Home » Software » Ontoterm

Ontoterm

WHAT IS ONTOTERM?

OntoTerm is a Terminology Management System (TMS) developed to overcome some of the problems that existing TMSs have. In particular it addresses two major issues:

  • Conceptual modeling: the object domain or subject field must be conceptually structured prior to entering language-specific terms. The construction of an ontology, i.e. a conceptually structured body of language-independent knowledge, is thus the first step in the construction of a termbase.
  • Terminology information exchange: OntoTerm implements the recent ISO standard for terminology exchange: Martif (ISO 1220) and all the data categories from the CLS Framework (ISO 1620).

If you are familiar with conventional terminology database management systems, you will probably find OntoTerm rather different. To begin with, OntoTerm does not allow you to enter terms in a termbase unless you have previously entered and defined a concept explic itly in the ontology. We can then say that OntoTerm is an ontology-based TMS (thereby its name). How do I use it? Are there any user manuals? On-line help?

At present the OntoTerm team consists of just one person (me). This means that I hardly have the time write the software, let alone manuals or on-line help. Here's a Getting Started Guide describing how you go about using OntoTerm. The Screenshots page contains also a few tips on how to use the program.

At this early stage of development the only on-line help available is on data categories. You can have a better description of them where I ripped it from.

The aim of offering the program for evaluation is to improve it in as many ways as users may suggest. That's why I need your feedback. Please send me bugs, comments or any sort of improvements you would like to get in future versions.

 

ARE THERE ANY PAPERS DESCRIBING IT?

Not yet, but I am working hard on this. I just haven't had the time to write anything but software! However, it is next on my priorities list.

The documentation page contains some relevant documents that you may find useful to unders tand the rationale behind OntoTerm. Do also have a look at the links page.

 

HOW MUCH DOES IT COST?

Nothing! At this moment OntoTerm is completely free! This is so because it is still at a very early stage of development. You are encouraged to download the program, use it and tell me what you think about it. OntoTerm 0.98 beta is available for download here. This evaluation version is not without restrictions, though: even though the Ontology Editor allows you to create as large as an ontology as you wish, it will not allow you to enter more than ten concepts in the TermBase Editor.

If you are interested and want to be informed as soon as new versions become available, do send me e-mail and I will add you to my OntoTerm-users mailing list. E-mail me also if you desperately need an unrestricted version.

 

WILL IT EVER BE COMMERCIALLY AVAILABLE?

That depends very much on whether you think it's useful. Some people seem to think that it is better than existing commercial TMSs. Again, download it and judge for yourself.

FEATURES

Concept-based

The Ontology Editor provides true conceptual modeling. You can start with a pre-defined set of top-level, generic concepts to help you start out or you can start from scratch and create your own concepts. Concepts are defined by the relationships they hold to one another, rather than by their name, which should be regarded solely as a mnemonic, rather than a language-specific. You are free to enrich this conceptual relationships as much as you want, so that the resulting ontology provides domain-speci fic knowledge about a given domain.

HTML publishing

You are able to publish your data easily and effortlessly. OntoTerm generates web pages using the information you have entered in the databases. Generate pages individually, as subsets, or the whole termbase using the HTML Report Generator.

Easy-to-use user interface

Unlike other TMSs, everything in OntoTerm is where you would expect it to be. It is highly graphical and uses context menus extensively. If you get lost try right-clicking on objects, that may give you a hint as to how to proceed from there.

ISO compliant

OntoTerm implements the recent ISO standard for terminology exchange: Martif (ISO 1220) and all the data categories from the CLS Framework (ISO 1620). In the future you will be able to define your own set of data categories, but it is very unlikely that you will ever need to do so, as the full set of the comprehensive data categories from the CLS Framework are already at your fingertips. The TermBase Editor makes the complexity of using these data categories properly a trifle by allowing the user to assign only the data categories appropriate for the selected element.

SCREENSHOTS

The Ontology Editor

This is a shot of the Ontology Editor at work. Typing text in the SEARCH box will easily find concepts. Selecting a concept from the list will make it the active concept, which you can then edit by assigning properties (attributes and relations to other concep ts). This is actually what really defines a concept in the ontology: its position in the hierarchy and its attributes, rather than the definition, which should be regarded only as an indication for the user to understand what is meant by it.

The editor is kept compact by using property pages (tabs) which together show -and allow to edit- all the information assigned to a concept.

New concepts are added using the "Add Concept" window, which may stay open while you browse and edit the ontology. All windows are updated as you add, delete, or modify concepts.

Concepts are defined by the set relations and attributes assigned to it. This will be inhereted by children concepts:

In heritance of properties is shown in the Inheritance tab. This is normally set to "Off" for faster browsing. Clicking on the "Full" or "Override" option buttons will compute and display inherited attributes and relations.

Trees

The IS-A relation defines the structure of your hierarchy, which can be displayed as a tree. You can have any number of trees open at any one time. You may choose to view the full ontology tree, as in the shot above or select a concept to be top node if you wish to concentrate on a section of the hierarchy:

In addition to searching in the Ontology Editor, you may search in the tree windows too. Right-clicking on a concept allows you to open that concept either in the Ontology Editor or in the Ontology Navigator.

All trees are dynamically loaded, which means that you won't have to wait for them to load. When a subsection of a tree changes, just collapse and expand again the node immediately above to reflect the changes.

The Ontology Navigator

The Ontology Navigator allows you to view the information you have entered in a readable format, as well as navigate using either the hierarchy tree or the hyperlinks on the pages:

Web pages are generated on the fly as you click on the tree nodes on the left. If the page exists already, it will just retrieve it from disk. The refresh button can be used to generate again that page so that it reflects changes. The arrows on the toolbar allow you to navigate as you would in any web browser.

Using the "Publish" menu you can choose to generate all pages for the current Working Ontology. However, you may choose to generate only a set of concepts, using the Report Generator.

The HTML Report Generator

This tool allows you to select which concepts you want to generate web pages for. Selection can be made by clicking on the concept names or by entering a regular (wildcarded) expression. You may also choose where you wanted published, which might as well be a web folder, and also what you want the reports to include.

The TermBase Editor

The TermBase Editor is a highly polymorphic tool that allows you to create, edit, and browse termbases in an extremel y efficient way. Using a standard set of data categories you can assign bits of information to concepts, terms and any other element. It is you who decides how much detail you want to express. Clicking on any element on the termbase tree (left pane) will show you the data categories available for that element (with a blue icon) and which are banned (red cross). All editing is performed in the same window, which changes its form depending on whether you are browsing or editing.

Shared elements, such as images or multimedia elements, bibliographic references, URLs, etc. are kept separately in what is called the back matter. After adding a new element to this colection you can include references to it to describe concepts, terms, and any other elements that allows them.

Cross-references are ext remely easy to add: you simply click on the element and select "cross-reference", then you click again on the cross-referenced element and -if necessary- give it a label. Right-clicking on a cross-reference brings up a menu that'll take you to the cross-referenced element.

The type of flexibility that this tool provides is unparalleled. If you have had enough of flat, rigid database str uctures, OntoTerm will give you the expressive power you need.

At any time you may wish to view the information you have entered in a readable format. Clicking on the "Browse Mode" button on the toolbar will turn the left pane into a full fledged web-browser that you can also use to navigate your termbase or simply to view any multimedia element included in the back matter. 

DOCUMENTS

Leveraging terminological data for use in conjunction with lexicographical resources. This document presents infor maiton on MARTIF and OLIF. The authors (Alan K. Melby and Sue Ellen Wright) welcome comments, which can be sent to Alan Melby.

Computer applications in terminology -- Method for sharing terminological data. By alan K. Melby (In postscript - zipped)

Kavi Mahesh and Sergei Nirenburg. (1995). A Situated Ontology f or Practical NLP. In Proc. Workshop on Basic Ontological Issues in Knowledge Sharing, International Joint Conference on Artificial Intelligence (IJCAI-95), Aug. 19-20, 1995. Montreal, Canada. Download Postscript Paper.

Viegas, E., Mahesh, K., Nirenburg, S. and S. Beale. 1999. Semantics in Action. In P. Saint-Dizier (ed). Predicative Forms in Natural Language and in Lexical Knowledge Bases, Kluwer Academic Press. Download ps file

GUM. The Generalized Upper Model 2.0. A general task and domain independent `linguistically motivated ontology'

An article of mine and Chantal Pérez (1998, in Spanish) about using ontologies for establishing translation equivalents. Download postscript file.

My PhD thesis, (1997, in Spanish), about the design of lexicons for computational lexicography and MT. All files are zipped postscript.

DOWNLOAD

The latest version available is 1.0. You must agree to the the license terms, which you should read during the installation process.

I would like to hear from you. Tell me what you're doing with OntoTerm.

LINKS

Ontologies

  • Mikrokosmos. A major Machine Translation project at New Mexico State University that inspired the creation of OntoTerm.
  • Stanford Knowledge Systems Laboratory. Create your own ontologies on the web.
  • Ontolingua. A standard for knowledge representation.
  • CODE4. A Knowledge Base Management System that has also been used for terminology.
  • Loom. A powerful KBMS.
  • Ontosaurus. A web browser for Loom knowledge bases.

Terminology

  • TTT.org -- Translation, Theory & Technology. Information about language theory and language technology, particularly relating to translation.
  • The CLS Framework. The result of a joint effort of the Brigham Young University Translation Research Group (BYU TRG) and the Kent State University Institute for Applied Linguistics (KSU IAL). The framework deals with the structure a nd content of terminological databases.
  • IULA. An outstanding Spanish terminology and lexicography institute.
  • SDL MultiTerm. A well-established Terminology Management System.
  • TermStar. Yet another TMS.