Abstract
One of the challenges in Machine Learning to find a classifier and parameter settings that work well on a given dataset. Evaluating all possible combinations typically takes too much time, hence many solutions have been proposed that attempt to predict which classifiers are most promising to try. As the first recommended classifier is not always the correct choice, multiple recommendations should be made, making this a ranking problem rather than a classification problem. Even though this is a well studied problem, there is currently no good way of evaluating such rankings. We advocate the use of Loss Time Curves, as used in the optimization literature. These visualize the amount of budget (time) needed to converge to a acceptable solution. We also investigate a method that utilizes the measured performances of classifiers on small samples of data to make such recommendation, and adapt it so that it works well in Loss Time space. Experimental results show that this method converges extremely fast to an acceptable solution.

Abstract
We present an Information Retrieval tool that facilitates the task of the user when searching for a particular information that is of interest to him. Our system processes a given set of documents to produce a graph, where nodes represent documents and links the similarities. The aim is to offer the user a tool to navigate in this space in an easy way. It is possible to collapse/expand nodes. Our case study shows affinity groups based on the similarities of text production of researchers. This goes beyond the already established communities revealed by co-authorship. The system characterizes the activity of each author by a set of automatically generated keywords and by membership to a particular affinity group. The importance of each author is highlighted visually by the size of the node corresponding to the number of publications and different measures of centrality. Regarding the validation of the method, we analyse the impact of using different combinations of titles, abstracts and keywords on capturing the similarity between researchers.

Abstract
Several algorithms based on PageRank algorithm have been proposed to rank the document sentences in the multi-document summarization field and LexRank and T-LexRank algorithms are well known examples. In literature different concepts such as weighted inter-cluster edge, cluster-sensitive graph model and document-sensitive graph model have been proposed to improve LexRank and T-LexRank algorithms (e.g. DsR-G, DsR-Q) for multi-document summarization. In this paper, a density-based graph model for multi-document summarization is proposed by adding the concept of density to LexRank and T-LexRank algorithms. The resulting generic multi-document summarization systems, DensGS and DensGSD were evaluated on DUC 2004 while the query-based variants, DensQS, DensQSD were evaluated on DUC 2006, DUC 2007 and TAC 2010 task A. ROUGE measure was used in the evaluation. Experimental results show that density concept improves LexRank and T-LexRank algorithms and outperforms previous graph-based models (DsR-G and DsR-Q) in generic and query-based multi-document summarization tasks. Furthermore, the comparison of the number of iterations indicates that the density-based algorithm is faster than the other algorithms based on PageRank.

Abstract
Most research on multi-document summarization explores methods that generate summaries based on queries regardless of the users' preferences. We note that, different users can generate somewhat different summaries on the basis of the same source data and query. This paper presents our study on how to exploit the information regards how users summarized their texts. Models of different users can be used either separately, or in an ensemble-like fashion. Machine learning methods are explored in the construction of the individual models. However, we explore yet another hypothesis. We believe that the sentences selected into the summary should be coherent and supplement each other in their meaning. One method to model this relationship between sentences is by detecting actor-object relationship (AOR). The sentences that satisfy this relationship have their importance value enhanced. This paper combines ensemble summarizing system and AOR to generate summaries. We have evaluated this method on DUC 2006 and DUC 2007 using ROUGE measure. Experimental results show the supervised method that exploits the ensemble summarizing system combined with AOR outperforms previous models when considering performance in query-based multi-document summarization tasks.

Abstract

Abstract
The pursuit of unhealthy behaviors, such as smoking or binge drinking, not only carries various downside risks, but also provides pleasure. A parsimonious model, used in the literature to explain the decision to pursue an unhealthy activity, represents that decision as a tradeoff between risks and benefits. We build on this literature by surveying a rural population in South Africa to elicit the perceived riskiness and the perceived pleasure for various risky activities and to examine how these perceptions relate to the pursuit of four specific unhealthy behaviors: frequent smoking, problem drinking, seatbelt nonuse, and risky sex. We show that perceived pleasure is a significant predictor for three of the behaviors and that perceived riskiness is a significant predictor for two of them. We also show that the correlation between the riskiness rating and behavior is significantly different from the correlation between the pleasure rating and behavior for three of the four behaviors. Finally, we show that the effect of pleasure is significantly greater than the effect of riskiness in determining drinking and risky sex, while the effects of pleasure and riskiness are not different from each other in determining smoking and seatbelt nonuse. We discuss how our findings can be used to inform the design of health promotion strategies.