This paper describes and evaluates privacy-friendly methods for extracting quasi-social networks from browser behavior on user-generated content sites, for the purpose of finding good audiences for brand advertising (as opposed to click maximizing, for example). Targeting social-network neighbors resonates well with advertisers, and on-line browsing behavior data counterintuitively can allow the identification of good audiences anonymously. Besides being one of the first papers to our knowledge on data mining for on-line brand advertising, this paper makes several important contributions. We introduce a framework for evaluating brand audiences, in analogy to predictive-modeling holdout evaluation. We introduce methods for extracting quasi-social networks from data on visitations to social networking pages, without collecting any information on the identities of the browsers or the content of the social-network pages. We introduce measures of brand proximity in the network, and show that audiences with high brand proximity indeed show substantially higher brand affinity. Finally, we provide evidence that the quasi-social network embeds a true social network, which along with results from social theory offers one explanation for the increases in audience brand affinity.

A common setting for novelty detection assumes that labeled examples from the nominal class are available, but that labeled examples of novelties are unavailable. The standard (inductive) approach is to declare novelties where the nominal density is low, which reduces the problem to density level set estimation. In this paper, we consider the setting where an unlabeled and possibly contaminated sample is also available at learning time. We argue that novelty detection in this semi-supervised setting is naturally solved by a general reduction to a binary classification problem. In particular, a detector with a desired false positive rate can be achieved through a reduction to Neyman-Pearson classification. Unlike the inductive approach, semi-supervised novelty detection (SSND) yields detectors that are optimal (e.g., statistically consistent) regardless of the distribution on novelties. Therefore, in novelty detection, unlabeled data have a substantial impact on the theoretical properties of the decision rule. We validate the practical utility of SSND with an extensive experimental study. We also show that SSND provides distribution-free, learning-theoretic solutions to two well known problems in hypothesis testing. First, our results provide a general solution to the general two-sample problem, that is, the problem of determining whether two random samples arise from the same distribution. Second, a specialization of SSND coincides with the standard p-value approach to multiple testing under the so-called random effects model. Unlike standard rejection regions based on thresholded p-values, the general SSND framework allows for adaptation to arbitrary alternative distributions in multiple dimensions.

Learning from Demonstration (LfD) is a popular technique for building decision-making agents from human help. Traditional LfD methods use demonstrations as training examples for supervised learning, but complex tasks can require more examples than is practical to obtain. We present Abstraction from Demonstration (AfD), a novel form of LfD that uses demonstrations to infer state abstractions and reinforcement learning (RL) methods in those abstract state spaces to build a policy. Empirical results show that AfD is greater than an order of magnitude more sample efficient than just using demonstrations as training examples, and exponentially faster than RL alone. 1

This paper studies the effects of training data on binary text classification and postulates that negative training data is not needed and may even be harmful for the task. Traditional binary classification involves building a classifier using labeled positive and negative training examples. The classifier is then applied to classify test instances into positive and negative classes. A fundamental assumption is that the training and test data are identically distributed. However, this assumption may not hold in practice. In this paper, we study a particular problem where the positive data is identically distributed but the negative data may or may not be so. Many practical text classification and retrieval applications fit this model. We argue that in this setting negative training data should not be used, and that PU learning can be employed to solve the problem. Empirical evaluation has been conducted to support our claim. This result is important as it may fundamentally change the current binary classification paradigm. 1

Abstract. Classifiers are traditionally learned using sets of positive and negative training examples. However, often a classifier is required, but for training only an incomplete set of positive examples and a set of unlabeled examples are available. This is the situation, for example, with the Transport Classification Database (TCDB, www.tcdb.org), a repository of information about proteins involved in transmembrane transport. This paper presents and evaluates a method for learning to rank the likely relevance to TCDB of newly published scientific articles, using the articles currently referenced in TCDB as positive training examples. The new method has succeeded in identifying 964 new articles relevant to TCDB in fewer than six months, which is a major practical success. From a general data mining perspective, the contributions of this paper are (i) devising and evaluating two novel approaches that solve the positive-only problem effectively, (ii) applying support vector machines in a state-ofthe-art way for recognizing and ranking relevance, and (iii) deploying a system to update a widely-used, real-world biomedical database. Supplementary information including all data sets are publicly available at www.cs.ucsd.edu/users/knoto/pub/ajcai08. 1

Abstract not found

Journal of Digital Information Management Abstract: We identify and explore an Information Retrieval paradigm called Query-By-Multiple-Examples (QBME) where the information need is described not by a set of terms but by a set of documents. Intuitive ideas for QBME include using the centroid of these documents or the well-known Rocchio algorithm to construct the query vector. We consider this problem from the perspective of text classification, and find that a better query vector can be obtained through learning with Support Vector Machines (SVMs). For online queries, we show how SVMs can be learned from one-class examples in linear time. For offline queries, we show how SVMs can be learned from positive and unlabeled examples together in linear or polynomial time, optimising some meaningful multivariate performance measures. The effectiveness and efficiency of the proposed approaches have been confirmed by our experiments on four real-world datasets.

The Protein-Protein Interaction tasks of BioCreative III: classification/ranking of articles and linking bio-ontology

reliability diagram, adaptive learning, structured learning, maximum margin optimization, convex optimizaton, Medications designed for a general population do not work the same for each individual. Similarly, patterns observed from naturally occurring disease outbreaks do not necessarily describe outbreaks of purposeful disease outbreaks (e.g. bioterrorism). To tackle challenges posed by individual differences, my thesis introduces data-driven paradigms that predict a particular case will have the outcome of interest. My insight is to accommodate individual differences by coherently leveraging information from complementary perspectives (e.g., temporal dependency, relational correlation, feature similarity, and estimation uncertainty) to provide more reliable predictions than possible with existing cohort-based approaches. Specifically, I carefully investigated two representative problems, bioterrorism-related disease outbreak and personalized clinical decision support, for which previous research does not provide satisfactory solutions. I developed a Temporal Maximum Margin Markov Network framework to consider the temporal correlation concurrently with relational dependency in bioterrorism-related diseases ’ outbreaks. This framework reduces the ambiguity in estimating

In scientific research, it is often difficult to express information needs as simple keyword queries. We present a more natural way of searching for relevant scientific literature. Rather than a string of keywords, we define a query as a small set of papers deemed relevant to the research task at hand. By optimizing an objective function based on a fine-grained notion of influence between documents, our approach efficiently selects a set of highly relevant articles. Moreover, as scientists trust some authors more than others, results are personalized to individual preferences. In a user study, researchers found the papers recommended by our method to be more useful, trustworthy and diverse than those selected by popular alternatives, such as Google Scholar and a state-of-the-art topic modeling approach.