Special issue on Meaning in Context: ontologically and linguistically motivated representations of objects and events


Guest editors

Valerio Basile, Tommaso Caselli and Daniele P. Radicioni

***************

Dealing with context is a key factor in the conceptualization of human experience, and a major issue for understanding natural language. It is well known that some properties of objects and events may have different cognitive salience according to their context of occurrence, thus determining access to partial relevant information rather than to all information. One typical example is that of an orange being passed between two children, or the same orange peeled on a table: in the former case the roundness prevails over other traits, and the orange is being used to play; in the latter one, the edible features are those mostly conveyed by the scene. Interpreting events poses contextual challenges as well: (in how far) does a given event allow for different interpretations, like it might happen for revenge/self defense? Similar selectional mechanisms underlie figurative uses of word meanings, such as metonymy and metaphors among others, that intrinsically characterize the interface between knowledge and language.

Contextual access to objects and events needs to be further investigated, shared conceptualizations and terminologies are needed, as well as more robust approaches, including connections to domain and formal ontologies. The design of ontological and linguistic resources that account for the semantic phenomena involved in the contextual interpretation of objects and events requires collecting information and devising context-aware procedures.

In an era where most research is committed to statistical approaches, e.g. vector representations of the linguistic context and neural architectures, pairing the natural language semantic interpretation process and formal ontology may improve the inferential capacities of artificial agents with the explanatory power that is less relevant in those mainstream approaches.

Methods traditionally adopted to elaborate text documents exhibit limitations in representing and processing objects and events. Many efforts are being put in grasping text documents’ semantics based on semantically shallow approaches, whilst natural language inference demands for deep interpretation models, allowing to handle properties, functions, and roles, among others, to deal with commonsense and to produce explanations.

A different approach relies on lexical information: several large-scale lexical resources, such as WordNet (https://wordnet.princeton.edu), BabelNet (http://babelnet.org), FrameNet (https://framenet.icsi.berkeley.edu/fndrupal/), and ImagAct (http://imagact.lablita.it/index.php?lang=en), among others, have been proposed in the last few years and have been successfully employed to bridge the gap between knowledge representations and natural language. However, to cope with contextual access to objects and events involves many additional features still lacking in such resources. Neither shallow representations of NL semantics nor lexical resources alone provide sufficient ground to account for contextual phenomena.

Relevant areas include, but are not limited to: events representation and retrieval, event sequences, contextual features representation, trend detection, knowledge discovery, word sense disambiguation, ontology alignment, opinion mining and sentiment analysis, and conceptual similarity, among others. All proposed approaches must address the issue of representation of context, and suitable procedures to use context and context aware meaning representations of objects and events. The ideal submission should provide evidence that context improves the performance of systems on real-world applications and/or provides useful insights and explanations on systems’ output.

Topics of interest

Research works submitted to the special issue should foster scientific advances whether and to what extent objects and events representation and processing can be linked to the context where they occur. The following is a tentative list of relevant topics:

  • theoretical foundations for the use of AI techniques to deal with context and with changing/evolving objects and events
  • KR frameworks to represent mutable/evolving objects and events, including formal ontologies, conceptual spaces and distributed representations
  • formal methods for reasoning in evolving scenarios
  • theoretical, methodological, experimental, and application-oriented aspects of knowledge engineering and knowledge management centered on events and evolving objects
  • use cases and application scenarios (e.g., in law, medicine) where contextual information impacts on objects/events representation and processing
  • linguistic approaches to context analysis
  • context-aware lexical resources to describe objects and events
  • context-aware topic and event detection and tracking, knowledge discovery
  • context-aware frame semantics
  • entity linking and word sense disambiguation
  • representation of context in the Semantic Web
  • surveys on the adoption of contextual information in Cognitive Science, NLP and Ontological Modeling
  • context-based explainable Artificial Intelligence

Timeline

Manuscript Submission Deadline: July 23rd 2018
Acceptance Notification: November 26th 2018
Final Manuscript Due: February 26th 2019

Submission Guidelines

Submission guidelines can be found on the Journal Site: https://www.iospress.nl/journal/applied-ontology/?tab=submission-of-manu...

This special issue welcomes original high-quality contributions that have been neither published in nor submitted to any journals or refereed conferences. Extended versions of (properly referenced) conference papers should include at least 30% of new material. Please, clearly specify in the cover letter that the paper is to be considered for the special issue on "Meaning in Context: ontologically and linguistically motivated representations of objects and events."

Guest Editors

Valerio Basile, Sapienza University of Rome, Italy, basile at di.uniroma1.it
Tommaso Caselli, Rijksuniversiteit Groningen, The Netherlands, t.caselli at rug.nl
Daniele P. Radicioni, University of Turin, Italy, radicion at di.unito.it