"We introduce an algorithm that, given n objects, learns a similarity matrix over all n^2 pairs, from crowdsourced data alone. The algorithm samples responses to adaptively chosen triplet-based relative-similarity queries. Each query has the form "is object 'a' more similar to 'b' or to 'c'?" and is chosen to be maximally informative given the preceding responses. The output is an embedding of the objects into Euclidean space (like MDS); we refer to this as the "crowd kernel." SVMs reveal that the crowd kernel captures prominent and subtle features across a number of domains, such as "is striped" among neckties and "vowel vs. consonant" among letters."

"Yet, the co-training algorithm in this paper also makes the same assumptions (as it too has underlying naive Bayes clas- sifiers), but does not suffer from the violations. Thus we hypothesize that the co-training algorithm succeeds in part because it is more robust to the assumptions made by its underlying classifiers. This can be understood by looking at the differences in how EM and co-training use the underly- ing assumptions."