The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Regresi ordinal digunakan untuk mengklasifikasikan kejadian dengan mempertimbangkan hubungan ordinal antara label. Kaedah sedia ada cenderung untuk mengurangkan ketepatan apabila mereka mematuhi pemeliharaan hubungan ordinal. Oleh itu, kami mencadangkan rangkaian berasaskan pengetahuan pengedaran (DK-net) yang mempertimbangkan hubungan ordinal sambil mengekalkan ketepatan yang tinggi. DK-net memfokuskan pada set data imej. Walau bagaimanapun, dalam aplikasi industri, seseorang boleh mencari bukan sahaja data imej tetapi juga data jadual. Dalam kajian ini, kami mencadangkan DK-neural oblivious decision ensemble (NODE), versi DK-net yang lebih baik untuk data jadual. DK-NODE menggunakan NODE untuk pengekstrakan ciri. Di samping itu, kami mencadangkan kaedah untuk melaraskan parameter yang mengawal tahap pematuhan dengan hubungan ordinal. Kami bereksperimen dengan tiga set data: WineQuality, Abalone, dan dataset Eucalyptus. Eksperimen menunjukkan bahawa kaedah yang dicadangkan mencapai ketepatan yang tinggi dan MAE kecil pada tiga set data. Terutama, kaedah yang dicadangkan mempunyai purata MAE terkecil pada semua set data.
Yoshiyuki TAJIMA
Yokohama National University
Tomoki HAMAGAMI
Yokohama National University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Salinan
Yoshiyuki TAJIMA, Tomoki HAMAGAMI, "Ordinal Regression Based on the Distributional Distance for Tabular Data" in IEICE TRANSACTIONS on Information,
vol. E106-D, no. 3, pp. 357-364, March 2023, doi: 10.1587/transinf.2022EDP7071.
Abstract: Ordinal regression is used to classify instances by considering ordinal relation between labels. Existing methods tend to decrease the accuracy when they adhere to the preservation of the ordinal relation. Therefore, we propose a distributional knowledge-based network (DK-net) that considers ordinal relation while maintaining high accuracy. DK-net focuses on image datasets. However, in industrial applications, one can find not only image data but also tabular data. In this study, we propose DK-neural oblivious decision ensemble (NODE), an improved version of DK-net for tabular data. DK-NODE uses NODE for feature extraction. In addition, we propose a method for adjusting the parameter that controls the degree of compliance with the ordinal relation. We experimented with three datasets: WineQuality, Abalone, and Eucalyptus dataset. The experiments showed that the proposed method achieved high accuracy and small MAE on three datasets. Notably, the proposed method had the smallest average MAE on all datasets.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2022EDP7071/_p
Salinan
@ARTICLE{e106-d_3_357,
author={Yoshiyuki TAJIMA, Tomoki HAMAGAMI, },
journal={IEICE TRANSACTIONS on Information},
title={Ordinal Regression Based on the Distributional Distance for Tabular Data},
year={2023},
volume={E106-D},
number={3},
pages={357-364},
abstract={Ordinal regression is used to classify instances by considering ordinal relation between labels. Existing methods tend to decrease the accuracy when they adhere to the preservation of the ordinal relation. Therefore, we propose a distributional knowledge-based network (DK-net) that considers ordinal relation while maintaining high accuracy. DK-net focuses on image datasets. However, in industrial applications, one can find not only image data but also tabular data. In this study, we propose DK-neural oblivious decision ensemble (NODE), an improved version of DK-net for tabular data. DK-NODE uses NODE for feature extraction. In addition, we propose a method for adjusting the parameter that controls the degree of compliance with the ordinal relation. We experimented with three datasets: WineQuality, Abalone, and Eucalyptus dataset. The experiments showed that the proposed method achieved high accuracy and small MAE on three datasets. Notably, the proposed method had the smallest average MAE on all datasets.},
keywords={},
doi={10.1587/transinf.2022EDP7071},
ISSN={1745-1361},
month={March},}
Salinan
TY - JOUR
TI - Ordinal Regression Based on the Distributional Distance for Tabular Data
T2 - IEICE TRANSACTIONS on Information
SP - 357
EP - 364
AU - Yoshiyuki TAJIMA
AU - Tomoki HAMAGAMI
PY - 2023
DO - 10.1587/transinf.2022EDP7071
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E106-D
IS - 3
JA - IEICE TRANSACTIONS on Information
Y1 - March 2023
AB - Ordinal regression is used to classify instances by considering ordinal relation between labels. Existing methods tend to decrease the accuracy when they adhere to the preservation of the ordinal relation. Therefore, we propose a distributional knowledge-based network (DK-net) that considers ordinal relation while maintaining high accuracy. DK-net focuses on image datasets. However, in industrial applications, one can find not only image data but also tabular data. In this study, we propose DK-neural oblivious decision ensemble (NODE), an improved version of DK-net for tabular data. DK-NODE uses NODE for feature extraction. In addition, we propose a method for adjusting the parameter that controls the degree of compliance with the ordinal relation. We experimented with three datasets: WineQuality, Abalone, and Eucalyptus dataset. The experiments showed that the proposed method achieved high accuracy and small MAE on three datasets. Notably, the proposed method had the smallest average MAE on all datasets.
ER -