Top 8 most active users in MovieLens data set and statistical analysis results with paired t-test.

The effect size is calculated as (Measurement|t = 0 − Measurement|t = 200)/standard.dev. Bold means significance, and ES means the effect size. Filter bias has a consistently significant and sizable effect on the three measurements across all 8 users. Random selection has less impact on the boundary shift and blind spot. However it significantly decreases the inequality. Active learning aims to help learn correct boundary, therefore it highly depends on the initial data points. Active learning affects points close to the boundary, thus it has limited effects overall.本数据集相关分析中，效应量（effect size）的计算公式为：(t=0时刻的测量值 − t=200时刻的测量值)/标准差（standard deviation）。结果以加粗形式标注时，表示该效应具有统计学显著性，ES即效应量（effect size）。过滤偏差（Filter bias）对全部8名用户的三项测量指标均存在持续显著且规模可观的影响。随机选择（Random selection）对边界偏移（boundary shift）与盲区（blind spot）的影响相对较弱，但可显著降低数据不平等程度。主动学习（Active learning）旨在帮助习得正确的边界，因此其性能高度依赖初始数据点。主动学习仅对靠近该边界的数据点产生影响，故而整体效果有限。