LICIC: Less Important Components for Imbalanced Multiclass Classification

Dentamaro, Vincenzo; Impedovo, Donato; Pirlo, Giuseppe

doi:10.3390/info9120317

Cited by 11 publications

(6 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The classification is performed with 10-Fold cross validation with 70-30 ratio maintaining the inter-patient separation scheme: 28 subjects in training and 12 in test set randomly kept 10 different times. The produced train-test sets within 10-Fold cross validation may be imbalanced [54]. For this reason and only when there was relevant imbalance between healthy and sick, it was decided to perform, before feature selection, a novel oversampling technique called LICIC [54].…”

Section: Classifiers and Experimental Setupmentioning

confidence: 99%

“…The produced train-test sets within 10-Fold cross validation may be imbalanced [54]. For this reason and only when there was relevant imbalance between healthy and sick, it was decided to perform, before feature selection, a novel oversampling technique called LICIC [54]. This oversampling technique creates new instances balancing the minority classes by preserving nonlinearities and the particular pattern present in each specific class.…”

Section: Classifiers and Experimental Setupmentioning

confidence: 99%

See 1 more Smart Citation

Gait Analysis for Early Neurodegenerative Diseases Classification Through the Kinematic Theory of Rapid Human Movements

2020

Self Cite

View full text Add to dashboard Cite

Neurodegenerative diseases are particular diseases whose decline can partially or completely compromise the normal course of life of a human being. In order to increase the quality of patient's life, a timely diagnosis plays a major role. The analysis of neurodegenerative diseases, and their stage, is also carried out by means of gait analysis. Performing early stage neurodegenerative disease assessment is still an open problem. In this paper, the focus is on modeling the human gait movement pattern by using the kinematic theory of rapid human movements and its sigma-lognormal model. The hypothesis is that the kinematic theory of rapid human movements, originally developed to describe handwriting patterns, and used in conjunction with other spatio-temporal features, can discriminate neurodegenerative diseases patterns, especially in early stages, while analyzing human gait with 2D cameras. The thesis empirically demonstrates its effectiveness in describing neurodegenerative patterns, when used in conjunction with state-of-the-art pose estimation and feature extraction techniques. The solution developed achieved 99.1% of accuracy using velocity-based, angle-based and sigma-lognormal features and left walk orientation. INDEX TERMS pose-estimation, computer vision, computer aided diagnosis, gait analysis, machine learning, early neurodegenerative diseases assessment, kinematic theory of rapid human movements, sigma-lognormal

show abstract

Section: Classifiers and Experimental Setupmentioning

confidence: 99%

Section: Classifiers and Experimental Setupmentioning

confidence: 99%

Gait Analysis for Early Neurodegenerative Diseases Classification Through the Kinematic Theory of Rapid Human Movements

2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…Dentamaro et al [18] propose a new oversampling technique called Less Important Components for Imbalanced Multiclass Classification-LICIC to cope with both class imbalance and the famous "curse of dimensionality" problem. The method enables preservation of non-linearities within the dataset, while creating new instances without adding noise.…”

Section: Contributionmentioning

confidence: 99%

eHealth and Artificial Intelligence

Impedovo

Pirlo

2019

Information

Self Cite

View full text Add to dashboard Cite

Artificial intelligence is changing the healthcare industry from many perspectives: diagnosis, treatment, and follow-up. A wide range of techniques has been proposed in the literature. In this special issue, 13 selected and peer-reviewed original research articles contribute to the application of artificial intelligence (AI) approaches in various real-world problems. Papers refer to the following main areas of interest: feature selection, high dimensionality, and statistical approaches; heart and cardiovascular diseases; expert systems and e-health platforms.

show abstract

“…24,26 Finally, support vector machines, random forest classifiers, and novel oversampling techniques have been implemented to improve the characterization of spectra from highly similar bacteria. 27 We sought to determine whether the Aristotle Classifier, which was developed for glycomics studies, could provide additional benefits to the field of bacterial identification. To test this hypothesis, we used nine different data sets containing MALDI-TOF MS data of bacterial proteins, where the all the members of a given data set originate from the same genus.…”

Section: Introductionmentioning

confidence: 99%

“…In many cases, machine learning methods have provided researchers better ability to discriminate bacterial samples than what would be gained by simply comparing user-generated spectra to a set of prototypes. − In one case, the use of the XGBoost classifier was pivotal in enabling the identification of polymicrobial species based on MS data of their membrane glycolipids . In other examples, researchers have used a variety of machine learning strategies to discriminate samples of bacterial mixtures. , Finally, support vector machines, random forest classifiers, and novel oversampling techniques have been implemented to improve the characterization of spectra from highly similar bacteria …”

mentioning

confidence: 99%

Adaption of the Aristotle Classifier for Accurately Identifying Highly Similar Bacteria Analyzed by MALDI-TOF MS

Desaire

Hua

2019

Anal. Chem.

View full text Add to dashboard Cite

MALDI-TOF MS has shown great utility for rapidly identifying microbial species. It can be used to successfully type bacteria and fungi from a variety of sources more rapidly and cost-effectively than traditional methods. One area where improvements are necessary is in the typing of highly similar samples, such as those samples from the same genus but different species or samples from within a single species but from different strains. One promising way to address this current limitation is by using advanced machine learning techniques. In this work, we adapt a newly developed machine learning tool, the Aristotle Classifier, to bacterial classification of MALDI-TOF MS data. This tool was originally developed for classifying glycomics and glycoproteomics data, so we modified it to be well-suited for assigning mass spectral data from bacterial proteins. The classifier exceeds existing benchmarks in classifying bacteria, and it shows particularly strong performance when the samples to be identified are highly similar. The combination of mass spectrometry data and tools like the Aristotle Classifier could ameliorate the ambiguities associated with challenging bacterial classification problems.

show abstract

LICIC: Less Important Components for Imbalanced Multiclass Classification

Cited by 11 publications

References 18 publications

Gait Analysis for Early Neurodegenerative Diseases Classification Through the Kinematic Theory of Rapid Human Movements

Gait Analysis for Early Neurodegenerative Diseases Classification Through the Kinematic Theory of Rapid Human Movements

eHealth and Artificial Intelligence

Adaption of the Aristotle Classifier for Accurately Identifying Highly Similar Bacteria Analyzed by MALDI-TOF MS

Contact Info

Product

Resources

About