Issues in learning an ontology from text

Ontology construction for any domain is a labour intensive and complex process. Any methodology that can reduce the cost and increase efficiency has the potential to make a major impact in the life sciences. This paper describes an experiment in ontology construction from text for the animal behaviour domain. Our objective was to see how much could be done in a simple and relatively rapid manner using a corpus of journal papers. We used a sequence of pre-existing text processing steps, and here describe the different choices made to clean the input, to derive a set of terms and to structure those terms in a number of hierarchies. We describe some of the challenges, especially that of focusing the ontology appropriately given a starting point of a heterogeneous corpus. Results - Using mainly automated techniques, we were able to construct an 18055 term ontology-like structure with 73% recall of animal behaviour terms, but a precision of only 26%. We were able to clean unwanted terms from the nascent ontology using lexico-syntactic patterns that tested the validity of term inclusion within the ontology. We used the same technique to test for subsumption relationships between the remaining terms to add structure to the initially broad and shallow structure we generated. All outputs are available at http://thirlmere.aston.ac.uk/~kiffer/animalbehaviour/ webcite. Conclusion - We present a systematic method for the initial steps of ontology or structured vocabulary construction for scientific domains that requires limited human effort and can make a contribution both to ontology learning and maintenance. The method is useful both for the exploration of a scientific domain and as a stepping stone towards formally rigourous ontologies. The filtering of recognised terms from a heterogeneous corpus to focus upon those that are the topic of the ontology is identified to be one of the main challenges for research in ontology learning.

Issues in learning an ontology from text

Ontology construction for any domain is a labour intensive and complex process. Any methodology that can reduce the cost and increase efficiency has the potential to make a major impact in the life sciences. This paper describes an experiment in ontology construction from text for the Animal Behaviour domain. Our objective was to see how much could be done in a simple and rapid manner using a corpus of journal papers. We used a sequence of text processing steps, and describe the different choices made to clean the input, to derive a set of terms and to structure those terms in a hierarchy. We were able in a very short space of time to construct a 17000 term ontology with a high percentage of suitable terms. We describe some of the challenges, especially that of focusing the ontology appropriately given a starting point of a heterogeneous corpus.

Tetrahydrobiopterin metabolism, neurotransmitters and behaviour in the rat

Various neurotoxins were investigated to assess their suitability for developing an animal model to study partial brain BH4 deficiency, neurotransmitters and behavioural alterations. Acute dosing with lead, diethylstilboestrol (DES), amphetamine and scopolamine produced no significant changes in rat brain BH4 metabolism though total biopterins in the liver were significantly reduced by lead and DES. Acute starvation of adult rats decreased brain biopterins. This loss of biopterins may be due to enhanced oxidative catabolism of the active cofactor caused by glutathione depletion. Dietary administration of a BH4 biosynthesis inhibitor, DAHP, consistently decreased brain total biopterins in weaner rats but did not alter the levels of DA, NA, 5-HT or metabolites. However the DAHP diet also induced a marked reduction in food intake. Rats subjected to an equivalent degree of food restriction without inhibitor showed significant but less severe reductions in brain biopterins and again no effect on transmitter levels. DAHP produced a significant decrease in locomotor activity and rearing. This could not be ascribed to reduction in food intake as animals subjected to just dietary restriction showed an increase in these activities. As gross brain levels of DA, NA and 5-HT were unaltered by DAHP the behavioural changes associated with the induced deficiency in brain total biopterins might not have been mediated through the action of these compounds. Although localised changes in neurotransmitter levels may have been obscured by gross analysis it is also possible that the behaviour changes were mediated by a role of BH4 not yet elucidated. Long-term administration of a high aluminium low calcium diet to mice produced no effect on gross brain total biopterins, catecholamines, serotonin or choline acetyltransferase activity though significant behavioural changes were observed.

An X-ray micro-fluorescence study to investigate the distribution of Al, Si, P and Ca ions in the surrounding soft tissue after implantation of a calcium phosphate-mullite ceramic composite in a rabbit animal model

Synthetic calcium phosphates, despite their bioactivity, are brittle. Calcium phosphate-mullite composites have been suggested as potential dental and bone replacement materials which exhibit increased toughness. Aluminium, present in mullite, has however been linked to bone demineralisation and neurotoxicity: it is therefore important to characterise the materials fully in order to understand their in vivo behaviour. The present work reports the compositional mapping of the interfacial region of a calcium phosphate-20 wt% mullite biocomposite/soft tissue interface, obtained from the samples implanted into the long bones of healthy rabbits according to standard protocols (ISO-10993) for up to 12 weeks. X-ray micro-fluorescence was used to map simultaneously the distribution of Al, P, Si and Ca across the ceramic-soft tissue interface. A well defined and sharp interface region was present between the ceramic and the surrounding soft tissue for each time period examined. The concentration of Al in the surrounding tissue was found to fall by two orders of magnitude, to the background level, within similar to 35 mu m of the implanted ceramic.