Authors
Deborah Markham, Andrew Muddiman
Abstract
Social Media has the potential to provide beneficial information to emergency services during a crisis. However, the quantity and quality of the data make it a source that is challenging to manage. Thus, we need to develop effective information mining techniques to transform these large quantities of low value information into low quantities of high value information. In this deliverable we discuss techniques we have implemented, as well as those we can implement in the remainder of the project, to decrease the quantity and increase the value of the information presented to emergency services.
Purpose of the Document
The purpose of this document is to describe and assess techniques we have implemented for extracting and refining information present in social media streams, as well as providing guidance for our next steps. We intend for this deliverable to inform readers about the details of the algorithms we have chosen, their strengths and weaknesses as demonstrated by their application to test data, as well as their context in the field as a whole. Moreover, we intend for readers to realise the benefits of the algorithms we have implemented to emergency services, as well as the potential benefits of our future work.