We have big data, but we need big knowledge: weaving surveys into the semantic web

2/5 ( ratings)

The semantic web aspires to improve the accessibility of information to automated systems as much as the regular WWW makes information accessible to human users. Moreover, the standards and tools of the semantic web are designed to exploit so far as possible the resources that are already in place delivering that regular web.
In this brave new world a question to your smartphone like "What is the readership of The Times in Humberside?" can be answered as cogently as "What is the population of Brazil?" - and both of them better than they are at present .
At the other end of research, as soon as you learn a respondent lives in postcode sector KT6 your questionnaire can be steered by knowing this is a low-crime, high-income, highly-urbanised area.
This paper shows how behind the scenes the tools to realise this and other ways to enhance the collection and usage of our data are falling into place.
It introduces the technologies that have been developed for knowledge processing and points the way to how they may be used to great advantage within survey organisations.
Finally, it presents the results of a practical project that has developed a tool to convert Triple-S survey data into semantic data and shows how a diversity of analyses of data and metadata across multiple surveys can be greatly simplified using semantic web tools.