Abstrakt
We consider the problem of user-item recommendation as a multiuser instance ranking learning. A user-item preference is monotonizable if the learning can restrict to monotone models.
A preference model is monotone if it is a monotone composition of rankings on domains of explanatory attributes (possibly describing user behavior, item content but also data aggregations). Target preference ordering of users on items is given by a preference indicator (e.g. purchase, rating).
To increase intuitiveness, we focus on a special class of monotone models which can be expressed as rules of generalized annotated programs. In this paper we focus on learning the (partial) order of vectors of rankings of user-item attribute values.
We measure degree of agreement of comparable vec-tors with ordering given by preference indicators for each user. We are interested in distribution of this degree across users.
We provide sets of experiments on user behavior data from an e-shop and on a subset of the semantically enriched Movie Lens 1M data.