Digital platforms have a widely-documented issue with bias, which can impact their bottom line via missed opportunities, bad publicity, or even legal action. There is evidence that bias decreases when a platform conveys a signal of quality for underrepresented groups. This paper studies the...

A main goal of online display advertising is to drive purchases (etc.) following ad engagement. However, there often are too few purchase conversions for campaign evaluation and optimization, due to low conversion rates, cold start periods, and long purchase cycles (e.g., with brand...

This paper addresses the repeated acquisition of labels for data itemswhen the labeling is imperfect. We examine the improvement (or lackthereof) in data quality via repeated labeling, and focus especially onthe improvement of training labels for supervised induction. With theoutsourcing of...

Traditional event models underlying naive Bayes classifiers assume probability distributions that are not appropriate for binary data generated by human behaviour. In this work, we develop a new event model, based on a somewhat forgotten distribution created by Kenneth Ted Wallenius in 1963. We...

This paper addresses the classification of linked entities. Weintroduce a relational vector (VS) model (in analogy to theVS model used in information retrieval) that abstracts the linkedstructure, representing entities by vectors of weights. Givenlabeled data as background knowledge training...

For many supervised learning tasks, the cost of acquiringtraining data is dominated by the cost of class labeling.In this work, we explore active learning forclass probability estimation (CPE). Active learning acquiresdata incrementally, using the model learned sofar to help identify especially...

Tree induction and logistic regression are two standard, off-the-shelf methodsfor building models for classification. We present a large-scale experimentalcomparison of logistic regression and tree induction, assessing classification accuracyand the quality of rankings based on class-membership...

This paper addresses focused information acquisition for predictive data mining. Asbusinesses strive to cater to the preferences of individual consumers, they often employpredictive models to customize marketing efforts. Building accurate models requiresinformation about consumer preferences...