"Characterising Response Types and Revealing Noun Ambiguity in
 German Association Norms"

The talk uses semantic associations as a tool for automatically
identifying ambiguous nouns: Standard statistical analyses are
combined with the LSC soft clustering approach, to systematically
predict ambiguity within association norms for German nouns. In
contrast to prior association elicitation studies, the associations
were collected not only to written representations of target nouns but
also to their pictorial representations. In a first analysis,
systematic differences in the type and distribution of associate
responses for the two presentation forms are identified. In a second
analysis, a cluster analysis is applied to the collected
target-response pairs, and the clustering is used to predict noun
ambiguity and to discriminate senses in the target nouns.