Using Regression Error Analysis and Feature Selection to Automatic Cluster Labeling

Silva, Lúcia Emília Soares; Machado, Vinícius Ponte; Araujo, Sidiney Souza; Lima, Bruno Vicente Alves de; Veras, Rodrigo

doi:10.1007/978-3-030-86230-5_30

Cited by 3 publications

(2 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Despite a slightly lower average hit rate, the proposed model does not have overlapping labels, considering the same attribute. It was observed that in the model proposed by [21], there is an overlapping range of values, which compromises the interpretation of the label since the same label referring to the UCS attribute belongs to more than one cluster, that is is, UCS(c 1 =[1∼5]) and UCS(c 2 =[1.9∼10]).…”

Section: B Wine Datasetmentioning

confidence: 99%

See 1 more Smart Citation

Method for Inferring the Optimal Number of Clusters with Subsequent Automatic Data Labeling based on Standard Deviation

Silva¹,

Sousa²,

Santos³

et al. 2023

IJACSA

View full text Add to dashboard Cite

Machine learning is a suitable pattern recognition technique for detecting correlations between data. In the case of unsupervised learning, the groups formed from these correlations can receive a label, which consists of describing them in terms of their most relevant attributes and their respective ranges of values so that they are understood automatically. In this research work, this process is called labeling. However, a challenge for researchers is establishing the optimal number of clusters that best represent the underlying structure of the data subjected to clustering. This optimal number may vary depending on the data set and the grouping method used and influences the data clustering process and, consequently, the interpretability of the generated groups. Therefore, this research aims to provide an inference approach to the number of clusters to be used in the grouping based on the range of attribute values, followed by automatic data labeling based on the standard deviation to maximize the understanding of the groups obtained. This methodology was applied to four databases. The results show that it contributes to the interpretation of the groups since it generates more accurate labels without any overlap between ranges of values, considering the same attribute in different groups.

show abstract

Section: B Wine Datasetmentioning

confidence: 99%

“…The Table XXIII shows the result of labeling the method of [21] for the Breast Cancer dataset. No criteria were used to infer the optimal number of groups in [21], so the author used K=2 for clustering.…”

Section: Seeds Datasetmentioning

confidence: 99%