection Online platform provide unprecedented access to collections information.

Opening up our data for analysis will also help museums build new partnerships, for example with higher education and high-tech industries. But we mustn’t lose sight of the small, highly personal experiences that people expect from us. We can seize the opportunity of big data, so long as we remember to focus on quality as well as quantity.