There are two well-known characteristics about text classification. One is that the dimension of the sample space is very high, while the number of examples available usually is very small. The other is that the examp...There are two well-known characteristics about text classification. One is that the dimension of the sample space is very high, while the number of examples available usually is very small. The other is that the example vectors are sparse. Meanwhile, we find existing support vector machines active learning approaches are subject to the influence of outliers. Based on these observations, this paper presents a new hybr/d active learning approach. In this approach, to select the unlabelled example(s) to query, the learner takes into account both sparseness and high-dimension characteristics of examples as well as its uncertainty about the examples' categorization. This way, the active learner needs less labeled examples, but still can get a good generalization performance more quickly than competing methods. Our empirical results indicate that this new approach is effective.展开更多
背景以往研究表明精神分裂症患者的神经认知缺损和精神病性症状会导致其职业和社会功能降低。目的评估中国精神分裂症男性住院患者的神经认知和精神病性症状与社会心理功能之间的关系。方法选取上海市精神卫生中心的51例住院男性精神分...背景以往研究表明精神分裂症患者的神经认知缺损和精神病性症状会导致其职业和社会功能降低。目的评估中国精神分裂症男性住院患者的神经认知和精神病性症状与社会心理功能之间的关系。方法选取上海市精神卫生中心的51例住院男性精神分裂症患者,其中40例患者最终完成了个体和社会功能量表(Personal and Social Performance Scale,PSP)中文版、临床疗效总评量表-病情严重程度量表(Clinical Global Impression-Severity,CGI-S)、阳性和阴性症状量表(Positive and Negative Symptom Scale,PANSS)、字母-数字排序以及香港文字记忆学习测试等项目的评定。结果患者PANSS量表的3个临床分量表的分值和社会功能总体评估(PSP总分和CGI-S分值)之间存在明显负相关。患者的神经认知测定结果与症状或社会功能状况均无关。结论对于急性期住院精神分裂症患者而言,临床症状的严重度—而非神经认知缺损程度,与其社会功能水平密切相关。展开更多
文摘There are two well-known characteristics about text classification. One is that the dimension of the sample space is very high, while the number of examples available usually is very small. The other is that the example vectors are sparse. Meanwhile, we find existing support vector machines active learning approaches are subject to the influence of outliers. Based on these observations, this paper presents a new hybr/d active learning approach. In this approach, to select the unlabelled example(s) to query, the learner takes into account both sparseness and high-dimension characteristics of examples as well as its uncertainty about the examples' categorization. This way, the active learner needs less labeled examples, but still can get a good generalization performance more quickly than competing methods. Our empirical results indicate that this new approach is effective.
基金supported by funds awarded to Dr.Linda Byrne from the Adult Mental Health Strategic Research Centrethe School of Psychology at Deakin University,Melbourne,AustraliaDr.Daihui Peng from the Shanghai City Health Bureau Fund(grant number:2010086)
文摘背景以往研究表明精神分裂症患者的神经认知缺损和精神病性症状会导致其职业和社会功能降低。目的评估中国精神分裂症男性住院患者的神经认知和精神病性症状与社会心理功能之间的关系。方法选取上海市精神卫生中心的51例住院男性精神分裂症患者,其中40例患者最终完成了个体和社会功能量表(Personal and Social Performance Scale,PSP)中文版、临床疗效总评量表-病情严重程度量表(Clinical Global Impression-Severity,CGI-S)、阳性和阴性症状量表(Positive and Negative Symptom Scale,PANSS)、字母-数字排序以及香港文字记忆学习测试等项目的评定。结果患者PANSS量表的3个临床分量表的分值和社会功能总体评估(PSP总分和CGI-S分值)之间存在明显负相关。患者的神经认知测定结果与症状或社会功能状况均无关。结论对于急性期住院精神分裂症患者而言,临床症状的严重度—而非神经认知缺损程度,与其社会功能水平密切相关。
