Multi-dimensional classification via stacked dependency exploitation
- PDF / 330,412 Bytes
- 14 Pages / 595 x 842 pts (A4) Page_size
- 14 Downloads / 213 Views
. RESEARCH PAPER .
December 2020, Vol. 63 222102:1–222102:14 https://doi.org/10.1007/s11432-019-2905-3
Multi-dimensional classification via stacked dependency exploitation Bin-Bin JIA1,2,3 & Min-Ling ZHANG1,3,4* 1 School of Computer Science and Engineering, Southeast University, Nanjing 210096, China; College of Electrical and Information Engineering, Lanzhou University of Technology, Lanzhou 730050, China; 3 Key Laboratory of Computer Network and Information Integration (Southeast University), Ministry of Education, China; 4 Collaborative Innovation Center of Wireless Communications Technology, Nanjing 210096, China
2
Received 14 November 2019/Revised 15 February 2020/Accepted 21 March 2020/Published online 9 November 2020
Abstract Multi-dimensional classification (MDC) aims to build classification models for multiple heterogenous class spaces simultaneously, where each class space characterizes the semantics of an object w.r.t. one specific dimension. Modeling dependencies among class spaces plays a key role in solving MDC tasks, where most approaches work by assuming directed acyclic graph (DAG) structure or random chaining structure over class spaces. Different from existing probabilistic strategies, a deterministic strategy named Seem for dependency modeling is proposed in this paper via stacked dependency exploitation. In the first-level, pairwise dependencies are considered which can be modeled more reliably than modeling full dependencies among all class spaces by DAG or chaining structure. In the second-level, the class label of unseen instance w.r.t. each class space is determined by adaptively stacking predictive outputs from first-level pairwise classifiers. Experimental results show that stacked dependency exploitation leads to superior performance against stateof-the-art MDC approaches. Keywords machine learning, multi-dimensional classification, class dependencies, deterministic strategy, stacked dependency exploitation Citation Jia B-B, Zhang M-L. Multi-dimensional classification via stacked dependency exploitation. Sci China Inf Sci, 2020, 63(12): 222102, https://doi.org/10.1007/s11432-019-2905-3
1
Introduction
Multi-class classification is an important learning task in traditional supervised learning. Sometimes, however, we need to classify the same object from different dimensions. For example, when conducting demographic census, the Census Bureau needs to classify people from the occupation dimension (with possible classes teacher, lawyer, farmer, salesman, etc.), from the marital-status dimension (with possible classes unmarried, married, divorced, etc.), and from the education dimension (with possible classes bachelor, master, doctor, etc.). This particular problem can be naturally formalized under multidimensional classification framework [1–3]. Specifically, multi-dimensional classification deals with the problem where each training example is represented by a single instance while associated with multiple class variables. Here, each class variable corresponds to one specific class s
Data Loading...
