ISWC 2017

Hybrid Statistical Semantic Understanding and Emerging Semantics(HSSUES)


Held in conjunction with ISWC’17, October 21-22, 2017

International Workshop on Hybrid Statistical Semantic Understanding and Emerging Semantics (HSSUES)

Introduction

Understanding the semantics of Web content is at the core of many applications, ranging from Web search, news aggregation and machine translation to personal assistant services such as Amazon Echo, Cortana, Siri, and Google Home. Presently, two different approaches apply to this task. The first approach utilizes a rich suite of information retrieval and machine learning techniques that capture meaning through powerful statistical tools like neural networks and distributional semantics (e.g., word2vec). Recently, such emerging semantic models have achieved state-of-the-art results in several predictive applications (e.g. recommendation, node classification, knowledge graph completion) relevant not just to the Semantic Web, but allied communities such as data mining and natural language processing. The second, more traditional, approach conveys meaning in a structured form through embedded data markup (using Schema.org, OGP, etc.) and ontologies, and can be further enhanced through available knowledge bases such as Freebase and DBpedia. Ontological notions of semantics play a central role in the Semantic Web community, permitting publishers and consumers of data to interact using a well-defined ontology. Such models give practitioners the powerful ability to reason about, and represent, instances in a well-defined manner. HSSUES is a half-day workshop that will explore the synergy, from perspectives of theory, application, experiments (including negative results) and vision, between both approaches, and how such synergies can be exploited to advance the state-of-the-art. We are interested in mechanisms, both theoretical and experimental, that range the spectrum of possible strategies and provide novel functionalities through hybrid approaches to statistical and symbolic understanding. The broader goal is to foster a discussion that will lead to cross-cutting ideas and collaborations at a timely moment when Semantic Web research has significantly started intersecting with the natural language processing and knowledge discovery communities.

Format

We are looking forward to a highly interactive workshop with participants from a wide range of backgrounds, from researchers to application builders, and from academia to industry. Researchers with an interdisciplinary interest in exploring synergies between declarative semantics such as ontologies and machine learning and knowledge discovery methods, especially graph (including knowledge graph) embeddings and deep neural networks, are especially invited to participate and submit papers. In addition to oral presentations selected from submitted papers, we will have poster sessions, and an invited talk from industry.

CFP

Short and long papers are solicited for the following set of non-exhaustive topics:

Theory, Algorithms and Methods:


• Emerging semantic models e.g., distributional semantics, onto-distributional and other non-declarative or pseudo-declarative semantics
• Synergies between emerging and ontological semantics e.g., using semantic data and technologies to explain or improve distributional models, using distributional approaches to acquire knowledge bases
• Foundational proposals for content models that combine statistical and symbolic representations
• Novel embedding algorithms, especially for diverse data such as knowledge graphs, RDF, and ontologies
• Statistical machine learning methods and algorithms for symbolic representations

Applications

• Creating symbolic representations from machine learning
• Web search
• Question answering
• Personalization
• Data Mining
• User interfaces and visualization
• Semantic recommendations
• Link prediction
• Node classification
• Instance matching/Entity resolution
• Knowledge graph embeddings
• Knowledge graph completion

Experiments, Systems and Data


• Novel datasets, especially datasets acquired through, or useful for evaluating, hybrid approaches
• Novel methodologies, concerning both evaluations and data curation/collection
• Experimental results using existing methods, including negative results of interest
• Systems issues in hybrid systems, including best practices, case studies, lessons learned, and feature descriptions

We will also accept a small number of vision, opinion and position papers that provide discussions on challenges and roadmaps (for hybrid systems, and emerging semantic models).

All papers should be formatted according to the standard LNCS Style. All papers will be peer reviewed, single-blinded. Authors whose papers are accepted to the workshop will have the opportunity to participate in a poster session, and some set may also be chosen for oral presentation. Long papers should not exceed than 12 pages, and short papers should not exceed 6 pages, including all references. The accepted papers will be published online in CEUR-WS and will not be considered archival. Proceedings will be available for download after the conference.

All papers will be peer reviewed, single-blinded. Authors whose papers are accepted to the workshop will have the opportunity to participate in a poster session, and some set may also be chosen for oral presentation. Long papers should not exceed than 12 pages, and short papers should not exceed 6 pages, including all references. The accepted papers will be published online and will not be considered archival. Proceedings will be available for download after the conference.

We are using the EasyChair system for submissions. Please upload your submissions to our EasyChair submission system.

Please email any enquiries to kejriwal@isi.edu

Key Dates

May 6: First call for papers
June 21: Submissions open
August 6 (extended): Workshop papers due
August 24: Notification of accepted papers
September 21: Publication of workshop proceedings
Oct. 21 (9:00 AM-12:20 PM): Workshop

Accepted Papers

Flexible Similarity Search of Semantic Vectors Using Fulltext Search Engines. Michal Růžička, Vít Novotný, Petr Sojka, Jan Pomikálek and Radim Řehůřek. [Paper]

Related Fact Checks. Sreya Guha. [Paper]

Reconciling Event-Based Knowledge through RDF2VEC. Mehwish Alam, Diego Reforgiato, Misael Mongiovì, Aldo Gangemi and Petar Ristoski. [Paper]

Towards a Vecsigrafo: Portable Semantics in Knowledge-based Text Analytics. Ronald Denaux and Jose Manuel Gómez-Pérez. [Paper]

Statistical Semantic Classification of Crisis Information. Prashant Khare, Miriam Fernandez and Harith Alani. [Paper]

Evolution of Semantically Identified Topics. Victor Mireles and Artem Revenko. [Paper]

Partial Knowledge in Embeddings. Ramanathan Guha. [Paper]

Semantic Web: Learning from Machine Learning. Dan Brickley. [Paper]

Schedule

9:00: Introduction and Welcome

9:05: Oral Presentations I (17+17+8 minutes)
Flexible Similarity Search of Semantic Vectors Using Fulltext Search Engines. Michal Růžička, Vít Novotný, Petr Sojka, Jan Pomikálek and Radim Řehůřek. [Paper][Slides]
Towards a Vecsigrafo: Portable Semantics in Knowledge-based Text Analytics. Ronald Denaux and Jose Manuel Gómez-Pérez. [Paper][PPTX Slides][Slides]
Partial Knowledge in Embeddings. Ramanathan Guha. [Paper]

9:50: Keynote: Michael Witbrock (IBM)

10:30: Coffee Break

11:00: Oral Presentations II (17+17+17+17+8 minutes)
Related Fact Checks. Sreya Guha. [Paper]
Reconciling Event-Based Knowledge through RDF2VEC. Mehwish Alam, Diego Reforgiato, Misael Mongiovì, Aldo Gangemi and Petar Ristoski. [Paper]
Statistical Semantic Classification of Crisis Information. Prashant Khare, Miriam Fernandez and Harith Alani. [Paper][PPTX Slides]
Evolution of Semantically Identified Topics. Victor Mireles and Artem Revenko. [Paper][Slides]

12:15~12:20: Closing Notes

Keynote Speaker

Michael Witbrock (IBM) is currently a distinguished staff member at IBM Research, working at the intersection of learning and reasoning. Earlier, he was VP of Research at Cycorp, principal scientist at Lycos and systems scientist at Carnegie Mellon University.

Organization

Chair

Program Committee

Dan Brickley, Google
Derek Doran, Wright State University
Stefano Faralli, University of Manheim
Evgeniy Gabrilovich, Google
Majid Ghasemi-Gol, Information Sciences Institute/USC
Goran Glavaš, University of Manheim
Vineet Gupta, Google
Sabrina Kirrane, Vienna University of Economics and Business
Brigitte Krenn, Austrian Research Institute for Artificial Intelligence
Steve Macbeth, Microsoft
John McCrae, Insight Center for Data Analytics, National University of Ireland Galway
Vivek Raghunathan, Google
Enrico Santus, Singapore University of Technology and Design
Michael Spranger, Sony Computer Science Laboratories Inc.
Srinivasan Venkatachary, Laserlike Inc.
Jiewen A. Wu, Accenture