Contextual Representation of Events and Objects in Language

CREOL@JOWO 2019


Artificial Intelligence



The Joint Ontology WOrkshops (JOWO, https://www.iaoa.org/jowo/2019/) is a venue of workshops that, together, address a wide spectrum of topics related to ontology research, ranging from Cognitive Science to Knowledge Representation, Natural Language Processing, Artificial Intelligence, Logic, Philosophy, and Linguistics.
### SCOPE
CREOL aims at gathering together researchers from different communities (Applied Ontology, NLP, AI, Semantic Web) to investigate the relationship between the semantic representation of objects and events and their context, from an ontological and linguistic perspective.
Dealing with context is a key factor in the conceptualisation of human experience, and thus a major issue for understanding natural language. It is well known that some properties of objects and events may be activated according to their context of occurrence, and thus determine selective access to salient aspects.
Current ontologies and lexicons (e.g. DOLCE [1], Unified Verb Index [2])) offer limited (meaning) representations of events and objects that then may find different realisations in text. Additionally, the growing interest towards multimodal information systems requires devising approaches and resources aimed at representing context by considering different sources (e.g., textual description, image, video) as a whole.
Contextual access to objects and events needs to be investigated at its interface with language and visual scene, as well. Recently, several advanced approaches have been proposed to model meaning representations of lexical items in their context (e.g., Deep contextualised word representations [3]). Likewise, approaches and resources have been designed to represent and make explicit the relations intervening between objects in scenes depicting events (e.g., Visual Genome [4]). Moreover, established theories of meaning representation allow for the representation of context to varying extent (e.g., Abstract Meaning Representation [5], Discourse Representation Theory [6]).
The goal of this workshop is to promote and discuss the design of ontologies, linguistic resources, and computational methods that treat context as a primary focus of interest.
[1] Gangemi et al. (2002) “Sweetening Ontologies with DOLCE”
[2] Trumbo (2006) “Increasing the usability of research lexica”
[3] Peters et al. (2018) “Deep contextualized word representations”
[4] Krishna et al. (2017) "Visual genome: Connecting language and vision using crowdsourced dense image annotations"
[5] Banarescu et al. (2013) “Abstract Meaning Representation for Sembanking”
[6] Kamp and Reyle (1993) “From Discourse to Logic”
### SUBMISSION
We solicit contributions of up to 8 pages plus references. All submissions must be in PDF format and must follow the IOS Press FOIS formatting guidelines, available at https://goo.gl/qkTpT7.
All accepted contributions will appear in the proceedings, which will be published by CEUR. (For previous editions of the JOWO proceedings, see www.iaoa.org/jowo/).
### APPLICATION AREAS
Contributions are solicited that cover a variety of topics, including but not limited to:
- ontologies of objects, qualities, and events;
- context-aware ontologies and ontologies of context;
- 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;
- linguistic approaches to context analysis;
- context-aware lexical resources to describe events;
- context-aware topic and event detection and tracking, knowledge discovery;
- context-aware frame semantics, and ontology-based frame design;
- context in typicality-based knowledge;
- contextual knowledge for eXplainable AI;
- formal ontologies and models for context-aware accountability;
- entity linking, keyword linking, word sense disambiguation;
- context-aware multimodal resources and applications;
- contextualised meaning representations (formal and distributional);
- use cases (such as Natural Language Processing, Computer Vision, Robotics) and application scenarios (e.g., in law, medicine) where contextual information impacts on concepts/events representation and processing.
### IMPORTANT DATES
- February 22, 2019 - First call for papers
- May 15, 2019 - Submission of contributions to workshops
- July 1 , 2019 - Workshop paper acceptance notification
- August 1, 2019 - Deadline for final camera ready
- September 23-25, 2019 - CREOL@JOWO
### WORKSHOP ORGANIZERS
Valerio Basile, University of Turin, Italy
Tommaso Caselli, University of Groningen, NL
Daniele P. Radicioni, University of Turin, Italy
Andrea Amelio Ravelli, University of Florence, Italy
### PROGRAM COMMITTEE
[TBA]