VocabControl

I will probably be on the other side of the Atlantic when the ISKO UK conference takes place in July in London, UK. I will be sorry to miss it, because the committee have brought together a diverse, topical, and fascinating collection of speakers.

ISKO UK excels in unifying academic and practitioner communities, and the conference promises to investigate the barriers that separate research from practice and to seek out boundary objects that can bring the communities together.

This is demonstrated in person by the keynote speakers Patrick Lambe of Straits Knowledge and Martin White of Intranet Focus Ltd - both respected for their commercial as well as academic contributions to the field of Knowledge Organization.

Amidst what is already shaping up to be a very full and varied programme, the presentations by Jeremy Tarling and Matt Shearer (BBC News) and Jarred McGinnis and Helen Lippell (Press Association) will show how research in semantic techniques is now being put to practical use in managing the fast-flowing oceans of information that news organizations handle.

The programme also includes a whole session on combining ontologies with other tools, as well as papers on facet analysis and construction of controlled vocabularies. There’s even some epistemology to please pure theoreticians.

David Riecks of Controlled Vocabulary gave a presentation about indexing images. He pointed out that metadata is all around us, but we don’t tend to notice it. He described the sort of metadata needed to make an asset “smart” and how organizations like the PLUS registry are attempting to provide a simple, one-stop shop for rights and licensing metadata. The Embedded Metadata Manifesto sets out details of metadata that needs to be included in image files to promote easy and legal re-use of content and so protect the rights of photographers and others in the content creation and related industries.

David also provided an extremely useful list of metadata resources , including a handy link to a website that checks whether metadata is being stripped from files at the point of upload.

Laura Fu talked us through the latest Digital Asset Management (DAM) implementation at Randall Marcinko of Marcinko Enterprises Inc. then talked about using different elements of assets to act as indexing mechanisms. He gave an example of where they were able to use the images associated with pieces of text as disambiguators to distinguish between the text. He also pointed out the dangers of trying to make every information project the same, and to think carefully about what is needed. It is easey to fall into the trap of simply offering all clients the same solution, whether that works best for them or not. Depending on what you are trying to achieve, a simple list is all that is needed, not a complex taxonomy or thesaurus, and the simpler the method of solving a problem, the easier and cheaper it is likely to be to implement.

Last month I had a wonderful time at the SLA (Special Libraries Association) conference in Chicago. I had never previously been to an SLA conference, even though there is a lively SLA Europe division. SLA is very keen to be seen as “not just for librarians” and the conference certainly spanned a vast range of information professions. The Taxonomy Division is thriving and there seem to be far more American than British taxonomists, which, although not surprising, was a pleasure as I don’t often find myself as one of a crowd! The conference has a plethora of receptions and social events, including the “legendary” IT division dance party.

There were well over 100 presentation sessions, as well as divisional meetings, panel discussions, and networking events that ranged from business breakfasts to tours of Chicago’s architectural sights. There was plenty of scope to avoid or embrace the wide range of issues and areas under discussion and I focused on taxonomies, Linked Data, image metadata, and then took a diversion into business research and propaganda.

I also thoroughly enjoyed the vendor demonstrations, especially the editorially curated and spam-free search engine Blekko, FastCase, and Law360 legal information vendors, and EOS library management systems.

My next posts will cover a few of the sessions I attended in more detail. Here’s the first:

Adding Value to Content through Linked Data

Joseph Busch of Taxonomy Strategies offered an overview of the world of Linked Data. The majority of Linked Data available in the “Linked Data Cloud” is US government data, with Life Sciences data in second place, which reflects the communities that are willing and able to make their data freely and publicly available. It is important to keep in mind the distinction between concept schemes - Dublin Core, FOAF, SKOS, which provide structures but no meanings - and semantic schemes - taxonomies, controlled vocabularies, ontologies, which provide meanings. Meanings are created through context and relationships, and many people assume that equivalence is simple and association is complex. However, establishing whether something is the “same” as something else is often far more difficult than simply asserting that two things are related to each other.

Many people also fail to use the full potential of their knowledge organization work. Vocabularies are tools that can be used to help solve problems by breaking down complex issues into key components, giving people ways of discussing ideas, and challenging perceptions.

The presentation by Joel Richard, web developer at the Smithsonian Libraries, focused on their botanic semantic project – digitizing and indexing Taxonomic Literature II. (I assume they have discussed taxonomies of taxonomy at some point!) This is a fifteen-volume guide to the literature of systemic botany published between 1753 and 1940. The International Association for Plant Taxonomy (IAPT) granted permission to the Smithsonian to release the work on the web under an open licence.

The books were scanned using OCR, which produced 99.97% accuracy, which sounds impressive but that actually means 5,000-12,000 errors – far too many for serious researchers. Errors in general text were less of a concern than errors in citations and other structured information, where – for example, mistaking an 8 for a 3 could be very misleading. After some cleanup work, the team next identified terms such as names and dates that could be parsed and tagged, and selected sets of pre-existing identifiers and vocabularies. They are continuing to look for ontologies that may be suitable for their data set. Other issues to think about are software and storage. They are using Drupal rather than a triplestore, but are concerned about scalability, so are trying to avoid creating billions of triples to manage.

Joel also outlined some of the benefits of using Linked Data, gave some examples of successful projects, and provided links to further resources.

In the last two weeks I have attended three very different conferences, with DAM as the common thread. The first was Media Pro Expo, where I spoke on a panel with the DAM Foundation, alongside Mark Davey, Madi Solomon, and David Lipsey. The second was Createasphere’s first European DAM conference, and the third (co-located with the Createasphere event) was the SPAR Europe Conference on 3D Imaging and Data Management for Engineering, Construction, Manufacturing, and Security.

The contrast between Media Pro and SPAR, and their respective audiences, was striking, but so were the similarities of the problems they faced, such as the common need to manage rich media assets and huge volumes of data. Media Pro was aimed at marketing companies, and had lots of amusing exhibits showcasing ways of using technology to create engaging and entertaining campaigns. (I enjoyed playing with an interactive magazine cover linked to a camera that allowed you to put your picture “on the cover” and select your favourite headlines.) Marketing companies are concerned with keeping, curating and mining data not just about customers’ contact details, but also their likes, social connections, and shopping habits in order to create personalised campaigns, so they have become great consumers of metadata.

3D Imaging and Data Management

SPAR was all about scanning and mapping, not in the sense that I am familiar with, but literally surveying the Earth and making maps. There were companies that use lasers to create roadmaps, others that carry out aerial surveys, and some that create 3-D representations of buildings. There are systems for surveying and modelling building sites to make sure that construction avoids sewers, pipes, and underground cables, and even a system for creating 3-D photosets of crime scenes to help the police in investigation and evidence gathering.

Createasphere

At Createasphere I talked about managing metadata in complex information environments and how we need to treat metadata as content in its own right. There were a range of excellent and diverse presentations, covering topics from the potential of immersive virtual worlds and the huge volumes of data they produce, to descriptions of technical metadata exchange projects.

I began to think about the crossover point between the creativity and imagination of the media and marketing companies and the power and accuracy of the surveying companies and how this is going to bring about hugely powerful fantasy “Holodeck” worlds that will make Second Life and the Sims look quainter than the Mickey Mouse cartoons of the 1930s.

Better than the real world

One challenge for information professionals is to think about how we can create navigation and search systems that do more than just replicate the real-world paradigms we are used to at the moment - I am thinking of things like road signs and timetables - but how to harness the best of semantic techniques and data mining processes to create reactive intuitive worlds that work better than the real one. Ed Lantz of Vortex Immersion Media spoke of “intelligent spaces” that automatically access our data, our assets, information about us, and arrange themselves to suit us. How do we prepare for a world when the likes of Apple’s speech recognition system Siri aren’t genies in bottles, but are the environment around us? We used to worry about ghosts in the machine, but will we end up as the ghosts inside the machine? We worry about putting our assets out there into the cloud, but perhaps we should be thinking more about what it will be like when we step inside the cloud or bring the cloud into our homes?

There was a post circulating on Twitter recently describing the library of the future as a hellish place where characters from books come alive and stalk the readers in the rooms. It was somewhat derided as a childish joke, but if we create Holodecks and then try to live in them, it could well come true. The implicit warning it contains that we could inadvertently trap ourselves in such a hellish place where privacy, rights, control, and manipulation are so hidden from view that we lose our sense of self seems to be very mature and insightful. Another post I read was about how interface designers are currently working on “pictures under glass” and need to start to use the full tactile, haptic, and 360 degree expressivity of our physical bodies, such as we are beginning to with technologies like the Wii and Kinect.

Making work fun

Theresa Regli of the Real Story Group pointed out that the world we are in now is one in which people still don’t grasp the importance of labelling their images, so immersive virtual worlds seem a long way off, but she also talked of the need for corporate interfaces to embrace “gamification”, as employees are far more productive when their jobs are fun. It may take some time, but I like the idea of a Holodeck meeting room where people make presentations and collaborate on plans by dancing around, rather than sitting staidly at a table. Rather than the hellish library where AI brings fictional monsters to life, it might turn out to be a lot of fun and all that movement may even be good for our health!

This post is the last in a series about the UDC consortium international seminar in The Hague, 19-20 September, 2011

Joan S. Mitchell, OCLC (USA), and Marcia Lei Zeng, Kent State University (USA), supported by Maja Žumer, University of Ljubljana (Slovenia), talked about extending models for controlled vocabularies to classification systems: modelling DDC with FRSAD, which led to interesting discussions about their concepts of “nomen” and “thema”.

Along with my former colleague Andy Heather, now CTO at DODS Parliamentary Communications Ltd, I talked about our work on the data migration of classifications from a legacy database into new taxonomy management software, presenting our paper: Transformation of a legacy UDC-based classification system: exploiting and remodelling semantic relationships.

Conclusions

The key ideas I took away from the conference were:
1) Classifications and ontologies are not an either/or choice. They have different properties and different strengths and weaknesses and so should be chosen according to the task in hand.
2) It is difficult to turn a classification into an ontology, but easy to turn an ontology into a taxonomy, so if you don’t have either to start with and can’t decide, an ontology is a safer bet. If you already have a classification, you need to think carefully about whether it is worth turning it into a fully modelled ontology, as converting it to RDF or SKOS is likely to be much easier. However, at the moment, RDF and SKOS have limitations, especially in handling faceted taxonomies, so beware of losing semantic richness in the conversion process. Polyhierarchies offer a way of expressing facets in SKOS.
4) Vocabulary control and alignment continue to be significant issues for the Semantic Web.
5) Ontology curation, management, and semantic alignment will be increasingly important issues for the Semantic Web.

Slides and audio recordings of all 21 talks can be now downloaded from the conference website.

Conference proceedings are published by Ergon Verlag and can now be
purchased/ordered online from http://seminar.udcc.org/2011/php/proceedings.php.

Day two of the UDC consortium international seminar opened with two subject-specific talks – Wolfram Sperber described a classification of mathematics and Andrew Buxton showed how similar chemistry classification and ontologies are, using the ChEBI ontology. He also described the different ways classifications and ontologies could be used to support each other and about the lack of good graphical tools and visualisations to represent ontologies.

Categories and relations: key elements of ontologies - Categorial Distinctions

Roberto Poli, University of Trento (Italy) talked about the compliexisties of part-whole relationships. There are simple wholes, composed of a sum of their parts, but some parts of wholes cannot simply be added together – for example, the social, psychological, and physical aspects of a person. He also discussed the difference between science as epistemological – dealing with what can be known – and ontological – deraling with what exists.

Towards a relation ontology for the Semantic Web

Dagobert Soergel made a bold claim that the only way for the Semantic Web to deliver its promise is if we adopt a relation ontology and map each dataset to the standard, to allow interoperability. He pointed out that you “do not getting semantics from syntax alone”.

Relations in the notational hierarchy of the Dewey Decimal Classification

Rebecca Green from OCLC described the difficulties encountered when trying to automatically create ontologies from the Dewey Decimal Classification. These included semantic differences in the way subclasses had been defined, meaning that no single rule would handle them all appropriately.

Modelling concepts and structures in analytico-synthetic classifications

The eminent Ingetraut Dahlberg compared Aristotle and Ranganathan’s key facets and UDC and Colon Classification systems. She also presented a survey of academic subject areas analysed into facets.

Representing the structural elements of a freely faceted classification

Claudio Gnoli of the University of Pavia, talked about freely faceted classifications, in comparison with systems such as UDC. He emphasised the urgency of publishing classifications on line, but highlighted the limitations of SKOS and OWL to fully expressed faceted systems despite the fact that faceted systems are extremely good tools for obtaining precise search results. Faceted systems are also excellent for combining information across disciplines, allowing you to combine aspects of one subject areas with aspects of a different one, and interdisciplinarity is becoming increasingly important as an approach, as innovation often happens at the boundaries between disciplines.

He pointed out that a polyhierarchical approach can be modelled in SKOS as a way of representing facets, but that this approach is often overlooked. He also called for more work to be done on SKOS so that it can represent facets directly.

Facet analysis as a tool for modelling subject domains and terminologies

Vanda Broughton, University College London, offered the Bliss Classification as a useful tool for online subject classification, but called for help in how best to publish it for general use. Should it be released as a text document, database, or should work be done to convert it to an ontology – and if so, in what form?

She stressed how the logical approach of facet analysis and regular syntax makes it predictable and hence ideal for machine manipulation.

Analytico-synthetic approach for handling knowledge diversity in media content analysis

Devika P. Madalli, Indian Statistical Institute, DRTC (India), described the Living Knowledge project that used an analytico-synthetic approach in order to bring together around useful themes diverse content from different sources using varied means of expression. This supported a rich faceted search system.

Slides and audio recordings of all 21 talks can be now downloaded from the conference website.

Conference proceedings are published by Ergon Verlag and can now be
purchased/ordered online from http://seminar.udcc.org/2011/php/proceedings.php.

This post is 4th in a series about the UDC consortium international seminar in The Hague, 19-20 September, 2011.

Interoperability of knowledge organization systems with and through ontologies

Daniel Kless from the University of Melbourne pointed out that problems with ontologies arise when combining them, as errors in combination can have disastrous effects on subsequent reasoning. A well-defined modelling method is needed to minimise this. Standards such as OWL and RDF do not address the problems of methodology or terminology control.

Towards the integration of knowledge organization systems with the linked data cloud

Vincenzo Maltese of the University of Trento, Italy, explained how it is vital to make clear the semantics and purpose of any ontology when attempting to share Linked Data. Ontologies may differ in their scope, purpose, structure, terminology, language, coverage, formality, and conceptualization. He drew a distinction between descriptive ontologies and classification ontologies. It is very easy to convert a descriptive ontology to a classification ontology and the process can be automated, but extremely difficult to convert a classification ontology to a descriptive one and the process requires human intellectual and editorial effort.

Classification and reference vocabulary in linked environment data

Joachim Fock of the Federal Environment Agency (Germany) talked about how they transformed their keyword thesaurus to a Linked Data format.

This post is the third in a series about the UDC consortium international seminar in The Hague, 19-20 September, 2011.

Approaches to providing context in knowledge representation structures

Barbara Kwasnik, Syracuse University (USA), talked about ways that context can be used as a disambiguation tool, and described different kinds of contexts: warrant, scientific, educational, cultural, etc. However, interdisciplinary approaches can be difficult. It is easy to have different ontological commitments, but you need a mapping to know when and which bits need to work across domains. Ontologies will need updating as the world and world views shift and change, so we need ways of defining their scope, as well as provenance and mappings. There are also difficulties in establishing the neutrality of ontologies.

Interaction between elementary structures in universes of knowledge

Richard P. Smiraglia, University of Wisconsin (USA),
talked about how people want to turn the multidimensional world into a unidimensional top-down model. He pointed out that people tend to assume UDC is like Dewey, but it actually works far more like Ranganathan’s Colon Classification. He called for new theories of organizing knowledge in shifting contexts and theories about how to mediate between concepts and structures like UDC.

Demystifying ontology

Emad Khazraee, Drexel University (USA), talked about how ontological approaches are as old as literature itself, showing a picture of what I think was the ancient Sumerian king list. He talked about boundary objects and the overlap between different academic areas that are interested in knowledge organisation and learning. He also discussed the differences between ontology-as-categorial-analysis and ontology-as-technology.

I thoroughly enjoyed the third biennial International UDC Consortium seminar at the National Library of the Netherlands, The Hague, last Monday and Tuesday. The UDC conference website includes the full programme and slides and the proceedings have been published by Ergon Verlag.

This is a first of a series of posts covering the conference.

Aida Slavic, UDC editor-in-chief, opened the conference by pointing out that classification is supposed to be an ordered place, but systems and study of it are difficult and complex. We still lack terminology to express and discuss our work clearly. There is now an obvious need to instruct computers to use and interpret classifications and perhaps our work to make our classifications machine readable will also help us explain what we do to other humans.

On being the same as

Professor Patrick Hayes of the Florida Institute for Machine Learning and Cognition delivered the keynote address, pointing out that something so simple as asserting that one thing is the same as another is actually incredibly difficult and one of the problems facing the development of the Semantic Web is that people are asserting that two things are the same when actually they are merely similar.

He explained that the formalisms and logic underpinning the Semantic Web are all slimmed down versions of modern 20th century logic based on a particular world view and set of assumptions. This works very well in theory, but once you start applying such logics to the real messy and complex world with real objects, processes, and ideas, the logics are put under increasing stress.

In logic, when two things are referred to as the same, this means they are two different names for the same thing, not that there are two things that are logically equivalent. So, Paris, the city of my dreams, and Paris the administrative area, Paris throughout history, and Paris – the capital of France are not necessarily all the same. This means that in logic we have to separate out into different versions aspects of an idea that in ordinary language we think of as the same thing.

He described this as the problem of “logic versus Occam” (as in Occam’s razor). Logic drives us to create complexity, in that we have to precisely define every aspect of a concept as a different entity. In order for the Semantic Web to work, we need to be very clear about our definitions so that we don’t muddle up different aspects of a concept.

Semantic enrichment

Guy Maréchal continued the Linked Data theme by talking in more detail about how flat data models can be semantically enriched. He pointed out that if you have good structured catalogue records, it takes very little effort to give concepts URIs and to export this data as sets of relationships. This turns your database into a graph, ready for semantic search and querying.

He argued that “going to semantics cannot be avoided” and that “born digital” works will increasingly be created with semantically modelled metadata.

From Mass Digitisation to Mass Content Enrichment

The next talk was a description of the SONUMA digitisation and metadata enhancement project. Sonuma and Memnon Archiving Services have been working on inventories and dictionaries to help them index audio visual assets. They have been converting speech to text, holding the text as XML files, and then associating sections of the XML with the appropriate point in the AV content, so that it can be searched.

They identify breaks in programmes by looking for the time stamps using OCR techniques, and then looking for jumps in the numerical sequences. They assume that jumps in the numbers are breaks in programmes. This enables them to break up long tapes into sections, which usually correspond to programmes.

Social networking and Knowledge Management

Tom Adami described Knowledge Management projects at the United Nations Mission in Sudan (Best Practice Lessons Learnt: How the Exit Interview and Oral History Project at UNMIS is building a knowledge database). The UN in Africa faces problems of high staff turnover, remote locations, and difficulties in maintaining infrastructure. However, they have been using social networking to encourage people to share their knowledge and experience in a user-friendly way and so add to the official knowledge base.

Archive as a social media lab: Creative dissemination of digital sound and audiovisual collections

Budhaditya Chattopadhyay talked about a project to bring together archival practice, artistic practice, and social media. He also referred to the problems of preserving social media which is in essence ephemeral but may be an integral part of an artwork.