Learning concept hierarchies from text corpora using formal concept analysis
Journal of Artificial Intelligence Research
| Hi-index | 0.00 |
We describe an approach designed to reduce the costs of ontology development through the use of untrained, volunteer knowledge engineers. Results are provided from an experiment in which volunteers were asked to judge the correctness of automatically inferred subsumption relationships in the biomedical domain. The experiment indicated that volunteers can be recruited fairly easily but that their attention is difficult to hold, that most do not understand the subsumption relationship without training, and that incorporating learned estimates of trust into voting systems is beneficial to aggregate performance.