Martin Romacker

CREATING KNOWLEDGE REPOSITORIES FROM BIOMEDICAL REPORTS: THE MEDSYNDIKATE TEXT MINING SYSTEM

MEDSYNDIKATE is a natural language processor for automatically acquiring knowledge from medical finding reports. The content of these documents is transferred to formal representation structures which constitute a corresponding text knowledge base. The system architecture integrates requirements from the analysis of single sentences, as well as those of referentially linked sentences forming cohesive texts. The strong demands MEDSYNDIKATE poses to the availability of expressive knowledge sources are accounted for by two alternative approaches to (semi)automatic ontology engineering. We also present data for the knowledge extraction performance of MEDSYNDIKATE for three major syntactic patterns in medical documents.

MEDSYNDIKATE--a natural language system for the extraction of medical information from findings reports.

MEDSYNDIKATE is a natural language processor, which automatically acquires medical information from findings reports. In the course of text analysis their contents is transferred to conceptual representation structures, which constitute a corresponding text knowledge base. MEDSYNDIKATE is particularly adapted to deal properly with text structures, such as various forms of anaphoric reference relations spanning several sentences. The strong demands MEDSYNDIKATE poses on the availability of expressive knowledge sources are accounted for by two alternative approaches to acquire medical domain knowledge (semi)automatically. We also present data for the information extraction performance of MEDSYNDIKATE in terms of the semantic interpretation of three major syntactic patterns in medical documents.

Content management in the SYNDIKATE system —: how technical documents are automatically transformed to text knowledge bases

Abstract S YN D I KAT E is a family of natural language understanding systems for automatically acquiring knowledge from real-world texts (e.g., information technology test reports, medical finding reports), and for transferring their content to formal representation structures which constitute a corresponding text knowledge base. We present a general system architecture which integrates requirements from the analysis of single sentences, as well as those of referentially linked sentences forming cohesive texts. Properly accounting for text cohesion phenomena is a prerequisite for the soundness and validity of the generated text representation structures. It is also crucial for any information system application making use of automatically generated text knowledge bases in a reliable way, e.g., by inferentially supported fact retrieval.

Part-whole reasoning: a case study in medical ontology engineering

Clinical computing requires effective medical ontologies that can support large-scale formal reasoning. Our proposal lets the knowledge engineer, on demand, enable or disable transitivity of part-whole reasoning and part-whole induced concept specialization and role propagation, with respect to commonly shared medical conceptualizations.

Discourse structures in medical reports—Watch out! The generation of referentially coherent and valid text knowledge bases in the medSYNDIKATE system

The automatic analysis of medical narratives currently suffers from neglecting text structure phenomena such as referential relations between discourse units. This has unwarranted effects on the descriptional adequacy of medical knowledge bases automatically generated from texts. The resulting representation bias can be characterized in terms of incomplete, artificially fragmented and referentially invalid knowledge structures. We focus here on four basic types of textual reference relations, viz. pronominal and nominal anaphora, textual ellipsis and metonymy and show how to deal with them in an adequate text parsing device. Since the types of reference relations we discuss show an increasing dependence on conceptual background knowledge, we stress the need for formally grounded, expressive conceptual representation systems for medical knowledge. Our suggestions are based on experience with MEDSYNDIKATE, a medical text knowledge acquisition system designed to properly deal with various sorts of discourse structure phenomena.

The SYNDIKATE text knowledge base generator

SYNDIKATE comprises a family of text understanding systems for automatically acquiring knowledge from real-world texts, viz. information technology test reports and medical finding reports. Their content is transformed to formal representation structures which constitute corresponding text knowledge bases. SYNDIKATEs architecture integrates requirements from the analysis of single sentences, as well as those of referentially linked sentences forming cohesive texts. Besides centering-based discourse analysis mechanisms for pronominal, nominal and bridging anaphora, SYNDIKATE is supplied with a learning module for automatically bootstrapping its domain knowledge as text analysis proceeds.

Context-Based Ambiguity Management for Natural Language Processing

We introduce a formal context mechanism, embedded into a description logics framework, which allows to uniformly represent and manage different kinds of natural language ambiguities as they occur in the course of text understanding. Multiple lexical, syntactic and semantic interpretations are separated by assigning each of these alternatives a single context space for local reasoning and incremental disambiguation.

Small Is Beautiful - Compact Semantics for Medical Language Processing

We introduce two abstraction mechanisms by which the process of semantic interpretation of medical documents can be simplified and optimized. One relates to the linguistic generality, the other to the inheritance-based specification of semantic rules. The proposed methodology leads to a parsimonious inventory of abstract, simple and domain-independent semantic interpretation schemata whose effectiveness has been evaluated on a medical text corpus.

Part-Whole Reasoning in Medical Knowledge Bases Using Description Logics

The development of powerful, ubiquitous and comprehensive medical ontologies that support formal reasoning on a large scale is one of the key requirements for clinical computing. Taxonomic medical knowledge, a major portion of these ontologies, is fundamentally characterized by is—a and part-whole relationships between concepts. While reasoning in generalization hierarchies is a well-understood process, no fully conclusive mechanism yet exists for part-whole reasoning. We here propose a new representation construct for part-whole relations based on the formal framework of description logics, i.e. the well-known concept language ALC, and show how part-whole reasoning can naturally be emulated via classification-based reasoning without extending the expressiveness of the underlying terminological system.

A Natural Language Understanding System for Knowledge-Based Analysis of Medical Texts

An approach to knowledge-based text understanding of real-world texts from the medical domain (viz. gastro-intestinal findings) is presented. We survey major methodological features of an object-oriented, fully lexicalized, dependency-based grammar model which is tightly linked to domain knowledge representations based on description logics. The parser adheres to the principles of robustness, incrementality and concurrency. The substrate of automatic knowledge acquisition are text knowledge bases generated by the parser from medical narratives, which represent major portions of the content of these documents.