Li-min Liu

Representing the UMLS as an Object-oriented Database: Modeling Issues and Advantages

OBJECTIVE The Unified Medical Language System (UMLS) combines many well-established authoritative medical informatics terminologies in one knowledge representation system. Such a resource is very valuable to the health care community and industry. However, the UMLS is very large and complex and poses serious comprehension problems for users and maintenance personnel. The authors present a representation to support the users comprehension and navigation of the UMLS. DESIGN An object-oriented database (OODB) representation is used to represent the two major components of the UMLS-the Metathesaurus and the Semantic Network-as a unified system. The semantic types of the Semantic Network are modeled as semantic type classes. Intersection classes are defined to model concepts of multiple semantic types, which are removed from the semantic type classes. RESULTS The authors provide examples of how the intersection classes help expose omissions of concepts, highlight errors of semantic type classification, and uncover ambiguities of concepts in the UMLS. The resulting UMLS OODB schema is deeper and more refined than the Semantic Network, since intersection classes are introduced. The Metathesaurus is classified into more mutually exclusive, uniform sets of concepts. The schema improves the users comprehension and navigation of the Metathesaurus. CONCLUSIONS The UMLS OODB schema supports the users comprehension and navigation of the Metathesaurus. It also helps expose and resolve modeling problems in the UMLS.

Controlled Vocabularies in OODBs: Modeling Issues and Implementation

A major problem that arises in many large application domains is the discrepancy among terminologies of different information systems. The terms used by the information systems of one organization may not agree with the terms used by another organization even when they are in the same domain. Such a situation clearly impedes communication and the sharing of information, and decreases the efficiency of doing business. Problems of this nature can be overcome using a controlled vocabulary (CV), a system of concepts that consolidates and unifies the terminologies of a domain. However, CVs are large and complex and difficult to comprehend. This paper presents a methodology for representing a semantic network-based CV as an object-oriented database (OODB). We call such a representation an Object-Oriented Vocabulary Repository (OOVR). The methodology is based on a structural analysis and partitioning of the source CV. The representation of a CV as an OOVR offers both the level of support typical of database management systems and an abstract view which promotes comprehension of the CVs structure and content. After discussing the theoretical aspects of the methodology, we apply it to the MED and InterMED, two existing CVs from the medical field. A program, called the OOVR Generator, for automatically carrying out our methodology is described. Both the MED-OOVR and the InterMED-OOVR have been created using the OOVR Generator, and each exists on top of ONTOS, a commercial OODBMS. The OOVR derived from the InterMED is presently available on the Web.

Modeling a vocabulary in an object-oriented database

Controlled vocabularies have been used as the means for unifying disparate terminologies found within an application field. This unification leads to better administration of information and enhanced communication among various parties. Semantic networks have been shown to be excellent vehicles for modeling controlled vocabularies. However, they often lack the necessary access flexibility and robustness required by external agents such as intelligent information-locators and decision-support systems. In this paper, we describe the process of mapping an existing medical vocabulary based on a semantic network model into an Object-Oriented Database (OODB) system. We first consider two straightforward approaches to carrying out this task and describe their deficiencies. We then present a new approach which yields a very compact OODB schema for the representation of the vocabulary’s entire hierarchy and inter-connectivity. We refer to the resulting OODB as the Object-Oriented Healthcare Vocabulary Repository (OOHVR), which is currently up and running in the context of ONTOS, a commercially available OODB system.

Using OODB modeling to partition a vocabulary into structurally and semantically uniform concept groups

Controlled vocabularies (CVs) are networks of concepts that unify disparate terminologies and facilitate the process of information sharing within an application domain. We describe a general methodology for representing an existing CV as an object-oriented database (OODB), called an object-oriented vocabulary repository (OOVR). A formal description of the OOVR methodology, which is based on a structural abstraction technique, is given, along with an algorithmic description and a number of theorems pertaining to some of the methodologys formal characteristics. An OOVR offers a two-level (concept level and schema level) view of a CV, with the schema-level view serving as an important abstraction that can aid in orientation to the CVs contents. While an OOVR can also assist in traversals of the CV, we have identified certain special CV configurations where such traversals can be problematic. To address this, we introduce - based on the original methodology - an enhanced OOVR methodology that utilizes both structural and semantic features to partition and model a CVs constituent concepts. With its basis in the notions of area and the recursively defined articulation concept, an enhanced OOVR representation provides users with an improved CV view comprising groups of concepts that are uniform both in their structure and semantics. An algorithmic description of the singly-rooted OOVR methodology and theorems describing some of its formal properties are given. The results of applying it to a large existing CV are discussed.

Incorporating semantic relationships into an object-oriented database system

Semantic relationships, those class-to-class connections that carry inherent support for constraints and various other functionalities, play an important role when building information models for applications. This is true whether one employs traditional data modeling techniques, knowledge representation languages, or object-oriented modeling methodologies. An example of such a semantic relationship is the part-whole relationship. In fact, most of the popular object-oriented modeling approaches include such constructs in their repertoire of data modeling primitives. However commercial object-oriented database (OODB) systems ordinarily do not provide built-in support for them. We present a methodology by which a semantic relationship can be incorporated into an existing OODB system. At first, we give an overview of the general methodology for carrying out this integration task. Then, in order to ground our work in a real system, we show the addition of a part-whole semantic relationship to the ONTOS DB/Explorer OODB management system. This implementation is up and running, and an example application demonstrating its use is available on the Web.

Frameworks for incorporating semantic relationships into object-oriented database systems

A semantic relationship is a data modeling construct that connects a pair of classes or categories and has inherent constraints and other functionalities that precisely reflect the characteristics of the specific relationship in an application domain. Examples of semantic relationships include part–whole, ownership, materialization and role‐of. Such relationships are important in the construction of information models for advanced applications, whether one is employing traditional data‐modeling techniques, knowledge‐representation languages or object‐oriented modeling methodologies. This paper focuses on the issue of providing built‐in support for such constructs in the context of object‐oriented database (OODB) systems. Most of the popular object‐oriented modeling approaches include some semantic relationships in their repertoire of data‐modeling primitives. However, commercial OODB systems, which are frequently used as implementation vehicles, tend not to do the same. We will present two frameworks by which a semantic relationship can be incorporated into an existing OODB system. The first only requires that the OODB system support manifest type with respect to its instances. The second assumes that the OODB system has a special kind of metaclass facility. The two frameworks are compared and contrasted. In order to ground our work in existing systems, we show the addition of a part–whole semantic relationship both to the ONTOS DB/Explorer OODB system and the VODAK Model Language. Copyright

Using a Similarity Measurement to Partition a Vocabulary of Medical Concepts

Controlled medical vocabularies have become increasingly important in a range of medical informatics applications. However, the extensive size of most vocabularies often makes it diffcult for users to gain an understanding of their contents. In previous work, we have investigated the partitioning of a large semantic-network based medical vocabulary into smaller units, for the purpose of easier graphical display and comprehension. The partitioning process relied heavily on a domain expert. In this paper, we propose a structural method for automating the partitioning of a vocabulary. The structural method is based on a definition of the similarity of a pair consisting of a child concept and its parent concept in the semantic network. A distribution over these similarities for all pairs in the semantic network is then computed. Based on this distribution, the semantic network can be partitioned into more manageable pieces. The approach has been applied to the InterMED and a complex portion of the MED, two large medical vocabularies.

Ownership as a conceptual modeling construct

Ownership is a relationship that pervades many aspects of our lives, from the personal to the economic, and is particularly important in the realm of the emerging electronic economy. As it is understood on an intuitive level, ownership exhibits a great deal of complexity and carries a rich semantics with respect both to the owner and the possession. A formal model of an ownership relationship that inherently captures varied ownership semantics is presented. This ownership relationship expands the repertoire of available conceptual data modeling primitives. It is built up from a set of characteristic dimensions, namely, exclusiveness, dependency, documentation, transferability, and inheritance, each of which focuses on a specific aspect of ownership semantics. The data modeler has the ability to make a variety of choices along these five dimensions, and thus has access to a wide range of available ownership features in a declarative fashion. These choices ultimately impose various constraints (specified in OCL) on the states of data objects and their respective ownership activities, including transactions such as acquiring and relinquishing ownership. To complement the formal aspects of the ownership model and enhance its usability, we present a graphical ownership notation that augments the Unified Modeling Language (UML) class diagram formalism. An implementation of the ownership relationship in a commercial object-oriented database system is discussed.

Enhancing OODB semantics to support browsing in an OODB vocabulary representation

In previous work, we have modeled a vocabulary given as a semantic network by an object‐oriented database (OODB). The OODB schema thus obtained provides a compact abstract view of the vocabulary. This enables the fast traversal of the vocabulary by a user. In the semantic network vocabulary, the IS‐A relationships express the specialization hierarchy. In our OODB modeling of the vocabulary, the SUBCLASS relationship expresses the specialization hierarchy of the classes and supports the inheritance of their properties. A typical IS‐A path in the vocabulary has a corresponding shorter SUBCLASS path in the OODB schema.

The Importance of Ownership for E-Commerce

This paper examines the nature and characteristics of the ownership relationship between objects in an object-oriented database. The ownership relationship is distinguished from the Part-Whole relationship since these two are often confused. Dimensions of the ownership relationship are detailed and described with clarifying examples from the financial securities markets. These dimensions are: 1) exclusiveness, 2) dependency, 3) limitation, 4) status, 5) settlement, 6) documentation and 7) inheritance. Formal definitions of the dimensions of the ownership relationship are sketched out for the purpose of extending current Object-Oriented Database Models (OODBs).