Call for Papers
The graph structure naturally models connections. In natural language processing, connections are ubiquitous, on anything between small and web-scale. We find them between words as grammatical, collocation or semantic relations; between clauses, sentences or larger text fragments as discourse, entailment, similarity or other relations; between complete texts or web pages; between entities in a social network; between concepts in ontologies or other knowledge repositories. Beyond the more often encountered “regular” graphs, hypergraphs have also appeared in our field, useful for capturing relations between more than two units.
Graphs have been rigorously studied, both mathematically and computationally. Such a well developed base could be – indeed has been – very useful for the field of Natural Language Processing: existing proofs and solutions could solve the problems that we manage to map into the graph framework. The relation need not be one way: the graph form of an NLP problem may lead to interesting computational research, maybe particularly when dealing with very large scale structures.
It is not surprising then that graphs occur often, and successfully, in NLP work. A concentrated evidence of this fact is the ongoing TextGraphs workshops. With the 9th edition taking place this year collocated with EMNLP 2014, this series of workshops has exposed and encouraged the synergy between the fields of Graph Theory and Natural Language Processing. The work presented shows a nice progression – from “small” graphs that provided efficient and elegant solutions for NLP applications that focused on single documents for POS tagging, word sense disambiguation, or semantic role labeling, to increasingly larger structures for ontology learning through open IE, analysis of information propagation in social networks, to name but a few.
We think the time has arrived to summarize the most successful graph-based approaches to a variety of Natural Language Processing problems, both small and large scale, and everything in between.
Topics of Interest
To reflect the many applications of graph structure analysis in NLP, we propose the following topics, which cover the entire range from small sentence-level graphs to very large web-scale graphs:
∙ Graph-based methods for Information Retrieval, Information Extraction and Text Mining
- Graph-based methods for word sense disambiguation
- Graph-based representations for ontology learning
- Graph-based strategies for semantic relations identification
- Encoding semantic distances in graphs
- Graph-based techniques for document navigation and visualization – Reranking with graphs
- Label Propagation, etc.
∙ New graph-based methods for NLP applications
- Random walk methods in graphs
- Spectral graph clustering
- Semi-supervised graph-based methods
- Methods and analyses for statistical networks
- Small world graphs
- Dynamic graph representations
- Topological and pretopological analysis of graphs, etc.
∙ Graph-based methods for applications on social networks
- Rumor proliferation
- Community detection
- Information diffusion
- Network evolution
- Multiple identity detection – Language dynamics studies – Surveillance systems, etc.
Zornitsa Kozareva, Yahoo! Inc, USA Vivi Nastaste, FBK, Italy Rada Mihalcea, University of Michigan, USA
First call for papers: February 20, 2014 Second call for papers: April 17, 2014 Submission deadline: June 1, 2014 Initial decisions: August 15, 2014 Submission of revised versions: October 15, 2014 Final decisions: December 15, 2015 Submission of camera-ready versions: January 30, 2015
Dr. Natalia Konstantinova EXPERT Network Training Coordinator Editorial Assistant for the Journal of Natural Language Engineering Research Group in Computational Linguistics Research Institute of Information and Language Processing University of Wolverhampton Stafford Street WOLVERHAMPTON WV1 1LY Email: n.konstantinova at wlv.ac.uk<mailto:n.konstantinova at wlv.ac.uk> Tel: + 44 1902 322967 Fax: 01902 323 543
-------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/html Size: 11290 bytes Desc: not available URL: <https://mailman.uib.no/public/corpora/attachments/20140419/04ee9354/attachment.txt>