TextGraphs-11: Graph-based Methods for Natural Language Processing
Workshop at ACL 2017, Vancouver, Canada, August 2017


The workshops in the TextGraphs series have published and promoted the synergy between the field of Graph Theory and Natural Language Processing. Besides traditional NLP applications like word sense disambiguation and semantic role labeling, and information extraction graph-based solutions nowadays also target new web-scale applications  like information propagation in social networks, rumor proliferation, e-reputation, language dynamics learning, and future events prediction, to name a few.

       >> Previous editions of the TextGraphs workshop

The eleventh edition of the TextGraphs workshop aims to extend the focus on issues and solutions for large-scale graphs, such as those derived for web-scale knowledge acquisition or social networks. We encourage the description of novel NLP problems or applications that have emerged in recent years, which can be addressed with graph-based methods. Furthermore, we also encourage research on  applications of graph-based methods in the area of Semantic Web in order to link them to related NLP problems and applications.


In an exciting and completely novel extension, we encourage graph-based interpretations of deep learning models for NLP tasks. Though deep learning models are displaying state of the art performance on many NLP tasks, they are often criticized for not being interpretable. In TextGraphs-11 we introduce a new challenge for graph-based methods: reasoning and interpretation of the layers used in deep learning models. Given that a neural network is, from one point of view, a graph through which activation scores are propagated, many of the existing graph-based methods used in our workshop community could potentially be  directly applicable. Can graph-based methods be harnessed to provide information to make deep processing comprehensible for humans and computers? What are the capabilities and limits when graph-based methods are applied to neural networks in general? Which aspects of the networks are not susceptible to such treatment and why not?


TextGraphs-11 invites submissions on (but not limited to) the following topics:

  • Graph-based methods for providing reasoning and interpretation of deep learning methods
    • Graph-based methods for reasoning and interpreting deep processing by neural networks,
    • Explorations of the capabilities and limits of graph-based methods applied to neural networks in general
    • Investigation of which aspects of neural networks are not susceptible to graph-based methods.    
  • 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 text summarization, simplification, and paraphrasing
    • Graph-based techniques for document navigation and visualization
    • Reranking with graphs
    • Applications of label propagation algorithms, 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
    • Graph kernels, etc.
  • Graph-based methods for applications on social networks
    • Rumor proliferation
    • E-reputation
    • Multiple identity detection
    • Language dynamics studies
    • Surveillance systems, etc.
  • Graph-based methods for NLP and Semantic Web
    • Representation learning methods for knowledge graphs (i.e., knowledge graph embedding)
    • Using graphs-based methods to populate ontologies using textual data,
    • Inducing knowledge of ontologies into NLP applications using graphs,
    • Merging ontologies with graph-based methods using NLP techniques.