TY - GEN
T1 - Clustering by Learning the Ordinal Relationships of Qualitative Attribute Values
AU - Wang, Pengkai
AU - Zhang, Yunfan
AU - Zhang, Yiqun
AU - Lu, Yang
AU - Li, Mengke
AU - Cheung, Yiu-ming
N1 - This work was supported in part by the National Natural Science Foundation of China (NSFC) under grants: 62102097, 62306181, and 62376233, the NSFC/Research Grants Council (RGC) Joint Research Scheme under the grant N HKBU214/21, the Natural Science Foundation of Guangdong Province under grants: 2024A1515010163, 2023A1515012855, and 2022A1515011592, the General Research Fund of RGC under grants: 12201321, 12202622, and 12201323, the RGC Senior Research Fellow Scheme with grant: SRFS2324-2S02, and the Science and Technology Program of Guangzhou under grant 202201010548.
Publisher Copyright:
© 2024 IEEE
PY - 2024/6/30
Y1 - 2024/6/30
N2 - In many real-world clustering tasks, data objects are described by both quantitative and qualitative attributes. Attributes with semantically ordered qualitative values are very common and are usually coded according to their order (i.e., consecutive integers) for clustering. However, semantic order is not always globally interdependent with a certain clustering task. An intuitive case is that level of income (attribute) is not always positively correlated with the level of mental health (label). Using mismatched order surely forms a bottleneck to clustering performance, and conversely, the unsupervised clustering process prevents understanding of "true" order. Therefore, we proposed a novel learning paradigm to tune the value order. More specifically, we adjust the intra-attribute orders, and let this process learn mutually with object clustering, thus bridging the gap between value order and clustering task. To the best of our knowledge, this is the first attempt to learn ordinal relationships among qualitative attribute values. Extensive experiments with significance tests show that our method outperforms the existing relevant clustering approaches on qualitative attribute data.
AB - In many real-world clustering tasks, data objects are described by both quantitative and qualitative attributes. Attributes with semantically ordered qualitative values are very common and are usually coded according to their order (i.e., consecutive integers) for clustering. However, semantic order is not always globally interdependent with a certain clustering task. An intuitive case is that level of income (attribute) is not always positively correlated with the level of mental health (label). Using mismatched order surely forms a bottleneck to clustering performance, and conversely, the unsupervised clustering process prevents understanding of "true" order. Therefore, we proposed a novel learning paradigm to tune the value order. More specifically, we adjust the intra-attribute orders, and let this process learn mutually with object clustering, thus bridging the gap between value order and clustering task. To the best of our knowledge, this is the first attempt to learn ordinal relationships among qualitative attribute values. Extensive experiments with significance tests show that our method outperforms the existing relevant clustering approaches on qualitative attribute data.
UR - https://ieeexplore.ieee.org/document/10650134/
U2 - 10.1109/IJCNN60899.2024.10650134
DO - 10.1109/IJCNN60899.2024.10650134
M3 - Conference proceeding
SN - 9798350359329
T3 - Proceedings of the International Joint Conference on Neural Networks
SP - 1
EP - 8
BT - 2024 International Joint Conference on Neural Networks (IJCNN)
PB - IEEE Canada
T2 - 2024 International Joint Conference on Neural Networks (IJCNN)
Y2 - 30 June 2024 through 5 July 2024
ER -