Research in imbalanced domain learning has almost exclusively focused on solving classification tasks for accurate prediction of cases labelled with a rare class. Approaches for addressing such problems in regression tasks are still scarce due to two main factors. First, standard regression tasks assume each domain value as equally important. Second, standard evaluation metrics focus on assessing the performance of models on the most common values of data distributions. In this paper, we present an approach to tackle imbalanced regression tasks where the objective is to predict extreme (rare) values. We propose an approach to formalise such tasks and to optimise/evaluate predictive models, overcoming the factors mentioned and issues in related work. We present an automatic and non-parametric method to obtain relevance functions, building on the concept of relevance as the mapping of target values into non-uniform domain preferences. Then, we propose SERA, a new evaluation metric capable of assessing the effectiveness and of optimising models towards the prediction of extreme values while penalising severe model bias. An experimental study demonstrates how SERA provides valid and useful insights into the performance of models in imbalanced regression tasks.

Time series forecasting is a challenging task, where the non-stationary characteristics of data portray a hard setting for predictive tasks. A common issue is the imbalanced distribution of the target variable, where some values are very important to the user but severely under-represented. Standard prediction tools focus on the average behaviour of the data. However, the objective is the opposite in many forecasting tasks involving time series: predicting rare values. A common solution to forecasting tasks with imbalanced data is the use of resampling strategies, which operate on the learning data by changing its distribution in favour of a given bias. The objective of this paper is to provide solutions capable of significantly improving the predictive accuracy on rare cases in forecasting tasks using imbalanced time series data. We extend the application of resampling strategies to the time series context and introduce the concept of temporal and relevance bias in the case selection process of such strategies, presenting new proposals. We evaluate the results of standard forecasting tools and the use of resampling strategies, with and without bias over 24 time series data sets from six different sources. Results show a significant increase in predictive accuracy on rare cases associated with using resampling strategies, and the use of biased strategies further increases accuracy over non-biased strategies.

Abstract-The process of decision making in humans involves a combination of the genuine information held by the individual, and the external influence from their social network connections. This helps individuals to make decisions or adopt behaviors, opinions or products. In this work, we seek to investigate under which conditions and with what cost we can form neighborhoods of influence within a social network, in order to assist individuals with little or no prior genuine information through a two-phase recommendation process. Most of the existing approaches regard the problem of identifying influentials as a long-term, network diffusion process, where information cascading occurs in several rounds and has fixed number of influentials. In our approach we consider only one round of influence, which finds applications in settings where timely influence is vital. We tackle the problem by proposing a two-phase framework that aims at identifying influentials in the first phase and form influential neighborhoods to generate recommendations to users with no prior knowledge in the second phase. The difference of the proposed framework with most social recommender systems is that we need to generate recommendations including more than one item and in the absence of explicit ratings, solely relying on the social network's graph.