Journal of Computer Applications ›› 2023, Vol. 43 ›› Issue (12): 3779-3789.DOI: 10.11772/j.issn.1001-9081.2022121841
Special Issue: 数据科学与技术
• Data science and technology • Previous Articles Next Articles
Lin SUN1(), Tianjiao MA2, Zhan’ao XUE2,3
Received:
2022-12-09
Revised:
2023-01-29
Accepted:
2023-01-31
Online:
2023-02-17
Published:
2023-12-10
Contact:
Lin SUN
About author:
MA Tianjiao, born in 1998, M. S. candidate. Her research interests include multilabel learning.Supported by:
通讯作者:
孙林
作者简介:
马天娇(1998—),女,河南信阳人,硕士研究生,主要研究方向:多标记学习基金资助:
CLC Number:
Lin SUN, Tianjiao MA, Zhan’ao XUE. Multilabel feature selection algorithm based on Fisher score and fuzzy neighborhood entropy[J]. Journal of Computer Applications, 2023, 43(12): 3779-3789.
孙林, 马天娇, 薛占熬. 基于Fisher score与模糊邻域熵的多标记特征选择算法[J]. 《计算机应用》唯一官方网站, 2023, 43(12): 3779-3789.
Add to citation manager EndNote|Ris|BibTeX
URL: https://www.joca.cn/EN/10.11772/j.issn.1001-9081.2022121841
算法 | 相同点 | 不同点 | 计算复杂度 |
---|---|---|---|
PMU[ | 过滤式方法,考虑了特征子集与标记集的相关性 | 1)运用多元互信息度量特征子集与标记集之间的相关性 2)基于高维联合熵的近似计算设计非转化多标记特征选择 3)需要提前给定特征数 | O(nmz2) |
MUCO[ | 考虑了标记相关性 | 1)使用最大相关最小冗余进行标记相关性分析 2)使用模糊互信息考虑候选特征与所选特征之间的冗余 3)基于杰卡德距离度量计算样本之间的相似度 4)基于模糊互信息设计多标记特征选择 | O(n2) |
MDDM[ | 考虑了特征与标记之间的相关性 | 1)使用希尔伯特-施密特独立性准则描述特征与关联标记之间的 依赖关系,并度量特征与标记之间的相关性 2)将特征空间映射到低维空间设计多标记特征提取方法 | O(m3) |
MFSMR[ | 过滤式方法,使用模糊邻域粗糙集理论,考虑了标记之间的相关性 | 1)使用模糊邻域相似度关系度量标记相关性 2)基于模糊邻域互信息评估特征之间的冗余性和特征与标记之间的相关性 3)基于模糊邻域粗糙集和最大相关性最小冗余度设计特征选择方法 | O(m2(n+m)+z2(n+m)+ nmz+m2n log n) |
MFSR[ | 过滤式方法,考虑了标记之间的相关性 | 1)利用杰卡德距离度量标记之间的相关性 2)使用余弦相似度函数衡量特征的相似度 3)基于样本间相似度函数度量样本在整个样本空间的相似关系 4)基于ReliefF设计多标记特征选择方法 | O(m(|U|+d)),其中d表示 样本xi 未拥有的标记数 |
Tab.1 Similarities and differences and computational complexities between MLFSF and five comparative algorithms
算法 | 相同点 | 不同点 | 计算复杂度 |
---|---|---|---|
PMU[ | 过滤式方法,考虑了特征子集与标记集的相关性 | 1)运用多元互信息度量特征子集与标记集之间的相关性 2)基于高维联合熵的近似计算设计非转化多标记特征选择 3)需要提前给定特征数 | O(nmz2) |
MUCO[ | 考虑了标记相关性 | 1)使用最大相关最小冗余进行标记相关性分析 2)使用模糊互信息考虑候选特征与所选特征之间的冗余 3)基于杰卡德距离度量计算样本之间的相似度 4)基于模糊互信息设计多标记特征选择 | O(n2) |
MDDM[ | 考虑了特征与标记之间的相关性 | 1)使用希尔伯特-施密特独立性准则描述特征与关联标记之间的 依赖关系,并度量特征与标记之间的相关性 2)将特征空间映射到低维空间设计多标记特征提取方法 | O(m3) |
MFSMR[ | 过滤式方法,使用模糊邻域粗糙集理论,考虑了标记之间的相关性 | 1)使用模糊邻域相似度关系度量标记相关性 2)基于模糊邻域互信息评估特征之间的冗余性和特征与标记之间的相关性 3)基于模糊邻域粗糙集和最大相关性最小冗余度设计特征选择方法 | O(m2(n+m)+z2(n+m)+ nmz+m2n log n) |
MFSR[ | 过滤式方法,考虑了标记之间的相关性 | 1)利用杰卡德距离度量标记之间的相关性 2)使用余弦相似度函数衡量特征的相似度 3)基于样本间相似度函数度量样本在整个样本空间的相似关系 4)基于ReliefF设计多标记特征选择方法 | O(m(|U|+d)),其中d表示 样本xi 未拥有的标记数 |
序号 | 数据集 | 样本数 | 特征数 | 标记数 | 平均标记数 | 类型 |
---|---|---|---|---|---|---|
1 | Birds | 645 | 260 | 20 | 1.47 | Audio |
2 | Cal500 | 502 | 68 | 174 | 26.04 | Music |
3 | Computer | 5 000 | 681 | 33 | 1.51 | Text |
4 | Emotion | 593 | 72 | 6 | 1.87 | Music |
5 | Enron | 1 702 | 1 001 | 53 | 3.38 | Text |
6 | Image | 2 000 | 294 | 5 | 1.24 | Image |
7 | Medical | 978 | 1 449 | 45 | 1.25 | Text |
8 | Recreation | 5 000 | 606 | 22 | 1.42 | Text |
9 | Reference | 5 000 | 793 | 33 | 1.17 | Text |
10 | Scene | 2 407 | 294 | 6 | 1.07 | Image |
11 | Yeast | 2 417 | 103 | 14 | 4.24 | Biology |
Tab.2 Details of multilabel datasets
序号 | 数据集 | 样本数 | 特征数 | 标记数 | 平均标记数 | 类型 |
---|---|---|---|---|---|---|
1 | Birds | 645 | 260 | 20 | 1.47 | Audio |
2 | Cal500 | 502 | 68 | 174 | 26.04 | Music |
3 | Computer | 5 000 | 681 | 33 | 1.51 | Text |
4 | Emotion | 593 | 72 | 6 | 1.87 | Music |
5 | Enron | 1 702 | 1 001 | 53 | 3.38 | Text |
6 | Image | 2 000 | 294 | 5 | 1.24 | Image |
7 | Medical | 978 | 1 449 | 45 | 1.25 | Text |
8 | Recreation | 5 000 | 606 | 22 | 1.42 | Text |
9 | Reference | 5 000 | 793 | 33 | 1.17 | Text |
10 | Scene | 2 407 | 294 | 6 | 1.07 | Image |
11 | Yeast | 2 417 | 103 | 14 | 4.24 | Biology |
数据集 | 原始特征数 | PMU | MUCO | MDDM-proj | MDDM-spc | MFSMR | MFSR | MLFSF |
---|---|---|---|---|---|---|---|---|
均值 | 534 | 260 | 181 | 333 | 351 | 111 | 105 | 89 |
Birds | 260 | 260 | 214 | 247 | 191 | 260 | 181 | 169 |
Cal500 | 68 | 20 | 17 | 16 | 16 | 10 | 25 | 23 |
Computer | 681 | 183 | 196 | 477 | 450 | 125 | 125 | 122 |
Emotion | 550 | 72 | 61 | 22 | 31 | 40 | 40 | 35 |
Enron | 72 | 70 | 71 | 150 | 261 | 100 | 125 | 121 |
Image | 1 001 | 81 | 77 | 111 | 133 | 100 | 125 | 113 |
Medical | 1 449 | 1 449 | 606 | 1 332 | 1 449 | 80 | 80 | 73 |
Recreation | 606 | 152 | 180 | 443 | 491 | 200 | 150 | 50 |
Reference | 793 | 196 | 180 | 488 | 473 | 150 | 150 | 150 |
Scene | 294 | 293 | 291 | 290 | 278 | 100 | 100 | 78 |
Yeast | 103 | 85 | 102 | 90 | 91 | 50 | 50 | 47 |
Tab.3 Comparison of seven algorithms on eleven multilabel datasets in terms of FN(↓)
数据集 | 原始特征数 | PMU | MUCO | MDDM-proj | MDDM-spc | MFSMR | MFSR | MLFSF |
---|---|---|---|---|---|---|---|---|
均值 | 534 | 260 | 181 | 333 | 351 | 111 | 105 | 89 |
Birds | 260 | 260 | 214 | 247 | 191 | 260 | 181 | 169 |
Cal500 | 68 | 20 | 17 | 16 | 16 | 10 | 25 | 23 |
Computer | 681 | 183 | 196 | 477 | 450 | 125 | 125 | 122 |
Emotion | 550 | 72 | 61 | 22 | 31 | 40 | 40 | 35 |
Enron | 72 | 70 | 71 | 150 | 261 | 100 | 125 | 121 |
Image | 1 001 | 81 | 77 | 111 | 133 | 100 | 125 | 113 |
Medical | 1 449 | 1 449 | 606 | 1 332 | 1 449 | 80 | 80 | 73 |
Recreation | 606 | 152 | 180 | 443 | 491 | 200 | 150 | 50 |
Reference | 793 | 196 | 180 | 488 | 473 | 150 | 150 | 150 |
Scene | 294 | 293 | 291 | 290 | 278 | 100 | 100 | 78 |
Yeast | 103 | 85 | 102 | 90 | 91 | 50 | 50 | 47 |
指标 | 数据集 | MLKNN | PMU | MUCO | MDDM-proj | MDDM-spc | MFSMR | MFRS | MLFSF |
---|---|---|---|---|---|---|---|---|---|
AP(↑) | Birds | 0.688 9 | 0.688 9 | 0.688 9 | 0.700 2 | 0.703 1 | 0.715 9 | 0.741 0 | |
Cal500 | 0.479 3 | 0.484 8 | 0.482 7 | 0.480 1 | 0.480 2 | 0.490 1 | 0.492 6 | ||
Computer | 0.639 0 | 0.644 7 | 0.658 0 | 0.643 2 | 0.643 8 | 0.638 5 | 0.642 2 | ||
Emotion | 0.757 8 | 0.757 8 | 0.771 5 | 0.772 2 | 0.778 8 | 0.771 6 | 0.799 8 | ||
Enron | 0.544 6 | 0.579 2 | 0.583 4 | 0.583 7 | 0.584 9 | 0.604 7 | 0.622 2 | ||
Image | 0.756 8 | 0.758 4 | 0.757 7 | 0.750 9 | 0.757 8 | 0.756 6 | 0.834 5 | ||
Medical | 0.770 0 | 0.770 0 | 0.778 7 | 0.770 0 | 0.423 6 | 0.423 5 | 0.851 0 | ||
Recreation | 0.469 9 | 0.480 7 | 0.492 4 | 0.488 8 | 0.464 7 | 0.478 7 | 0.520 0 | ||
Reference | 0.617 5 | 0.628 5 | 0.633 8 | 0.623 6 | 0.600 1 | 0.608 6 | 0.647 0 | ||
Scene | 0.851 4 | 0.852 1 | 0.852 5 | 0.852 2 | 0.770 2 | 0.770 2 | 0.856 3 | ||
Yeast | 0.754 1 | 0.757 4 | 0.755 8 | 0.756 9 | 0.733 8 | 0.752 2 | 0.759 4 | ||
均值 | 0.666 3 | 0.673 0 | 0.676 8 | 0.676 8 | 0.640 2 | 0.639 4 | 0.706 0 | ||
HL(↓) | Birds | 0.057 1 | 0.057 1 | 0.056 7 | 0.057 4 | 0.056 2 | 0.057 5 | 0.054 6 | |
Cal500 | 0.140 7 | 0.139 2 | 0.139 8 | 0.140 5 | 0.140 4 | 0.138 5 | |||
Computer | 0.040 2 | 0.038 6 | 0.039 5 | 0.039 7 | 0.038 3 | 0.038 7 | 0.037 8 | ||
Emotion | 0.228 1 | 0.228 1 | 0.230 1 | 0.234 0 | 0.219 8 | 0.218 3 | 0.196 8 | ||
Enron | 0.058 3 | 0.057 4 | 0.056 5 | 0.058 0 | 0.057 9 | 0.054 1 | 0.052 8 | ||
Image | 0.193 2 | 0.192 0 | 0.191 0 | 0.194 0 | 0.194 0 | 0.188 5 | 0.150 7 | ||
Medical | 0.017 3 | 0.017 3 | 0.017 2 | 0.017 3 | 0.027 5 | 0.028 1 | 0.012 6 | ||
Recreation | 0.061 2 | 0.061 1 | 0.060 8 | 0.060 3 | 0.063 6 | 0.063 2 | 0.058 7 | ||
Reference | 0.031 5 | 0.029 1 | 0.030 5 | 0.031 0 | 0.032 5 | 0.029 1 | |||
Scene | 0.097 0 | 0.096 8 | 0.102 1 | 0.092 4 | 0.127 6 | 0.130 2 | 0.092 4 | ||
Yeast | 0.202 1 | 0.200 0 | 0.199 4 | 0.208 8 | 0.201 7 | 0.199 7 | |||
均值 | 0.102 2 | 0.101 8 | 0.101 8 | 0.102 4 | 0.102 4 | 0.104 0 | 0.093 1 | ||
RL(↓) | Birds | 0.126 6 | 0.126 6 | 0.120 0 | 0.125 1 | 0.123 4 | 0.110 1 | 0.097 2 | |
Cal500 | 0.190 7 | 0.189 7 | 0.189 5 | 0.189 8 | 0.189 8 | 0.184 1 | 0.183 2 | ||
Computer | 0.089 1 | 0.087 7 | 0.084 8 | 0.087 9 | 0.089 6 | 0.091 2 | 0.090 4 | ||
Emotion | 0.195 5 | 0.195 5 | 0.230 1 | 0.183 7 | 0.187 1 | 0.192 9 | 0.165 5 | ||
Enron | 0.110 1 | 0.110 0 | 0.107 9 | 0.113 3 | 0.112 3 | 0.104 9 | 0.098 1 | ||
Image | 0.205 1 | 0.203 7 | 0.203 8 | 0.214 1 | 0.203 9 | 0.208 5 | 0.133 3 | ||
Medical | 0.061 4 | 0.061 4 | 0.060 7 | 0.061 4 | 0.132 7 | 0.137 5 | 0.037 3 | ||
Recreation | 0.187 9 | 0.180 1 | 0.180 0 | 0.181 9 | 0.189 3 | 0.180 3 | 0.170 5 | ||
Reference | 0.091 6 | 0.085 7 | 0.087 5 | 0.086 4 | 0.090 0 | 0.091 0 | 0.077 4 | ||
Scene | 0.087 6 | 0.088 0 | 0.091 4 | 0.087 9 | 0.148 2 | 0.142 3 | 0.083 6 | ||
Yeast | 0.174 8 | 0.174 9 | 0.175 0 | 0.172 4 | 0.191 0 | 0.178 0 | 0.172 4 | ||
均值 | 0.138 2 | 0.136 3 | 0.138 0 | 0.136 5 | 0.142 8 | 0.145 6 | 0.119 0 | ||
OE(↓) | Birds | 0.393 8 | 0.393 8 | 0.367 5 | 0.378 3 | 0.376 4 | 0.350 5 | 0.312 2 | |
Cal500 | 0.125 5 | 0.119 5 | 0.121 5 | 0.123 5 | 0.122 8 | 0.114 0 | 0.116 2 | ||
Computer | 0.434 3 | 0.429 7 | 0.413 8 | 0.429 7 | 0.439 1 | 0.436 8 | 0.433 4 | ||
Emotion | 0.362 8 | 0.362 8 | 0.342 8 | 0.340 0 | 0.302 9 | 0.309 5 | 0.272 1 | ||
Enron | 0.438 9 | 0.343 7 | 0.355 5 | 0.342 0 | 0.341 4 | 0.314 7 | 0.296 7 | ||
Image | 0.377 0 | 0.370 0 | 0.371 0 | 0.376 0 | 0.371 0 | 0.373 2 | 0.252 5 | ||
Medical | 0.295 5 | 0.295 5 | 0.275 1 | 0.295 5 | 0.700 0 | 0.690 5 | 0.183 0 | ||
Recreation | 0.684 9 | 0.673 7 | 0.653 0 | 0.659 8 | 0.693 2 | 0.668 2 | 0.614 0 | ||
Reference | 0.478 2 | 0.473 8 | 0.452 5 | 0.476 1 | 0.512 8 | 0.494 9 | 0.452 5 | ||
Scene | 0.248 8 | 0.245 5 | 0.243 5 | 0.243 8 | 0.372 1 | 0.373 8 | 0.239 3 | ||
Yeast | 0.244 5 | 0.229 6 | 0.241 8 | 0.235 2 | 0.235 0 | 0.254 0 | 0.235 9 | ||
均值 | 0.371 3 | 0.358 0 | 0.351 7 | 0.350 3 | 0.393 7 | 0.392 9 | 0.309 5 | ||
CV(↓) | Birds | 3.512 0 | 3.512 0 | 3.326 0 | 3.498 0 | 3.391 0 | 3.145 0 | 2.862 0 | |
Cal500 | 131.500 0 | 131.400 0 | 131.200 0 | 131.400 0 | 131.500 0 | 129.6000 | 129.600 0 | ||
Computer | 4.226 0 | 4.164 0 | 4.0890 | 4.223 0 | 4.252 0 | 4.345 0 | 4.272 0 | ||
Emotion | 1.944 0 | 1.944 0 | 1.901 0 | 1.900 0 | 1.919 0 | 1.970 0 | 1.788 0 | ||
Enron | 14.750 0 | 14.720 0 | 14.520 0 | 15.180 0 | 14.990 0 | 14.450 0 | 13.810 0 | ||
Image | 1.082 0 | 1.088 0 | 1.090 0 | 1.123 0 | 1.084 0 | 1.098 0 | 0.811 4 | ||
Medical | 3.634 0 | 3.634 0 | 3.615 0 | 3.634 0 | 6.846 0 | 7.160 0 | 2.4650 | ||
Recreation | 4.988 0 | 4.801 0 | 4.828 0 | 4.867 0 | 5.011 2 | 4.814 0 | 4.6680 | ||
Reference | 3.498 0 | 3.301 0 | 3.364 0 | 3.319 0 | 3.452 6 | 3.448 3 | 3.0560 | ||
Scene | 0.524 6 | 0.527 5 | 0.523 3 | 0.525 8 | 0.822 5 | 0.801 7 | 0.509 0 | ||
Yeast | 6.449 0 | 6.426 0 | 6.403 0 | 6.406 0 | 6.645 0 | 6.455 0 | 6.3630 | ||
均值 | 16.010 0 | 15.954 0 | 15.988 0 | 15.990 0 | 16.065 0 | 16.139 0 | 15.4760 |
Tab.4 Comparison of eight algorithms on eleven multilabel datasets in terms of five metrics
指标 | 数据集 | MLKNN | PMU | MUCO | MDDM-proj | MDDM-spc | MFSMR | MFRS | MLFSF |
---|---|---|---|---|---|---|---|---|---|
AP(↑) | Birds | 0.688 9 | 0.688 9 | 0.688 9 | 0.700 2 | 0.703 1 | 0.715 9 | 0.741 0 | |
Cal500 | 0.479 3 | 0.484 8 | 0.482 7 | 0.480 1 | 0.480 2 | 0.490 1 | 0.492 6 | ||
Computer | 0.639 0 | 0.644 7 | 0.658 0 | 0.643 2 | 0.643 8 | 0.638 5 | 0.642 2 | ||
Emotion | 0.757 8 | 0.757 8 | 0.771 5 | 0.772 2 | 0.778 8 | 0.771 6 | 0.799 8 | ||
Enron | 0.544 6 | 0.579 2 | 0.583 4 | 0.583 7 | 0.584 9 | 0.604 7 | 0.622 2 | ||
Image | 0.756 8 | 0.758 4 | 0.757 7 | 0.750 9 | 0.757 8 | 0.756 6 | 0.834 5 | ||
Medical | 0.770 0 | 0.770 0 | 0.778 7 | 0.770 0 | 0.423 6 | 0.423 5 | 0.851 0 | ||
Recreation | 0.469 9 | 0.480 7 | 0.492 4 | 0.488 8 | 0.464 7 | 0.478 7 | 0.520 0 | ||
Reference | 0.617 5 | 0.628 5 | 0.633 8 | 0.623 6 | 0.600 1 | 0.608 6 | 0.647 0 | ||
Scene | 0.851 4 | 0.852 1 | 0.852 5 | 0.852 2 | 0.770 2 | 0.770 2 | 0.856 3 | ||
Yeast | 0.754 1 | 0.757 4 | 0.755 8 | 0.756 9 | 0.733 8 | 0.752 2 | 0.759 4 | ||
均值 | 0.666 3 | 0.673 0 | 0.676 8 | 0.676 8 | 0.640 2 | 0.639 4 | 0.706 0 | ||
HL(↓) | Birds | 0.057 1 | 0.057 1 | 0.056 7 | 0.057 4 | 0.056 2 | 0.057 5 | 0.054 6 | |
Cal500 | 0.140 7 | 0.139 2 | 0.139 8 | 0.140 5 | 0.140 4 | 0.138 5 | |||
Computer | 0.040 2 | 0.038 6 | 0.039 5 | 0.039 7 | 0.038 3 | 0.038 7 | 0.037 8 | ||
Emotion | 0.228 1 | 0.228 1 | 0.230 1 | 0.234 0 | 0.219 8 | 0.218 3 | 0.196 8 | ||
Enron | 0.058 3 | 0.057 4 | 0.056 5 | 0.058 0 | 0.057 9 | 0.054 1 | 0.052 8 | ||
Image | 0.193 2 | 0.192 0 | 0.191 0 | 0.194 0 | 0.194 0 | 0.188 5 | 0.150 7 | ||
Medical | 0.017 3 | 0.017 3 | 0.017 2 | 0.017 3 | 0.027 5 | 0.028 1 | 0.012 6 | ||
Recreation | 0.061 2 | 0.061 1 | 0.060 8 | 0.060 3 | 0.063 6 | 0.063 2 | 0.058 7 | ||
Reference | 0.031 5 | 0.029 1 | 0.030 5 | 0.031 0 | 0.032 5 | 0.029 1 | |||
Scene | 0.097 0 | 0.096 8 | 0.102 1 | 0.092 4 | 0.127 6 | 0.130 2 | 0.092 4 | ||
Yeast | 0.202 1 | 0.200 0 | 0.199 4 | 0.208 8 | 0.201 7 | 0.199 7 | |||
均值 | 0.102 2 | 0.101 8 | 0.101 8 | 0.102 4 | 0.102 4 | 0.104 0 | 0.093 1 | ||
RL(↓) | Birds | 0.126 6 | 0.126 6 | 0.120 0 | 0.125 1 | 0.123 4 | 0.110 1 | 0.097 2 | |
Cal500 | 0.190 7 | 0.189 7 | 0.189 5 | 0.189 8 | 0.189 8 | 0.184 1 | 0.183 2 | ||
Computer | 0.089 1 | 0.087 7 | 0.084 8 | 0.087 9 | 0.089 6 | 0.091 2 | 0.090 4 | ||
Emotion | 0.195 5 | 0.195 5 | 0.230 1 | 0.183 7 | 0.187 1 | 0.192 9 | 0.165 5 | ||
Enron | 0.110 1 | 0.110 0 | 0.107 9 | 0.113 3 | 0.112 3 | 0.104 9 | 0.098 1 | ||
Image | 0.205 1 | 0.203 7 | 0.203 8 | 0.214 1 | 0.203 9 | 0.208 5 | 0.133 3 | ||
Medical | 0.061 4 | 0.061 4 | 0.060 7 | 0.061 4 | 0.132 7 | 0.137 5 | 0.037 3 | ||
Recreation | 0.187 9 | 0.180 1 | 0.180 0 | 0.181 9 | 0.189 3 | 0.180 3 | 0.170 5 | ||
Reference | 0.091 6 | 0.085 7 | 0.087 5 | 0.086 4 | 0.090 0 | 0.091 0 | 0.077 4 | ||
Scene | 0.087 6 | 0.088 0 | 0.091 4 | 0.087 9 | 0.148 2 | 0.142 3 | 0.083 6 | ||
Yeast | 0.174 8 | 0.174 9 | 0.175 0 | 0.172 4 | 0.191 0 | 0.178 0 | 0.172 4 | ||
均值 | 0.138 2 | 0.136 3 | 0.138 0 | 0.136 5 | 0.142 8 | 0.145 6 | 0.119 0 | ||
OE(↓) | Birds | 0.393 8 | 0.393 8 | 0.367 5 | 0.378 3 | 0.376 4 | 0.350 5 | 0.312 2 | |
Cal500 | 0.125 5 | 0.119 5 | 0.121 5 | 0.123 5 | 0.122 8 | 0.114 0 | 0.116 2 | ||
Computer | 0.434 3 | 0.429 7 | 0.413 8 | 0.429 7 | 0.439 1 | 0.436 8 | 0.433 4 | ||
Emotion | 0.362 8 | 0.362 8 | 0.342 8 | 0.340 0 | 0.302 9 | 0.309 5 | 0.272 1 | ||
Enron | 0.438 9 | 0.343 7 | 0.355 5 | 0.342 0 | 0.341 4 | 0.314 7 | 0.296 7 | ||
Image | 0.377 0 | 0.370 0 | 0.371 0 | 0.376 0 | 0.371 0 | 0.373 2 | 0.252 5 | ||
Medical | 0.295 5 | 0.295 5 | 0.275 1 | 0.295 5 | 0.700 0 | 0.690 5 | 0.183 0 | ||
Recreation | 0.684 9 | 0.673 7 | 0.653 0 | 0.659 8 | 0.693 2 | 0.668 2 | 0.614 0 | ||
Reference | 0.478 2 | 0.473 8 | 0.452 5 | 0.476 1 | 0.512 8 | 0.494 9 | 0.452 5 | ||
Scene | 0.248 8 | 0.245 5 | 0.243 5 | 0.243 8 | 0.372 1 | 0.373 8 | 0.239 3 | ||
Yeast | 0.244 5 | 0.229 6 | 0.241 8 | 0.235 2 | 0.235 0 | 0.254 0 | 0.235 9 | ||
均值 | 0.371 3 | 0.358 0 | 0.351 7 | 0.350 3 | 0.393 7 | 0.392 9 | 0.309 5 | ||
CV(↓) | Birds | 3.512 0 | 3.512 0 | 3.326 0 | 3.498 0 | 3.391 0 | 3.145 0 | 2.862 0 | |
Cal500 | 131.500 0 | 131.400 0 | 131.200 0 | 131.400 0 | 131.500 0 | 129.6000 | 129.600 0 | ||
Computer | 4.226 0 | 4.164 0 | 4.0890 | 4.223 0 | 4.252 0 | 4.345 0 | 4.272 0 | ||
Emotion | 1.944 0 | 1.944 0 | 1.901 0 | 1.900 0 | 1.919 0 | 1.970 0 | 1.788 0 | ||
Enron | 14.750 0 | 14.720 0 | 14.520 0 | 15.180 0 | 14.990 0 | 14.450 0 | 13.810 0 | ||
Image | 1.082 0 | 1.088 0 | 1.090 0 | 1.123 0 | 1.084 0 | 1.098 0 | 0.811 4 | ||
Medical | 3.634 0 | 3.634 0 | 3.615 0 | 3.634 0 | 6.846 0 | 7.160 0 | 2.4650 | ||
Recreation | 4.988 0 | 4.801 0 | 4.828 0 | 4.867 0 | 5.011 2 | 4.814 0 | 4.6680 | ||
Reference | 3.498 0 | 3.301 0 | 3.364 0 | 3.319 0 | 3.452 6 | 3.448 3 | 3.0560 | ||
Scene | 0.524 6 | 0.527 5 | 0.523 3 | 0.525 8 | 0.822 5 | 0.801 7 | 0.509 0 | ||
Yeast | 6.449 0 | 6.426 0 | 6.403 0 | 6.406 0 | 6.645 0 | 6.455 0 | 6.3630 | ||
均值 | 16.010 0 | 15.954 0 | 15.988 0 | 15.990 0 | 16.065 0 | 16.139 0 | 15.4760 |
指标 | FF | |
---|---|---|
AP | 21.010 4 | 4.670 0 |
HL | 23.045 6 | 5.365 1 |
RL | 20.931 8 | 4.644 5 |
OE | 15.624 8 | 2.545 8 |
CV | 19.297 9 | 3.344 4 |
Tab.5 Statistical results of five metrics for seven algorithms
指标 | FF | |
---|---|---|
AP | 21.010 4 | 4.670 0 |
HL | 23.045 6 | 5.365 1 |
RL | 20.931 8 | 4.644 5 |
OE | 15.624 8 | 2.545 8 |
CV | 19.297 9 | 3.344 4 |
1 | SUN L, YIN T, DING W, et al. Multilabel feature selection using ML-ReliefF and neighborhood mutual information for multilabel neighborhood decision systems[J]. Information Sciences, 2020, 537: 401-424. 10.1016/j.ins.2020.05.102 |
2 | 张志浩, 林耀进, 卢舜, 等. 缺失标记下基于类属属性的多标记特征选择[J]. 计算机应用, 2021, 41(10): 2849-2857. 10.11772/j.issn.1001-9081.2020111893 |
ZHANG Z H, LIN Y J, LU S, et al. Multi-label feature selection based on label-specific feature with missing labels[J]. Journal of Computer Applications, 2021, 41(10): 2849-2857. 10.11772/j.issn.1001-9081.2020111893 | |
3 | 孙林,黄苗苗, 徐久成. 基于邻域粗糙集和Relief的弱标记特征选择方法[J]. 计算机科学, 2022, 49(4): 152-160. 10.11896/jsjkx.210300094 |
SUN L, HUANG M M, XU J C. Weak label feature selection method based on neighborhood rough sets and Relief[J]. Computer Science, 2022, 49(4): 152-160. 10.11896/jsjkx.210300094 | |
4 | 阮梓航, 肖先勇, 胡文曦, 等. 基于多粒度特征选择和模型融合的复合电能质量扰动分类特征优化[J]. 电力系统保护与控制, 2022, 50(14): 1-10. |
RUAN Z H, XIAO X Y, HU W X, et al. Multiple power quality disturbance classification feature optimization based on multi-granularity feature selection and model fusion [J]. Power System Protection and Control, 2022, 50(14): 1-10. | |
5 | 滕俊元, 高猛, 郑小萌, 等. 噪声可容忍的软件缺陷预测特征选择方法[J]. 计算机科学, 2021, 48(12): 131-139. 10.11896/jsjkx.201000168 |
TENG J Y, GAO M, ZHENG X M, et al. Noise tolerable feature selection method for software defect prediction [J]. Computer Science, 2021, 48(12): 131-139. 10.11896/jsjkx.201000168 | |
6 | SUN L, WANG T, DING W, et al. Feature selection using Fisher score and multilabel neighborhood rough sets for multilabel classification [J]. Information Sciences, 2021, 578: 887-912. 10.1016/j.ins.2021.08.032 |
7 | 汪正凯, 沈东升, 王晨曦. 基于文本分类的Fisher Score快速多标记特征选择算法[J]. 计算机工程, 2022, 48(2): 113-124. 10.19678/j.issn.1000-3428.0060594 |
WANG Z K, SHEN D S, WANG C X. Fisher Score fast multi-label feature selection algorithm based on text classification[J]. Computer Engineering, 2022, 48(2): 113-124. 10.19678/j.issn.1000-3428.0060594 | |
8 | GUYON I, WESTON J, BARNHILL S, et al. Gene selection for cancer classification using support vector machines[J]. Machine Learning, 2002, 46: 389-422. 10.1023/a:1012487302797 |
9 | GÜNES S, POLAT K, YOSUNKAYA S. Multi-class f-score feature selection approach to classification of obstructive sleep apnea syndrome[J]. Expert Systems with Applications, 2010, 37(2): 998-1004. 10.1016/j.eswa.2009.05.075 |
10 | 孙林, 黄金旭, 徐久成. 基于邻域容差互信息和鲸鱼优化算法的非平衡数据特征选择[J].计算机应用, 2023, 43(6): 1842-1854. 10.11772/j.issn.1001-9081.2022050691 |
SUN L, HUANG J X, XU J C. Feature selection for imbalanced data based on neighborhood tolerance mutual information and whale optimization [J]. Journal of Computer Applications, 2023, 43(6): 1842-1854. 10.11772/j.issn.1001-9081.2022050691 | |
11 | HANCER E, XUE B, ZHANG M. Differential evolution for filter feature selection based on information theory and feature ranking[J]. Knowledge-Based Systems, 2018, 140: 103-119. 10.1016/j.knosys.2017.10.028 |
12 | 吴迪, 郭嗣琮. 改进的Fisher Score特征选择方法及其应用[J]. 辽宁工程技术大学学报(自然科学版), 2019, 38(5): 472-479. |
WU D, GUO S Z. An improved Fisher Score feature selection method and its application[J]. Journal of Liaoning Technical University (Natural Science), 2019, 38(5): 472-479. | |
13 | 段洁, 胡清华, 张灵均, 等. 基于邻域粗糙集的多标记分类特征选择算法[J]. 计算机研究与发展, 2015, 52(1): 56-65. 10.7544/issn1000-1239.2015.20140544 |
DUAN J, HU Q H, ZHANG L J, et al. Feature selection for multi-label classification based on neighborhood rough sets[J]. Journal of Computer Research and Development, 2015, 52(1): 56-65. 10.7544/issn1000-1239.2015.20140544 | |
14 | LIN Y, HU Q, LIU J, et al. Multi-label feature selection based on neighborhood mutual information[J]. Applied Software Computation, 2016, 38: 244-256. 10.1016/j.asoc.2015.10.009 |
15 | LIU J, LIN Y, LI Y, et al. Online multi-label streaming feature selection based on neighborhood rough set[J]. Pattern Recognition, 2018, 84: 273-287. 10.1016/j.patcog.2018.07.021 |
16 | HUANG M, SUN L, XU J, et al. Multilabel feature selection using Relief and minimum redundancy maximum relevance based on neighborhood rough sets [J]. IEEE Access, 2020, 8: 62011-62031. 10.1109/access.2020.2982536 |
17 | WU Y, LIU J, YU X, et al. Neighborhood rough set based multi-label feature selection with label correlation[J]. Concurrency and Computation: Practice and Experience, 2022, 34(22): e7162. 10.1002/cpe.7162 |
18 | CHEN P, LIN M, LIU J. Multi-label attribute reduction based on variable precision fuzzy neighborhood rough set [J]. IEEE Access, 2020, 8: 133565-133576. 10.1109/access.2020.3010314 |
19 | SUN L, YIN T, DING W, et al. Feature selection with missing labels using multilabel fuzzy neighborhood rough sets and maximum relevance minimum redundancy [J]. IEEE Transactions on Fuzzy Systems, 2022, 30(5): 1197-1211. 10.1109/tfuzz.2021.3053844 |
20 | XU J, SHEN K, SUN L. Multi-label feature selection based on fuzzy neighborhood rough sets [J]. Complex & Intelligent Systems, 2022, 8: 2105-2129. 10.1007/s40747-021-00636-y |
21 | 张大斌,张博婷,凌立文,等.基于二次分解聚合策略的我国碳交易价格预测[J].系统科学与数学, 2022, 42(11): 3094-3106. 10.12341/jssms22287 |
ZHANG D B, ZHANG B T, LING L W, et al. Carbon price forecasting based on secondary decomposition and aggregation strategy[J]. Journal of Systems Science and Mathematical Sciences, 2022, 42(11): 3094-3106. 10.12341/jssms22287 | |
22 | RESHEF D N, RESHEF Y A, FINUCANE H K, et al. Detecting novel associations in large data sets[J]. Science, 2011, 334(6062): 1518-1524. 10.1126/science.1205438 |
23 | 刘琨, 封硕. 加强局部搜索能力的人工蜂群算法[J]. 河南师范大学学报(自然科学版), 2021, 49(2): 15-24. 10.16366/j.cnki.1000-2367.2021.02.003 |
LIU K, FENG S. An improved artificial bee colony algorithm for enhancing local search ability[J]. Journal of Henan Normal University (Natural Science Edition), 2021, 49(2): 15-24. 10.16366/j.cnki.1000-2367.2021.02.003 | |
24 | SARWAR B, KARYPIS G, KONSTAN J, et al. Item-based collaborative filtering recommendation algorithms[C]// Proceedings of the 10th International Conference on World Wide Web. New York: ACM, 2001: 285-295. 10.1145/371920.372071 |
25 | 黄剑湘, 林铮, 刘可真, 等. 考虑换流站海量事件的关联规则挖掘分析方法[J]. 电力系统保护与控制, 2022, 50(12): 117-125. |
HUANG J X, LIN Z, LIU K Z, et al. Association rule mining analysis method considering massive events in a converter station [J]. Power System Protection and Control, 2022, 50(12): 117-125. | |
26 | 余鹰. 多标记学习研究综述[J]. 计算机工程与应用, 2015, 51(17): 20-27. 10.3778/j.issn.1002-8331.1506-0063 |
YU Y. Survey on multi-label learning[J]. Computer Engineering and Applications, 2015, 51(17): 20-27. 10.3778/j.issn.1002-8331.1506-0063 | |
27 | ZHENG T, ZHU L. Uncertainty measures of neighborhood system based rough sets[J]. Knowledge Based Systems, 2015, 86: 57-65. 10.1016/j.knosys.2015.05.021 |
28 | 刘艳, 程璐, 孙林. 基于K-S检验和邻域粗糙集的特征选择方法[J]. 河南师范大学学报(自然科学版), 2019, 47(2): 21-28. 10.16366/j.cnki.1000-2367.2019.02.004 |
LIU Y, CHENG L, SUN L. Feature selection method based on K-S test and neighborhood rough sets[J]. Journal of Henan Normal University (Natural Science Edition), 2019, 47(2): 21-28. 10.16366/j.cnki.1000-2367.2019.02.004 | |
29 | 姚晟, 徐风, 赵鹏, 等. 基于改进邻域粒的模糊熵特征选择算法[J].南京大学学报(自然科学), 2017, 53(4): 802-814. 10.13232/j.cnki.jnju.2017.04.024 |
YAO S, XU F, ZHAO P, et al. Fuzzy entropy feature selection algorithm based on improved neighborhood granule [J]. Journal of Nanjing University (Natural Science), 2017, 53(4): 802-814. 10.13232/j.cnki.jnju.2017.04.024 | |
30 | LEE J, KIM D-W. Feature selection for multi-label classification using multivariate mutual information[J]. Pattern Recognition Letters, 2013, 34(3): 349-357. 10.1016/j.patrec.2012.10.005 |
31 | LIN Y, HU Q, LIU J, et al. Streaming feature selection for multilabel learning based on fuzzy mutual information[J]. IEEE Transactions on Fuzzy Systems, 2017, 25(6): 1491-1507. 10.1109/tfuzz.2017.2735947 |
32 | ZHANG Y, ZHOU Z-H. Multilabel dimensionality reduction via dependence maximization[J]. ACM Transactions on Knowledge Discovery from Data, 2010, 4(3): Article No. 14. 10.1145/1839490.1839495 |
33 | 孙林, 陈雨生, 徐久成. 基于改进ReliefF的多标记特征选择算法[J]. 山东大学学报(理学版), 2022, 57(4): 1-11. |
SUN L, CHEN Y S, XU J C. Multilabel feature selection algorithm based on improved ReliefF [J]. Journal of Shandong University (Natural Science), 2022, 57(4):1-11. | |
34 | SCHAPIRE R E, SINGER Y. BoosTexter: a boosting-based system for text categorization[J]. Machine Learning, 2000, 39: 135-168. 10.1023/a:1007649029923 |
35 | TSOUMAKAS G, VLAHAVAS I. Random k-labelsets: an ensemble method for multilabel classification [C]// Proceedings of the 2007 European Conference on Machine Learning. Berlin: Springer, 2007: 406-417. |
36 | CHEN L, CHEN D. Alignment based feature selection for multi-label learning[J]. Neural Processing Letters, 2019, 50: 2323-2344. 10.1007/s11063-019-10009-9 |
[1] | Hong CHEN, Bing QI, Haibo JIN, Cong WU, Li’ang ZHANG. Class-imbalanced traffic abnormal detection based on 1D-CNN and BiGRU [J]. Journal of Computer Applications, 2024, 44(8): 2493-2499. |
[2] | Lin GAO, Yu ZHOU, Tak Wu KWONG. Evolutionary bi-level adaptive local feature selection [J]. Journal of Computer Applications, 2024, 44(5): 1408-1414. |
[3] | Mingzhu LEI, Hao WANG, Rong JIA, Lin BAI, Xiaoying PAN. Oversampling algorithm based on synthesizing minority class samples using relationship between features [J]. Journal of Computer Applications, 2024, 44(5): 1428-1436. |
[4] | Dapeng XU, Xinmin HOU. Feature selection method for graph neural network based on network architecture design [J]. Journal of Computer Applications, 2024, 44(3): 663-670. |
[5] | Shengjie MENG, Wanjun YU, Ying CHEN. Feature selection algorithm for high-dimensional data with maximum correlation and maximum difference [J]. Journal of Computer Applications, 2024, 44(3): 767-771. |
[6] | Lin SUN, Menghan LIU. K-means clustering based on adaptive cuckoo optimization feature selection [J]. Journal of Computer Applications, 2024, 44(3): 831-841. |
[7] | Jingxin LIU, Wenjing HUANG, Liangsheng XU, Chong HUANG, Jiansheng WU. Unsupervised feature selection model with dictionary learning and sample correlation preservation [J]. Journal of Computer Applications, 2024, 44(12): 3766-3775. |
[8] | Tian HE, Zongxin SHEN, Qianqian HUANG, Yanyong HUANG. Adaptive learning-based multi-view unsupervised feature selection method [J]. Journal of Computer Applications, 2023, 43(9): 2657-2664. |
[9] | Lin SUN, Jinxu HUANG, Jiucheng XU. Feature selection for imbalanced data based on neighborhood tolerance mutual information and whale optimization algorithm [J]. Journal of Computer Applications, 2023, 43(6): 1842-1854. |
[10] | Zhenhua YU, Zhengqi LIU, Ying LIU, Cheng GUO. Feature selection method based on self-adaptive hybrid particle swarm optimization for software defect prediction [J]. Journal of Computer Applications, 2023, 43(4): 1206-1213. |
[11] | Jingcheng XU, Xuebin CHEN, Yanling DONG, Jia YANG. DDoS attack detection by random forest fused with feature selection [J]. Journal of Computer Applications, 2023, 43(11): 3497-3503. |
[12] | Lei MA, Chuan LUO, Tianrui LI, Hongmei CHEN. Fuzzy-rough set based unsupervised dynamic feature selection algorithm [J]. Journal of Computer Applications, 2023, 43(10): 3121-3128. |
[13] | Liang CHEN, Xianfeng TANG. Improved sine cosine algorithm for optimizing feature selection and data classification [J]. Journal of Computer Applications, 2022, 42(6): 1852-1861. |
[14] | Jing ZHAO, Jingyu HAN, Long QIAN, Yi MAO. ECG diagnostic classification based on improved RAKEL algorithm [J]. Journal of Computer Applications, 2022, 42(6): 1892-1897. |
[15] | Li LI, Kexin SHI, Zhenkang REN. Cross-project defect prediction method based on feature selection and TrAdaBoost [J]. Journal of Computer Applications, 2022, 42(5): 1554-1562. |
Viewed | ||||||
Full text |
|
|||||
Abstract |
|
|||||