Multi-granularity distance metric learning via neighborhood granule margin maximization

Zhu, Pengfei; Hu, Qinghua; Zuo, Wangmeng; Yang, Meng

doi:10.1016/j.ins.2014.06.017

Cited by 34 publications

(9 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, the multigranularity structure will be constructed through employing different radii. As reported in [16,27], the radius-based neighborhood forms a information granule, then the neighborhood-based singlegranularity can be constructed by applying one and only one radius, and it follows that neighborhood based multigranularity can be constructed by applying a set of different radii. In this process, applying a smaller radius will generate a finer information granule while applying a greater radius will generate a coarser information granule.…”

Section: Introductionmentioning

confidence: 98%

“…However, it is worth noting that they all focus on one and only one xed parameter in attribute reduction [9][10][11], such as the xed Gaussian kernel parameter in fuzzy rough set or the xed radius in neighborhood rough set. From the viewpoint of Granular Computing (GrC) [12][13][14][15][16], applying one and only one parameter in rough set can only re ect the information over a fixed granularity [17][18][19]. erefore, the attribute reduction over a fixed granularity can be termed as the single-granularity attribute reduction.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Ensemble‐Based Neighborhood Attribute Reduction: A Multigranularity View

et al. 2019

View full text Add to dashboard Cite

Recently, multigranularity has been an interesting topic, since different levels of granularity can provide different information from the viewpoint of Granular Computing (GrC). However, established researches have focused less on investigating attribute reduction from multigranularity view. This paper proposes an algorithm based on the multigranularity view. To construct a framework of multigranularity attribute reduction, two main problems can be addressed as follows: (1) The multigranularity structure can be constructed firstly. In this paper, the multigranularity structure will be constructed based on the radii, as different information granularities can be induced by employing different radii. Therefore, the neighborhood-based multigranularity can be constructed. (2) The attribute reduction can be designed and realized from the viewpoint of multigranularity. Different from traditional process which computes reduct through employing a fixed granularity, our algorithm aims to obtain reduct from the viewpoint of multigranularity. To realize the new algorithm, two main processes are executed as follows: (1) Considering that different decision classes may require different key condition attributes, the ensemble selector is applied among different decision classes; (2) to accelerate the process of attribute reduction, only the finest and the coarsest granularities are employed. The experiments over 15 UCI data sets are conducted. Compared with the traditional single-granularity approach, the multigranularity algorithm can not only generate reduct which can provide better classification accuracy, but also reduce the elapsed time. This study suggests new trends for considering both the classification accuracy and the time efficiency with respect to the reduct.

show abstract

Section: Introductionmentioning

confidence: 98%

Section: Introductionmentioning

confidence: 99%

Ensemble‐Based Neighborhood Attribute Reduction: A Multigranularity View

et al. 2019

View full text Add to dashboard Cite

show abstract

“…According to such property, many knowledge discovery and data mining approaches involving parameter based granularity have been explored. For example, Liu et al [21] designed a framework of multi-granularity feature selection scheme which considers the variation of parameters; Zhu et al [40] learnt an effective distance metric based on neighborhood granule margin maximization from a group of parameters; Zhu et al [41] also presented an adaptive selection scheme of neighborhood granularity (i.e., parameters), and provided a solution to margin distribution optimization. Nevertheless, these approaches only exploited the external representations of parameter, instead of describing the reflected granularity, as well as the associated discrimination ability of knowledge.…”

Section: Introductionmentioning

confidence: 99%

Knowledge Granularity for Continuous Parameters

She

et al. 2021

IEEE Access

View full text Add to dashboard Cite

In the community of Granular Computing, knowledge is interpreted as one classification ability of realistic or abstract objects. Generally, the concept of granularity is used for characterizing such an ability, which has been widely explored in literatures. To calculate the parameterized granularity, a naive approach is to find the granularity in terms of the parameter one by one. Nevertheless, such approach can only generate the single parameter based knowledge granularity, and the difference of knowledge granularities among different parameters may be slight. It follows that the knowledge granularity derived from single parameter may be lack of representativeness. In this paper, the continuous parameters based knowledge granularity is proposed, and the corresponding calculation approach is presented. Inspired by the thinking of definite integral in mathematical problems, the calculation approach is mainly implemented by following steps: firstly, the graph formed by granularity and parameter interval is divided into several small rectangles whose length of interval tends to be 0; secondly, the sum of area values of all the small rectangles is calculated; finally, the obtained area value divided by the whole length of parameter interval can be considered as the continuous parameters based knowledge granularity. This study suggests a new trend of handling problems related to knowledge from the viewpoint of continuity.

show abstract

“…However, different data usually prefer different distance metrics to reflect different semantic concepts of dissimilarity or similarity in the context of problems, and hence adapting the distance metrics to different data can be expected to improve the classification performance of NSM. On the other hand, distance metric learning methods emerging in the machine learning community provide us a tool to learn tailored distance metrics automatically from data and to improve the classification performance [23,21,26,19,24].…”

Section: Introductionmentioning

confidence: 99%

Learning distance to subspace for the nearest subspace methods in high-dimensional data classification

Zhu

Dong

Xue

2019

Information Sciences

View full text Add to dashboard Cite

Nearest subspace methods (NSM) are a category of classification methods widely applied to classify high-dimensional data. In this paper, we propose to improve the classification performance of NSM through learning tailored distance metrics from samples to class subspaces. The learned distance metric is termed as 'learned distance to subspace' (LD2S). Using LD2S in the classification rule of NSM can make the samples closer to their correct class subspaces while farther away from their wrong class subspaces. In this way, the classification task becomes easier and the classification performance of NSM can be improved. The superior classification performance of using LD2S for NSM is demonstrated on three real-world high-dimensional spectral datasets.

show abstract

Multi-granularity distance metric learning via neighborhood granule margin maximization

Cited by 34 publications

References 44 publications

Ensemble‐Based Neighborhood Attribute Reduction: A Multigranularity View

Ensemble‐Based Neighborhood Attribute Reduction: A Multigranularity View

Knowledge Granularity for Continuous Parameters

Learning distance to subspace for the nearest subspace methods in high-dimensional data classification

Contact Info

Product

Resources

About