Network data is ubiquitous, encoding collections of relationships between entities such as people, places, genes, or corporations. While many resources for networks of interesting entities are emerging, most of these can only annotate connections in a limited fashion. Although relationships between entities are rich, it is impractical to manually devise complete characterizations of these relationships for every pair of entities on large, real-world corpora. In this paper we present a novel probabilistic topic model to analyze text corpora and infer descriptions of its entities and of relationships between those entities. We develop variational methods for performing approximate inference on our model and demonstrate that our model can be practically deployed on large corpora such as Wikipedia. We show qualitatively and quantitatively that our model can construct and annotate graphs of relationships and make useful predictions.

We describe the approach to event extraction which the JULIELab Team from FSU Jena (Germany) pursued to solve Task 1 in the “BioNLP’09 Shared Task on Event Extraction”. We incorporate manually curated dictionaries and machine learning methodologies to sort out associated event triggers and arguments on trimmed dependency graph structures. Trimming combines pruning irrelevant lexical material from a dependency graph and decorating particularly relevant lexical conceptual class information. Given that methodological framework, the JULIELab Team scored on 2nd rank among 24 competing teams, with 45.8 % precision, 47.5 % recall and 46.7 % F1-score on all 3,182 events. 1

This paper presents a comparative evaluation of several state-of-the-art English parsers based on different frameworks. Our approach is to measure the impact of each parser when it is used as a component of an information extraction system that performs protein-protein interaction (PPI) identification in biomedical papers. We evaluate eight parsers (based on dependency parsing, phrase structure parsing, or deep parsing) using five different parse representations. We run a PPI system with several combinations of parser and parse representation, and examine their impact on PPI identification accuracy. Our experiments show that the levels of accuracy obtained with these different parsers are similar, but that accuracy improvements vary when the parsers are retrained with domain-specific data.

© 2008 Roberts et al; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License

We explore a rule system and a machine learning (ML) approach to automatically harvest information on gene regulation events (GREs) from biological documents in two different evaluation scenarios – one uses self-supplied corpora in a clean lab setting, while the other incorporates a standard reference database of curated GREs from REGULONDB, real-life data generated independently from our work. In the lab condition, we test how feasible the automatic extraction of GREs really is and achieve F-scores, under different, not directly comparable test conditions though, for the rule and the ML systems which amount to 34 % and 44%, respectively. In the REGU-LONDB condition, we investigate how robust both methodologies are by comparing them with this routinely used database. Here, the best F-scores for the rule and the ML systems amount to 34 % and 19%, respectively. 1

Natural language processing modules such as part-of-speech taggers, named-entity recognizers and syntactic parsers are commonly evaluated in isolation, under the assumption that artificial evaluation metrics for individual parts are predictive of practical performance of more complex language technology systems that perform practical tasks. Although this is an important issue in the design and engineering of systems that use natural language input, it is often unclear how the accuracy of an end-user application is affected by parameters that affect individual NLP modules. We explore this issue in the context of a specific task by examining the relationship between the accuracy of a syntactic parser and the overall performance of an information extraction system for biomedical text that includes the parser as one of its components. We present an empirical investigation of the relationship between factors that affect the accuracy of syntactic analysis, and how the difference in parse accuracy affects the overall system. 1

This is an Open Access article distributed under the terms of the Creative Commons Attribution License

The system we propose to learning semantic relations consists of two parallel components. For our final submission we used components based on the similarity measures defined over WordNet and the patterns extracted from the Web and WMTS. Other components using syntactic structures were explored but not used for the final run. 1 Experimental Set-up The system we used to classify the semantic relations consists of two parallel binary classifiers. We ran this system for each of the seven semantic relations separately. Each classifier predicts for each instance of the relation whether it holds or not. The predictions of all the classifiers are aggregated for each instance by disjunction. That is to say, each instance is predicted to be false by default unless any of the classifiers gives evidence against this. To generate the submitted predictions we used two parallel classifiers: (1) a classifier that combines eleven WordNet-based similarity measures, see Sec. 2.1, and (2) a classifier that learns lexical patterns from Google and the Waterloo Multi-Text System (WMTS)(Turney, 2004) snippets and applies these on the same corpora, see Sec. 2.2. Three other classifiers we experimented with, but that were not used to generate the submitted predictions: (3) a classifier that uses string kernel methods on the dependency paths of the training sentences, see Sec. 3.1, (4) a classifier that uses string kernels on the local context of the subject and object nominals in the training sentences, see Sec. 3.2 and (5) a classifier that uses hand-made lexical patterns on

Evaluating contributions of natural language parsers to protein–protein interaction extraction