Cost-sensitive dual-bidirectional linear discriminant analysis

Li, Huaxiong; Zhang, Libo; Huang, Bing; Zhou, Xianzhong

doi:10.1016/j.ins.2019.09.032

Cited by 55 publications

(14 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Choose K distinct objects cen 1 , cen 2 ,...,cen K from the universe U as the initial set of modes (t = 1) CEN (t=1) = {cen 1 ,cen 2 ,...,cen K } ∈ U k . Determine W (1) such that F (W,cen (1) ) is minimized. t) ,CEN (t) ), then stop.…”

Section: The Clustering Model Of the Drk-mmentioning

confidence: 99%

“…Theorem 2 let cen lj = [cen l1 , cen l2 ,...,cen ld ] be the mode of the lth (1 ≤ l ≤ K) cluster and V a j = a (1) j , a (2) j , … , a (n j ) j the domain of attributes a j where |a j |= n j be where…”

Section: The Clustering Model Of the Drk-mmentioning

confidence: 99%

“…Cluster analysis, also known as unsupervised learning, is a branch of Machine Learning which has multiple applications in various domains, including financial fraud detection, medical diagnosis, image processing, information retrieval and bioinformatics [ 1 ]. Clustering is used to unlock initially hidden and undetectable patterns by identifying groupings within the dataset under investigation in an unsupervised manner: no labels (or classes) are initially provided as input parameters.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A rough set based algorithm for updating the modes in categorical clustering

Salem

Naouali²,

Chtourou

2021

Int. J. Mach. Learn. & Cyber.

View full text Add to dashboard Cite

The categorical clustering problem has attracted much attention especially in the last decades since many real world applications produce categorical data. The k-mode algorithm, proposed since 1998, and its multiple variants were widely used in this context. However, they suffer from a great limitation related to the update of the modes in each iteration. The mode in the last step of these algorithms is randomly selected although it is possible to identify many candidate ones. In this paper, a rough density mode selection method is proposed to identify the adequate modes among a list of candidate ones in each iteration of the k-modes. The proposed method, called Density Rough k-Modes (DRk-M) was experimented using real world datasets extracted from the UCI Machine Learning Repository, the Global Terrorism Database (GTD) and a set of collected Tweets. The DRk-M was also compared to many states of the art clustering methods and has shown great efficiency.

show abstract

Section: The Clustering Model Of the Drk-mmentioning

confidence: 99%

Section: The Clustering Model Of the Drk-mmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A rough set based algorithm for updating the modes in categorical clustering

Salem

Naouali²,

Chtourou

2021

Int. J. Mach. Learn. & Cyber.

View full text Add to dashboard Cite

show abstract

“…Linear discriminant analysis (LDA) is a classical method for discriminant analysis. It has been widely used in many areas, such as pattern recognition [13,14] and machine learning [10,27]. LDA seeks to find a linear combination of features that separates two or more classes of objects.…”

Section: Linear Discriminant Analysismentioning

confidence: 99%

Within-class multimodal classification

Wan

Wang

Scotney

et al. 2020

Multimed Tools Appl

View full text Add to dashboard Cite

In many real-world classification problems there exist multiple subclasses (or clusters) within a class; in other words, the underlying data distribution is within-class multimodal. One example is face recognition where a face (i.e. a class) may be presented in frontal view or side view, corresponding to different modalities. This issue has been largely ignored in the literature or at least under studied. How to address the within-class multimodality issue is still an unsolved problem. In this paper, we present an extensive study of within-class multimodality classification. This study is guided by a number of research questions, and conducted through experimentation on artificial data and real data. In addition, we establish a case for within-class multimodal classification that is characterised by the concurrent maximisation of between-class separation, between-subclass separation and within-class compactness. Extensive experimental results show that within-class multimodal classification consistently leads to significant performance gains when within-class multimodality is present in data. Furthermore, it has been found that within-class multimodal classification offers a competitive solution to face recognition under different lighting and face pose conditions. It is our opinion that the case for within-class multimodal classification is established, therefore there is a milestone to be achieved in some machine learning algorithms (e.g. Gaussian mixture model) when within-class multimodal classification, or part of it, is pursued.

show abstract

“…Decisions of composition extractions from the coffee bean dataset, topology of composition distributions on a 2D plane with four quadrants, selection of compositions for the ordinate and abscissa coordinates for the four quadrants, and appropriate statistical analysis for computation affect the conclusions from the experimental dataset. The approach usually involves data normalization, data matrix replacement, statistical and regression analysis (Li, Zhang, Huang, & Zhou, 2020) (Kalsoom, Maqsood, Ghazanfar, Aadil, & Rho, 2018). However, the computation process unintentionally distorts specific important data characteristics and causes information loss (Chen, Wu, & Liang, 2018) (Tharwat, Gaber, Ibrahim, & Hassanien, 2017).…”

mentioning

confidence: 99%

Classification of coffee bean categories based upon analysis of fatty acid ingredients

Hung

Lee

Lin

2021

J. Food Process. Preserv.

View full text Add to dashboard Cite

The research studies constituents of fatty acids (FA) in coffee beans to identify their categories. Since fatty acids are the fundamental constituents of coffee aroma and flavor, challenges occur to classify the beans' original species in the roasted state. The examined samples in this study cover 74 coffee beans from different origins and are separated into Arabica and Robusta species based on their fatty acid composition. This research develops a discriminant strategy to identify categories of examined coffee beans. This study analyzes an experimental dataset using multiple data structure strategies during the identification process, which are different from traditional approaches that aim to improve coffee bean species classification and recognition rate. Furthermore, the developed coffee bean identification strategies implement various normalization and error analyses during the data reasoning process. This research concludes that fatty acids C18:1, C18:2, and C18:3 own essential characteristics for the coffee beans. Practical applicationsThis research develops an innovative strategy to identify coffee beans in well-known categories and implements various normalization and error analyses during the data processing process. The developed model solves the information loss problem due to data normalization and improves the accuracy of coffee been classification.Consequently, the study concludes that fatty acids C18:1, C18:2, and C18:3 characterize essential features for coffee beans, and the research results balance fundamental chemistry and engineering principles and serve the general food processing and preservation technology industries.

show abstract

Cost-sensitive dual-bidirectional linear discriminant analysis

Cited by 55 publications

References 42 publications

A rough set based algorithm for updating the modes in categorical clustering

A rough set based algorithm for updating the modes in categorical clustering

Within-class multimodal classification

Classification of coffee bean categories based upon analysis of fatty acid ingredients

Contact Info

Product

Resources

About