Alex Mackie, a recently-completed GIS student from the University of Edinburgh has commercialised his dissertation project which gives books locations, and founded his own company mappit.net.
For his MSc dissertation, submitted only three months ago, Alex chose to mix two subjects close to his heart; books and maps. As well as a research exercise, he realised there was commercial value to mapping books, which has been largely ignored by the industry. If books are properly georeferenced then location-aware e-readers and tablets can use their user’s location to recommend locally relevant books or provide the option to search for books, relating to intended holiday destinations, favourite mountain or place-based Christmas present. This extends the principle that physical bookstores already recognize a demand for locally relevant books, with Waterstones and other retailers stocking shelves with books linked to the shop’s location.
Rather than having to have the text of thousands of books, the initial method involved extracting place names from Amazon book reviews, which total a remarkable 80 million words, on the basis that reviewers discuss the places books were about or set in. It may seem strange to be mining metadata for further, more specific metadata but the vast amount of review text is a potentially rich source of information. Yet there was a problem – while the reviews do indeed contain sufficient place-names to geolocate books, the error rate caused by misidentified mentions of things like Dundee cakes and Yorkshire terriers meant the data was not ideal for powering location-based book recommendation.
Thus an alternative approach was taken, disambiguating existing metadata. By taking metadata in the form of toponyms from book catalogues and using a custom algorithm to disambiguate these toponyms to real-world coordinates, an interactive “global book map” of 50,000 books had been built. This is growing all the time as more books are constantly processed. All of this in time for Scottish Book Week 2014 (Monday 24th – Sunday 30th November).
As the data grows, displaying tens of thousands of points elegantly and speedily becomes a significant challenge. Thus the book map takes a new approach to intelligently clustering these large point fields for web mapping. The PostGIS spatial extension to PostgreSQL was used to aggregate points by country and region, depending on the zoom level of the map. This was made possible because of lightning-fast point-in-polygon operations in PostGIS. While seemingly complicated, this approach gives a much more useful map than simpler alternatives, which tend to merge features together based on naive proximity rather than any real-world similarity.
Building on his academic background in philosophy and now GIS, Alex was able to learn further skills during his dissertation research, including natural language processing, how to handle ‘big’ data, spatial database design, building, testing and hosting dynamic web apps with a spatial component, geovisualisation and front-end web design.
Alex has built on his book map to add participatory mapping to the portfolio of his new company. He is creating a simple participatory mapping environment that allows users to quickly and easily create their own custom map layer, to which others can contribute. The intended application is for one-off consultation projects and niche interests with a spatial element. He is currently looking for beta-testers with interests in collaborative mapping, who can contact him via his web site or email@example.com.