ACCEPTOr : Automatic Construction of a Catalogue of EU ProjecTs' Outcomes

Abstract

The information found on the internet is increasing immensely on the daily. There is so much data found online that it impossible to be able to process it all. So much so, that there is a large portion of information which could be beneficial to people that is ignored. A substantial amount of information is left unused due to various reasons. Some of which may include overwhelming users with the numerous amount of data in existence, unstructured information and also a lack of awareness regarding certain beneficial parts of data found online. It is a known fact that every year, an immense part of the European Union’s budget is allocated to fund various research projects whose ultimate aim is to further improve the lives of the rest of the European Citizens. This report describes a system which is able to crawl different project websites, download their content and extract relevant information such as project objectives. The aim of this is to be able to showcase the main information from projects, in a catalogue, to these Citizens, who ultimately contribute to the funds needed for such projects. During the evaluation stage of our system we found that both the crawling component and scraping component are working well and producing a large amount of extractions. Having said this, it is due to production of an enormous number of results that the precision of the overall system did not turn out as satisfactory as what was hoped for.