Tags

, , , ,

If you are dealing with an hydric network, if you are trying to visualize the electric cables in the city in order to use less material as you can, if you are trying to understand the properties of a soil to choose the right place in which you will plant your product and in a lot of other cases connected to the most various fields of knowledge, then you know that mapping information is very important today.

The Geographic Information System (GIS) is a technology whose goal is to obtain conclusions from huge datasets mapping variables and information contained inside; today it is largely diffused because there are lots of open source software to do this. So GIS and technology like this help computer scientists, mathematicians and engineers to solve their problems having a concrete vision of the connection between their data. There are different ways of looking at a environment: at first, having a glance from the landscape at your window you can say “here I see a building, here I see a park” and so on; in this way, you are having a look to the environment around you as a collection of entities with associate properties (color, shape and so on), but you can see also the environment as inserted in a three dimensional coordinate system in which you want to evaluate some properties (air pressure, temperature and so on), without having a distinction of singular entities, and this is the so called continuous field vision, because you are evaluating continuous functions on a 3D space.

GIS applied to show the electric cables, the villages, the wind flow in all Uganda: work made by a friend, Niccolo Ficarelli, MSc in Energy Management and Sustainibility at EPFL while he was working for Deutsche GIZ in Kampala - Uganda.

GIS applied to show the electric cables, the villages, the wind flow in all Uganda: work made by a friend, Niccolo Ficarelli, MSc in Energy Management and Sustainibility at EPFL while he was working for Deutsche GIZ in Kampala – Uganda.

This latter vision can be helpful to make conclusions from problems in which you would never imagine to be possible to apply GIS, as for example archives about war and battles and related problems to the populations involved.

The first article, “Linked Open Data and First World War”, analyzes the information obtained in the archives of the Canadian province “Newfoundland and Labrador” about the Great War. The First World War, indeed, was the first war in which it was possible to document battles, movements of populations, decisions of armies and so on, so we have in our archives more information about this tragedy than obviously the most sadly famous war of the 19th century. But now the great news are that we can use this information, mapping them and get a summary: for example, in the archive in Canada, there are the information about a Newfoundland community that was constrained to move because of the War, so we can map this over time, and see which are the main causes, if there were natural causes or particular attacks through a boundary; moreover, it is possible to get access to the documents of British armies and German armies, and so this is useful to obtain information, because we don’ t know exactly how the generals were used to give orders to their army, but with a visual map we can understand better the movement of soldiers, the movement of the navy and the one of the air force and if there was a smart combination of them; the scientists in the project mapped, for istance, also the Western front, with number of British and German deaths and with number of soldiers involved.

The second article, which is the result of combined work of Huygens Institute and Leiden Institute in Netherlands, presents a project in which the main goal is to map the cultural migration of artists in Low Countries by using three complementary datasets, complementary in the sense they deal with different historical periods of Netherlands or in general of Europe, together with a kind of crowdsourcing of datasets made by other researchers in the Low Countries. Mauro Martino made an animated visualization of the movement of artists, mapping their places of birth and death, but there is a critique to this kind of technique: the datasets are, according to most, a biased sample, because they treat above all North America and Europe, and also because birth and death are too limiting data. The project is to integrate this work using more features, such as the relationships between the artists. But how can we understand when there was a relationship and when can we assess it is relevant or not? GIS takes an important role in this, because we can map the houses, streets and neighborhoods in which artists lived, in order to get at a first glance the possible links. In this case, GIS is necessary, because we have different datasets obtained from completely different projects, and we want a fast way to visualize them and compare them: for instance, the ECARTICO database visualizes where artists lived in Amsterdam, whereas the database in Rome makes us understand where Dutch artists had their houses in the eternal city. When they recollected their results, the summary was quite surprisingly: there were lots of artists living in the same house, or in the same building, or with a distance of a couple of streets.

Although they are talking about completely opposite subjects, art and war, paintings and battles, the philosophy of these two articles is not so different: they want to use information about people migration (a population in North America or artists throughout Europe) to assert conclusions about the culture and demographic data about population.

The third article, “Mapping and modeling centuries of literary geography across million of books” written by Matthew Wilkens, University of Notre Dame (USA), presents how GIS techniques can be used to visualize literary publications in the world at a national level (we color every state in the world according to a scale related to the number of books published) and a city level (drawing balls around cities dimensioned with the same criterion above) and it presents also how these summary help us to find links with cultural and demographic factors.

The third article, instead of the other ones, deals with documents related to book, papers and not strictly connected to people, but the aim is the same: we want to use GIS to get results about possible cultural changes. All these articles discuss cultural changes in the world by using GIS, but each one focuses on a difference subject: people migration constrained by war, migration of artists and their distribution in a map of a city and, the last one, distribution of published books.

If we have a look at the distributions of the number of the books published in each country, we notice that there is a bigger concentration in Europe, North America and the Mediterranean rim. This  show a kind of cultural conservatorism, because in that countries in which there are fewer books there are also lower attempts to change culture, economics and society. The second purpose of the research is to understand if it is possible to use datasets with cultural information to build predictive models. The work includes also a linear regression over an extremely huge number of variables and to use the best subset selection criterion with respect to the parameter adjusted R square (a parameter which quantifies how the data are fitted but penalizes a greater number of variables) to choose proper variables to compute the linear regression. Also in this article we have a quite incredible result: some variables such as the literacy rates or per capita newspaper publications are not relevant, whereas there are lots of variables which usually we don’ t expect to be important such as total population, immigrant population, degree of urbanization.

This a new vision, because we are trying to use statistics to move beyond the data and to use a predictive model to get some information about the future; in the summary, we have shown how nowadays GIS can be used for every purpose, for a better understanding of the past and of the world war, but also to make some prediction about areas with high cultural development in the future.

Advertisements