The amount of unstructured or semi-structured online information in the current Web is growing exponentially. Providing structure to this information by (semi)automatic means has become an increasingly appealing goal along the last decade. Deriving structure greatly enhances the utility and value of this wealth of information, as it enables a much more efficient, effective, and flexible use of online information by different types of applications. Automatically handling unstructured information sources, and extracting structure from them, are challenging tasks that have been addressed by Artificial Intelligence research. The problem involves issues of knowledge extraction, knowledge representation and analysis, which need to be addressed in order to subsequently exploit the extracted structure in a wide range of applications.
In this workshop we aim to organise a multidisciplinary event for researchers and specialists from different areas of Artificial Intelligence, interested in the extraction and exploitation of semi-structured online information sources. It aims to be a forum for the presentation and discussion of shared problems and techniques from different disciplines, seeking to stir synergies and collaborations among participants, under the common frame of semi-structured information: how can it be obtained? How can it be processed and elaborated? What can it be used for? How can it be applied?
Our first objective is to address questions such as how can structured information be obtained? From which online information sources? The workshop welcomes the submission of papers analysing techniques to build structured information from the Web, including –as of particular interest- social media sources, such as social networks, blogs, forums, wikis, and folksonomies. Furthermore, we want to identify which techniques are more appropriate for generating each type of structure and why.
Another objective of this workshop is to analyse the practical aspects of the exploitation of structured and semi-structured information, presenting and analysing specific methods, tools, and techniques, in order to identify the involved drawbacks and benefits. Several questions arise in this context: which research areas and applications can make a better use of structured information? Once structured data is available, do applications become easier or harder to manage? Because of this, we look for typical Artificial Intelligence fields and applications where the novelty is the fact of using structured information, in contrast with other, more classical sources of information. Examples of this situation are the standard natural language processing techniques and information retrieval systems.