Abstract

Abstract
Ensemble methods have been successfully used in the past to improve recommender systems; however, they have never been studied with incremental recommendation algorithms. Many online recommender systems deal with continuous, potentially fast, and unbounded flows of databig data streamsand often need to be responsive to fresh user feedback, adjusting recommendations accordingly. This is clear in tasks such as social network feeds, news recommender systems, automatic playlist completion, and other similar applications. Batch ensemble approaches are not suitable to perform continuous learning, given the complexity of retraining new models on demand. In this paper, we adapt a general purpose online bagging algorithm for top-N recommendation tasks and propose two novel online bagging methods specifically tailored for recommender systems. We evaluate the three approaches, using an incremental matrix factorization algorithm for top-N recommendation with positive-only user feedback data as the base model. Our results show that online bagging is able to improve accuracy up to 55% over the baseline, with manageable computational overhead.

Abstract
Exceptional preferences mining (EPM) is a crossover between two subfields of data mining: local pattern mining and preference learning. EPM can be seen as a local pattern mining task that finds subsets of observations where some preference relations between labels significantly deviate from the norm. It is a variant of subgroup discovery, with rankings of labels as the target concept. We employ several quality measures that highlight subgroups featuring exceptional preferences, where the focus of what constitutes exceptional' varies with the quality measure: two measures look for exceptional overall ranking behavior, one measure indicates whether a particular label stands out from the rest, and a fourth measure highlights subgroups with unusual pairwise label ranking behavior. We explore a few datasets and compare with existing techniques. The results confirm that the new task EPM can deliver interesting knowledge.

Abstract
Transthyretin Familial Amyloid Polyneuropathy (TTR-FAP) is a neurological genetic disease that propagates from one family generation to the next. The disease can have severe effects on the life of patients after the first symptoms (onset) appear. Accurate prediction of the age of onset for these patients can help the management of the impact. This is, however, a challenging problem since both familial and non-familial characteristics may or may not affect the age of onset. In this work, we assess the importance of sets of genealogical features used for Predicting the Age of Onset of TTR-FAP Patients. We study three sets of features engineered from clinical and genealogical data records obtained from Portuguese patients. These feature sets, referred to as Patient, First Level and Extended Level Features, represent sets of characteristics related to each patient's attributes and their familial relations. They were compiled by a Medical Research Center working with TTR-FAP patients. Our results show the importance of genealogical data when clinical records have no information related with the ancestor of the patient, namely its Gender and Age of Onset. This is suggested by the improvement of the estimated predictive error results after combining First and Extended Level with the Patients Features.

Abstract
This work describes a problem oriented approach to analyze and predict the Age of Onset of Patients diagnosed with Transthyretin Familial Amyloid Polyneuropathy (TTR-FAP). We constructed, from a set of clinical and familial records, three sets of features which represent different characteristics of a patient, before becoming symptomatic. Using those features, we tested a set of machine learning regression methods, namely Decision Tree (Regression Tree), Elastic Net, Lasso, Linear Regression, Random Forest Regressor, Ridge Regression and Support Vector Machine Regressor (SVM). Later, we defined a baseline model that represents the current medical practice to serve as a guideline for us to measure the accuracy of our approach. Our results show a significant improvement of machine learning methods when compared with the current baseline. © 2018 IEEE.

Abstract