Clustering objects on subsets of attributes (with discussion)

Summary.  A new procedure is proposed for clustering attribute value data. When used in conjunction with conventional distance-based clustering algorithms this procedure encourages those algorithms to detect automatically subgroups of objects that preferentially cluster on subsets of the attribute variables rather than on all of them simultaneously. The relevant attribute subsets for each individual cluster can be different and partially (or completely) overlap with those of other clusters. Enhancements for increasing sensitivity for detecting especially low cardinality groups clustering on a small subset of variables are discussed. Applications in different domains, including gene expression arrays, are presented.     

Autobiographical sociology

In autobiographical sociology a sociologist probes one or more past personal experiences as a way of identifying and analyzing something sociologically relevant. The general nature and importance of autobiographical sociology, as well as illustrations of and guidelines for it, are explored.     

Suburban landscape: Views of New Jersey artists 1

Suburban landscape predominates within New Jersey; people live in one suburb and drive to a job in another. Research on community ideologies suggests, however, that even suburban residents have vague images of suburbs. Vague images can mean, in turn, that the state's visual artists produce few images that incorporate the newly‐created suburban landscape that they see every day. Research on art worlds also suggests little encouragement for work that depicts the newer suburbs. Consistent with this, only a few of the visual artists living in New Jersey who have put slides in three large slide files include work that deals with the newer suburban landscape or with the process of creating suburbs. Instead, artists who depict New Jersey landscape concentrate on the state's “natural” landscape or on the state's older industrial suburbs. The privacy of newer suburbs no doubt also contributes to this pattern. In suburbs, even places designed for crowds, such as retail malls, are private.     

Utility functions that lead to the likelihood ratio as a relative model performance measure

In this paper, we adopt the point of view of a decision-maker who evaluates a probabilistic model based on the test-sample averaged utility of the expected-utility optimal strategy that the model suggests in a horse race setting. After briefly reviewing the basic properties of the resulting performance measure for probabilistic models, we show that such a measure ranks two models according to their likelihood ratio if and only if the performance measure is based on a generalized logarithmic utility function. Thus, we provide a new decision theoretic motivation of the likelihood ratio as a model performance measure.