Semantics 2017 - Keynote

Chris Welty

semantics_admin — Thu, 17 Aug 2017 07:34:52 +0000

DBPedia Keynote - Senior Research Scientist at Google

Even the changes are changing: A new age of cognitive computing

Watson, the computer that achieved prominence in 2011 by defeating the all time best players of the American TV quiz show, Jeopardy!, was a herald of a new age. Suddenly it was possible for machines to understand us, in our own language. In the six years since, new AI and cognitive capabilities for machines are being announced at a dizzying pace, yet in the 60 years before Watson, AI moved comparably slowly. There are a few simple factors that have helped move us into the cognitive computing age, such as availability of data and the power of computation, but there is a deeper reason. Artificial intelligence itself has had to change, from a view of machines as perfect rational thinkers, to an understanding that cognition is intimately tied to perception, which is imperfect, and that therefore knowledge and reasoning is inherently subjective and ambiguous. Once we accept this, we can continue changing.

CV

Dr. Chris Welty is a Sr. Research Scientist at Google in New York, and the Endowed Chair of Cognitive Computing at the VU University, Amsterdam. His main area of interest is using structured semantic information to improve semantic processing of unstructured information, such as using freebase to help improve web search. His latest work is on using crowdsourcing to form a new theory of truth based on diversity of perspectives, and he was part of the team that recently launched Explore in Docs.

Before Google, Dr. Welty was a member of the technical leadership team for IBM's Watson - the question answering computer that destroyed the all-time best Jeopardy! champions in a widely televised contest. He appeared on the broadcast, discussing the technology behind Watson, as well as many articles in the popular and scientific press. His proudest moment was being interviewed for StarTrek.com about the project. He is a recipient of the AAAI Feigenbaum Prize for his work.

Welty was one of the first to call attention to the new paradigm of Cognitive Computing that has emerged in computation, and he previously played a seminal role in the development of the Semantic Web and Ontologies, and co-developed OntoClean, the first formal methodology for evaluating ontologies. He is on the editorial board of AI Magazine, the Journal of Applied Ontology, the Journal of Web Semantics, and the Semantic Web Journal.

Sandeep Sacheti

semantics_admin — Mon, 31 Jul 2017 15:09:36 +0000

Executive Vice President, Customer Information Management & Operational Excellence (EVP, CIOx)

Wolters Kluwer

Keynote

How Wolters Kluwer’s AI aided human experts deliver increasing accuracy, speed and value for our customers

CV

Mr. Sandeep Sacheti has been Executive Vice President of Customer Information Management and Operational Excellence for Wolters Kluwer Governance, Risk & Compliance at Wolters Kluwer Financial Services, Inc. since June 2, 2016. Mr. Sacheti serves as the Vice President of Corporate Legal Services - New York at Wolters Kluwer N.V. Mr. Sacheti has been the Executive Vice President of Strategy and Customer Insights at Wolters Kluwer Corporate Legal Services since May 22, 2012. A well-known pioneer in data analytics, Mr. Sacheti has spent more than 15 years in customer-driven research and operations and has held a number of senior risk management, decision science and operational positions in financial services companies during that time. He served as the head of U.S. banking, business risk and analytics for UBS. He served in a number of roles at American Express, including vice president, risk management. He holds more than 20 patents in the areas of information management, customer relationship management, decision management and fraud detection and has published extensively in top academic journals. Mr. Sacheti has a master's degree in resource economics from the University of Massachusetts and a Ph.D. in Agricultural and Resource Economics from the University of California at Berkeley.

Keynote Speakers

semantics_admin — Tue, 27 Jun 2017 14:36:10 +0000

Keynote

Time:

Tuesday, September 12, 2017 - 09:00 to Wednesday, September 13, 2017 - 17:00

Place:

The Meervaart

Talks

Miel Hostens

Data Scientist at Department of Reproduction, Obstetrics and Herd Health

Ghent University: Department of Reproduction, Obstetrics and Herd Health

Keynote

Will semantics help disentangle the Gordian knot of Big Data in animal health

Nathalie Aussenac Gilles

CNRS senior researcher at Toulouse Institute of Computer Science Research

Institut de Recherche en Informatique de Toulouse

Keynote

New convergences between Natural language processing and knowledge engineering at the era of the Semantic Web.

An illustration with the extraction and representation of semantic relations

Utz Westermann

Senior Data Architect at Otto Group

Otto Group

Keynote

Buoyancy in Data Lakes - Agile Metadata Management in Hadoop Data Warehouses at Otto Group

Current metadata and information about data lineage are crucial for understanding and interpreting data in a Hadoop data warehouse. At the same time, Hadoop data warehouse projects sink or swim with the ability to continuously add new data sources and views as business requirements evolve.

Ger Baron

Chief Technology Officer at City of Amsterdam

City of Amsterdam

Industry

Aaron Bradley / Eamonn Glass

Senior Manager, Web Channel Strategy / Director, Online Platform Strategy

Electronic Arts

Keynote

Semantics at Play: Electronic Arts’ Linked Data Journey

Aaron Bradley / Eamonn Glass

semantics_admin — Thu, 01 Jun 2017 15:30:18 +0000

Senior Manager, Web Channel Strategy / Director, Online Platform Strategy

Electronic Arts

Keynote

Semantics at Play: Electronic Arts’ Linked Data Journey

Millions of players consume or create content related to EA’s video games every day. For EA, managing this content using traditional content management systems and databases had become increasingly difficult. The environments in which this content was created, stored and consumed kept growing in number and complexity. EA’s ability to reliably connect individual players with the content that mattered most to them remained elusive.

In response EA has set about changing its approach to the creation, classification and distribution of digital content. This has entailed structurally modelling and semantically describing the content in EA’s ecosystem. This approach has allowed EA to forge an entity-based understanding of available content to provide the right information to the right player at the right time.

In this talk Aaron and Eamonn will describe how they were able to initiate this journey at EA, discuss challenges and how they overcame them, share initial learnings, and look at how linked data will play a key role in the EA player network.

CV

Eamonn Glass
Director, Online Platform Strategy

Eamonn Glass has been leading teams at the intersection of technology, business and user needs for over 20 years. He has worked with start-ups and Fortune 500 companies in ecommerce, health care, business intelligence, retail, and utilities industry sectors. Eamonn now leads a team at Electronics Arts, merging his lifelong passion for gaming with his drive to create intelligent systems that deliver great customer experiences.

Aaron Bradley
Senior Manager, Web Channel Strategy

Aaron Bradley has been helping companies optimize their online presence for over a decade, including work in the ecommerce, education, publishing and entertainment sectors. A compulsive categorizer and one-time student of semiotics, Aaron's previous work as a web designer and technical services librarian have helped inform his approach to search engine and digital presence optimization, which has been increasingly focused on leveraging the power of structured data and knowledge graphs to help better connect businesses with their customers.

Utz Westermann

semantics_admin — Mon, 27 Mar 2017 08:12:19 +0000

Senior Data Architect at Otto Group

Otto Group

Keynote

Buoyancy in Data Lakes - Agile Metadata Management in Hadoop Data Warehouses at Otto Group

Conventional ETL workflow schedulers and metadata management approaches prove millstones round a project's neck, however:

with every rollout of new views significant effort has to be invested in one-time migration scripts for schema and data;
manual data documentation is laborious and gets out-of-sync quickly; and
automatic metadata derivation approaches are invasive, resource-taxing, and usually yield a perspective too technical for business.

The talk proposes integrated specification of data structure, data dependencies, and computation logic as a way to keep ETL productive and metadata current.

Based on such specifications, a scheduler can automatically detect changes and perform

appropriate schema migrations and
data recomputations automatically as necessary, significantly reducing rollout and operations effort.

Also, the very same specifications explicitly "program" rich metadata, avoiding

additional manual documentation labor,
automatic metadata derivation overhead and its low semantic level, greatly simplifying the implementation of metadata exploration tools.

The talk illustrates this approach with Schedoscope, a scheduler developed at Otto Group based on integrated view specification, and Metascope, a collaborative metadata exploration tool built on top of Schedoscope. Schedoscope and Metascope drive Otto Group BI's data platform, which processes clickstream, product, and CRM data from 120 online shops with a yearly revenue north of 5bn Euros. Schedoscope has enabled Otto Group BI's small team of data engineers to continuously release new data sources and view for more than 2 years now; with Metascope, Otto Group's analysts and data scientists have access to always up-to-date metadata and documentation.

Schedoscope and Metascope are available as open-source at http://schedoscope.org

CV

Utz Westermann is Senior Data Architect at Otto Group BI, Hamburg, with 18 years of experience in large-volume data processing.

He is the tech lead for Otto Group BI's data platform, which processes clickstream, product,and CRM data from 120 online shops.

Utz started out in academia researching multimedia databases, receiving a doctoral degree with distinction from Technical University of Vienna in 2004. After postdoctoral visits at VTT Oulu and UC Irvine, he left academia in 2006 to work as a technical consultant on EAI and as CTO of a startup developing a SEO tool monitoring millions of search rankings prior to joining Otto.

Utz has published regularly in peer-reviewed international scientific journals and presented at scientific conferences. Utz is the maintainer of the Schedoscope open-source project.

Nathalie Aussenac Gilles

semantics_admin — Fri, 24 Mar 2017 10:37:45 +0000

CNRS senior researcher at Toulouse Institute of Computer Science Research

Institut de Recherche en Informatique de Toulouse

Keynote

New convergences between Natural language processing and knowledge engineering at the era of the Semantic Web.

An illustration with the extraction and representation of semantic relations

Design the theories and systems required to process natural language (either to "understand" written or oral human language or to generate new text), or to perform reasoning and solve problems, have been two main research lines right from the beginning of AI. Today, they still form two separate research communities: Natural Language PRocessing (NLP) and Knowledge Engineering (KE). Although these activities are closely related in human cognition and activities, they have long been studied as two separate tasks that share a common background in knowledge representation and logic. Various limitations of knowledge based systems led AI research to investigate alternative approaches, and to privilege efficiency and user assistance when performing a task rather than cognitive likelihood and full autonomy of the systems. As a consequence, although early works in NLP and in KE relied on symbolic representations and algorithms strongly inspired by human behaviour, research studies are now involving a larger variety of techniques, including numeric machine learning, data analysis, constraint programming and planning, so that intelligent systems no longer mimic human performance.

A parallel movement accompanies this evolution, which faces a (natural) stronger convergence between NLP and KE. On the one hand, analyzing language requires to represent what has been understood and to reproduce some of the reasoning and tasks carried out when dealing with language, like solving references, understanding the often implicit relations between sentences, disambiguating the meaning of words regarding their context, recognizing entities, opinions, inferring more knowledge about the situation etc. On the other hand, building knowledge bases acquired from human expertise relies on language, thich conveys most of the required knowledge. This process has a high cost that can be reduced by extracting part of this knowledge from text. Worse, in many situations, it may not be relevant to refer only to persons when part of the knowledge is either tacit or already explicit in documents. This convergence has been stimulated by the availability of larger sets of data, huge amounts of textual documents in companies and most of all on the web. Convergence increased with the advance of mining algorithms in information extraction on the one hand, and fine grain understanding of natural language brought by linguistics and mainly computational linguistics, where corpus-based approaches proved to be very efficient to better understand language. Other influences and advances from research in terminology and knowledge representation or machine learning draw a new landscape where the frontiers between these "disciplines" are blurred. Current intelligent applications require to look for information on the basis of its meaning, its content and meta-data, which combines information retrieval and language analysis, but also reasoning based on back-ground knowledge, which means an access to a large knowledge base. Knowledge rich applications require language understanding and production to explore more knowledge sources or to better interact with the user and so on. In short, the notion of "semantics" that is shared by the two research communities is central, and the outcome of the Semantic Web is a new opportunity to investigate these topics.

Indeed, the Semantic Web promotes to better use the large resources available on the web, in particular thanks to machine learning, data classification or clustering and other mining techniques. The Semantic web working groups integrated many previous work on knowledge representation and reasoning to define standard formats that make it easier to share, store, query, compare, link, analyse or process knowledge and data, whether it be textual or numerical. This context is favorable to strengthen the relations between the studies about language and knowledge, encompassing knowledge representation; knowledge storage and linking; building, sharing and reusing semantic resources; using language to query or to exchange data; learning knowledge from text and data to better understand, process or produce language ...

The extraction of semantic relations is one of the research challenges in NLP and KE currently impacted by the Semantic Web. Semantic relations are a nice case-study not only because extracting semantic relations from text is still considered as a difficult task, but also because linking data on the Semantic Web has turned it into one of the important challenges to be addressed, and finally because the evolution of the extraction techniques is a nice example of the shift occurring when exploiting the web. These techniques evolved from linguistically grounded pattern based approaches matched on domain specific corpora to deep learning methods applied to the whole web. An unexpected feature of this impact is a shift in the disciplines collaborating to this field: whereas NLP and KE researchers used to collaborate with linguists and terminologists, they now work together with machine learning researchers and mathematicians. So far, there is not yet a cross-disciplinary fertilization of these two trends. For instance, the large efforts carried out in terminology and in computational linguistics to evaluate, define, improve pattern-based approaches are not much taken into account in machine learning for NLP or ontology learning. In a symmetric manner, relation extraction using machine learning has still little impact on linguistic analyses and when designing lexical resources.

In my talk, I will first sketch the historical relations between NLP and KE, highlighting research challenges at their confluence. I will then focus on semantic relations to illustrate how the vision, formats and approaches proposed by the Semantic Web impact several dimensions of this research topic: their extraction from text, their representation in knowledge bases and lexical resources as well as their use for richer semantic annotations. Thus I will draw an overview of the current approaches to identify semantic relations in text, with a special focus on the ways pattern-based solutions have evolved since early works, on machine learning methods and on the complementarity of various techniques to support this task. I will then report several representation models for semantic relations, that associate the conceptual part with linguistic information and lexical entries. I will finally present some models for a rich annotation including relations.

I will conclude by stressing the need to capitalize better all the experiments and tools developed up to now, in particular by sharing not only representations and data, but also patterns and learning methods, and also by investigating more systematically how existing techniques can be used together in a single platform, and mutually benefit of each other's results.

CV

Nathalie Aussenac-Gilles is researcher at IRIT, the laboratory in COmputer Science of Université Paul Sabatier in Toulouse, where she hired as a CNRS junior researcher from 1991 to 2010 and as a CNRS senior researcher since then. She heads the MELODI group (http://www.irit.fr/-Equipe-MELODI- ) with P. Muller and coordinates activities related to Big Data and Data processing. He research deals with knowledge engineering (KE), semantics in natural language processing and methods to build and populate knowledge bases and ontologies. Her work is influenced by long term cross-disciplinary collaborations with linguists and researchers in human factors. She designed methods and platforms for ontology engineering and evolution, tools for semantic relation and concept extraction from text, as well as on knowledge representation to include lexical data. She has been task manager in various French ANR and FUI projects with applications in areas as diverse as archaeology, car diagnosis, cartography, earth observation or scholar paper analysis. She chaired the GRACQ working groups (1988-2005) and later the TIA SIG (2005-2010) that gathered French researchers in KE and terminology for KE and AI. She was involved in the organization or chairing of conferences (EKAW, ESWC, TIA, IC, …) and she belongs to the editorial boards of scientific journals (formerly RevueI3 and IJHCS, currently Applied Ontology). She supervised 15 PhD. thesis and published more than 80 papers in Journals and conferences or workshops related to semantic web and knowledge engineering.

Miel Hostens

semantics_admin — Fri, 24 Mar 2017 10:27:19 +0000

Data Scientist at Department of Reproduction, Obstetrics and Herd Health

Ghent University: Department of Reproduction, Obstetrics and Herd Health

Keynote

Will semantics help disentangle the Gordian knot of Big Data in animal health

Since the 1950’s, computers have been used as a management tool in dairy farming. Over subsequent decades, dairy herd management software has evolved consistently and the personal computer has emerged as an important management tool to primarily monitor production, reproduction and health. In the meantime technologies to collect and store data have been evolving at a quicker pace compared to the speed at which new insights in animal science have been discovered. The exponentially increased volume and speed at which data is created in the post-dotcom decade, is commonly referred to as Big Data. One of the drivers of excitement around Big Data has been the expectation that we will be able to discover new insights in order to support decision making in animal health as ultimate goal. Albeit, integrated information and especially decision making support tools in the field are lacking. It illustrates that while the technology has been widely accepted and published in animal science, the effective implementation and validation is often minimal. Structural file format or database heterogeneity, syntax heterogeneity, implementation heterogeneity and semantic heterogeneity restrict proper use of real-world animal data. The use of ontologies (i.e. controlled vocabularies) is a means to make scientific descriptions more comparable. In dairy cows for example, most ontologies and trait definitions originate from milk recording organizations. At present, 30 to 40 mainly production related traits are commonly recorded in dairy cattle and used in selection schemes in many developed countries. Over the last decades, fertility and health-related phenotypes have been implemented with varying success. One of the main challenges in animals health is that even elementary concepts often have no canonical form which would be the preferred notation that encapsulates all equivalent forms of the same concept (eg. disease or health event definition). The semantic web could offer a common framework that allows data to be shared and reused across canonical form, application, enterprise, and community boundaries. Multiple technical and business challenges unique to the animal domain can be identified and need to be addressed before the semantic web can grow from a conceptual towards an applied framework in animal health management.

CV

Miel Hostens currently holds a Post-Doc assistant position at the ambulatory clinic of the Department of Reproduction, Obstetrics and Herd Health at the Faculty of Veterinary Medicine of Ghent University. He focusses on the optimization of productive and reproductive performances in small and large dairy herds with an emphasis on nutrition. He is one of the main data analysts for multiple work packages within the EU Framework 7 project www.GplusE.eu. Furthermore, he is actively involved in the education of Master students in Veterinary Medicine and statistical training of Ph.D. students. Besides that he is involved in post academic and extension services in the area of herd health management in dairy cows.