An ensemble method for concept drift in nonstationary environment

Most statistical and data-mining algorithms assume that data come from a stationary distribution. However, in many real-world classification tasks, data arrive over time and the target concept to be learned from the data stream may change accordingly. Many algorithms have been proposed for learning drifting concepts. To deal with the problem of learning when the distribution generating the data changes over time, dynamic weighted majority was proposed as an ensemble method for concept drift. Unfortunately, this technique considers neither the age of the classifiers in the ensemble nor their past correct classification. In this paper, we propose a method that takes into account expert's age as well as its contribution to the global algorithm's accuracy. We evaluate the effectiveness of our proposed method by using m classifiers and training a collection of n-fold partitioning of the data. Experimental results on a benchmark data set show that our method outperforms existing ones.     

Is there a protective role of testosterone against high-grade prostate cancer? Incidence and severity of prostate cancer in 553 patients who underwent prostate biopsy: a prospective data register

This study investigated the role of testosterone replacement therapy (TRT) in prostate safety and cancer progression. A cohort of 553 patients, 42 treated and 162 untreated hypogonadal men, and 349 eugonadal men were included. Pathological analysis of prostate biopsies examining the incidence and severity of PCa revealed that: 16.7% of treated hypogonadal men had a positive biopsy, a Gleason score of ≤6 in 71.4% and >6 in 28.6% of men, a predominant score of 3 and tumour staging of II in 85.7% men; 51.9% of untreated hypogonadal men had a positive biopsy, a Gleason score of ≤6 in 40.5% and >6 in 59.5% men, a predominant score of 3 (77.4%) and tumour staging of II (41.7%) or III (40.5%); 37.8% of eugonadal men had a positive biopsy, a Gleason score of ≤6 in 42.4% and >6 in 57.6% of men, a predominant score of 3 (82.6%) and tumour staging of II (44.7%) or III (47.7%). The incidence of positive prostate biopsies was lowest in hypogonadal men receiving TRT, with significantly lower severity of PCa in terms of staging and grading in the same group. These results suggest that TRT might have a protective effect against high-grade PCa.