Online Algorithms for Multiclass Classification Using Partial Labels
Ontology highlight
ABSTRACT:
SUBMITTER: Lauw H
PROVIDER: S-EPMC7206293 | biostudies-literature | 2020 Apr
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Modern survey methods may be subject to non-observable bias, from various sources. Among online surveys, for example, selection bias is prevalent, due to the sampling mechanism commonly used, whereby participants self-select from a subgroup whose characteristics differ from those of the target population. Several techniques have been proposed to tackle this issue. One such is Propensity Score Adjustment (PSA), which is widely used and has been analysed in various studies. The usual method of estimating the propensity score is logistic regression, which requires a reference probability sample in addition to the online nonprobability sample. The predicted propensities can be used for reweighting using various estimators. However, in the online survey context, there are alternatives that might outperform logistic regression regarding propensity estimation. The aim of the present study is to determine the efficiency of some of these alternatives, involving Machine Learning (ML) classification algorithms. PSA is applied in two simulation scenarios, representing situations commonly found in online surveys, using logistic regression and ML models for propensity estimation. The results obtained show that ML algorithms remove selection bias more effectively than logistic regression when used for PSA, but that their efficacy depends largely on the selection mechanism employed and the dimensionality of the data.
| S-EPMC7176094 | biostudies-literature
Project description:Dependence on the internet and computer programs demonstrates the significance of computer programs in our day-to-day lives. Such demands motivate malware developers to create more malware, both in terms of quantity and variety. Researchers are constantly faced with hurdles while attempting to protect themselves from potential hazards and risks due to malware authors' usage of code obfuscation techniques. Metamorphic and polymorphic variations are easily able to elude the widely utilized signature-based detection procedures. Researchers are more interested in deep learning approaches than machine learning techniques to analyze the behavior of such a vast number of virus variants. Researchers have been drawn to the categorization of malware within itself in addition to the classification of malware against benign programs to examine the behavioral differences between them. In order to investigate the relationship between the application programming interface (API) calls throughout API sequences and classify them, this work uses the one-dimensional convolutional neural network (1D-CNN) model to solve a multiclass classification problem. On API sequences, feature vectors for distinctive APIs are created using the Word2Vec word embedding approach and the skip-gram model. The one-vs.-rest approach is used to train 1D-CNN models to categorize malware, and all of them are then combined with a suggested ModifiedSoftVoting algorithm to improve classification. On the open benchmark dataset Mal-API-2019, the suggested ensembled 1D-CNN architecture captures improved evaluation scores with an accuracy of 0.90, a weighted average F1-score of 0.90, and an AUC score of more than 0.96 for all classes of malware.
| S-EPMC10702937 | biostudies-literature
Project description:Motor imagery (MI) allows the design of self-paced brain-computer interfaces (BCIs), which can potentially afford an intuitive and continuous interaction. However, the implementation of non-invasive MI-based BCIs with more than three commands is still a difficult task. First, the number of MIs for decoding different actions is limited by the constraint of maintaining an adequate spacing among the corresponding sources, since the electroencephalography (EEG) activity from near regions may add up. Second, EEG generates a rather noisy image of brain activity, which results in a poor classification performance. Here, we propose a solution to address the limitation of identifiable motor activities by using combined MIs (i.e., MIs involving 2 or more body parts at the same time). And we propose two new multilabel uses of the Common Spatial Pattern (CSP) algorithm to optimize the signal-to-noise ratio, namely MC2CMI and MC2SMI approaches. We recorded EEG signals from seven healthy subjects during an 8-class EEG experiment including the rest condition and all possible combinations using the left hand, right hand, and feet. The proposed multilabel approaches convert the original 8-class problem into a set of three binary problems to facilitate the use of the CSP algorithm. In the case of the MC2CMI method, each binary problem groups together in one class all the MIs engaging one of the three selected body parts, while the rest of MIs that do not engage the same body part are grouped together in the second class. In this way, for each binary problem, the CSP algorithm produces features to determine if the specific body part is engaged in the task or not. Finally, three sets of features are merged together to predict the user intention by applying an 8-class linear discriminant analysis. The MC2SMI method is quite similar, the only difference is that any of the combined MIs is considered during the training phase, which drastically accelerates the calibration time. For all subjects, both the MC2CMI and the MC2SMI approaches reached a higher accuracy than the classic pair-wise (PW) and one-vs.-all (OVA) methods. Our results show that, when brain activity is properly modulated, multilabel approaches represent a very interesting solution to increase the number of commands, and thus to provide a better interaction.
| S-EPMC7710761 | biostudies-literature
Project description:In previous works, boosting aggregation of classifier outputs from discrete brain areas has been demonstrated to reduce dimensionality and improve the robustness and accuracy of functional magnetic resonance imaging (fMRI) classification. However, dimensionality reduction and classification of mixed activation patterns of multiple classes remain challenging. In the present study, the goals were (a) to reduce dimensionality by combining feature reduction at the voxel level and backward elimination of optimally aggregated classifiers at the region level, (b) to compare region selection for spatially aggregated classification using boosting and partial least squares regression methods and (c) to resolve mixed activation patterns using probabilistic prediction of individual tasks. Brain activation maps from interleaved visual, motor, auditory and cognitive tasks were segmented into 144 functional regions. Feature selection reduced the number of feature voxels by more than 50%, leaving 95 regions. The two aggregation approaches further reduced the number of regions to 30, resulting in more than 75% reduction of classification time and misclassification rates of less than 3%. Boosting and partial least squares (PLS) were compared to select the most discriminative and the most task correlated regions, respectively. Successful task prediction in mixed activation patterns was feasible within the first block of task activation in real-time fMRI experiments. This methodology is suitable for sparsifying activation patterns in real-time fMRI and for neurofeedback from distributed networks of brain activation.
| S-EPMC3505582 | biostudies-literature
Project description:Background Studies have reported the use of photoplethysmography signals to detect atrial fibrillation; however, the use of photoplethysmography signals in classifying multiclass arrhythmias has rarely been reported. Our study investigated the feasibility of using photoplethysmography signals and a deep convolutional neural network to classify multiclass arrhythmia types. Methods and Results ECG and photoplethysmography signals were collected simultaneously from a group of patients who underwent radiofrequency ablation for arrhythmias. A deep convolutional neural network was developed to classify multiple rhythms based on 10-second photoplethysmography waveforms. Classification performance was evaluated by calculating the area under the microaverage receiver operating characteristic curve, overall accuracy, sensitivity, specificity, and positive and negative predictive values against annotations on the rhythm of arrhythmias provided by 2 cardiologists consulting the ECG results. A total of 228 patients were included; 118 217 pairs of 10-second photoplethysmography and ECG waveforms were used. When validated against an independent test data set (23 384 photoplethysmography waveforms from 45 patients), the DCNN achieved an overall accuracy of 85.0% for 6 rhythm types (sinus rhythm, premature ventricular contraction, premature atrial contraction, ventricular tachycardia, supraventricular tachycardia, and atrial fibrillation); the microaverage area under the microaverage receiver operating characteristic curve was 0.978; the average sensitivity, specificity, and positive and negative predictive values were 75.8%, 96.9%, 75.2%, and 97.0%, respectively. Conclusions This study demonstrated the feasibility of classifying multiclass arrhythmias from photoplethysmography signals using deep learning techniques. The approach is attractive for population-based screening and may hold promise for the long-term surveillance and management of arrhythmia. Registration URL: www.chictr.org.cn. Identifier: ChiCTR2000031170.
| S-EPMC9075456 | biostudies-literature
Project description:Vigorous sperm motility, including the transition from progressive to hyperactivated motility that occurs in the female reproductive tract, is required for normal fertilization in mammals. We developed an automated, quantitative method that objectively classifies five distinct motility patterns of mouse sperm using Support Vector Machines (SVM), a common method in supervised machine learning. This multiclass SVM model is based on more than 2000 sperm tracks that were captured by computer-assisted sperm analysis (CASA) during in vitro capacitation and visually classified as progressive, intermediate, hyperactivated, slow, or weakly motile. Parameters associated with the classified tracks were incorporated into established SVM algorithms to generate a series of equations. These equations were integrated into a binary decision tree that sequentially sorts uncharacterized tracks into distinct categories. The first equation sorts CASA tracks into vigorous and nonvigorous categories. Additional equations classify vigorous tracks as progressive, intermediate, or hyperactivated and nonvigorous tracks as slow or weakly motile. Our CASAnova software uses these SVM equations to classify individual sperm motility patterns automatically. Comparisons of motility profiles from sperm incubated with and without bicarbonate confirmed the ability of the model to distinguish hyperactivated patterns of motility that develop during in vitro capacitation. The model accurately classifies motility profiles of sperm from a mutant mouse model with severe motility defects. Application of the model to sperm from multiple inbred strains reveals strain-dependent differences in sperm motility profiles. CASAnova provides a rapid and reproducible platform for quantitative comparisons of motility in large, heterogeneous populations of mouse sperm.
| S-EPMC3099585 | biostudies-literature
Project description:The rapid growth of technology has massively increased the amount of text data. The data can be mined and utilized for numerous natural language processing (NLP) tasks, particularly text classification. The core part of text classification is collecting the data for predicting a good model. This paper collects Kurdish News Dataset Headlines (KNDH) for text classification. The dataset consists of 50000 news headlines which are equally distributed among five classes, with 10000 headlines for each class (Social, Sport, Health, Economic, and Technology). The percentage ratio of getting the channels of headlines is distinct, while the numbers of samples are equal for each category. There are 34 distinct channels that are used to collect the different headlines for each class, such as 8 channels for economics, 14 channels for health, 18 channels for science, 15 channels for social, and 5 channels for sport. The dataset is preprocessed using the Kurdish Language Processing Toolkit (KLPT) for tokenizing, spell-checking, stemming, and preprocessing.
| S-EPMC10147969 | biostudies-literature
Project description:As the complexity and heterogeneity of cancer is being increasingly appreciated through genomic analyses, microarray-based cancer classification comprising multiple discriminatory molecular markers is an emerging trend. Such multiclass classification problems pose new methodological and computational challenges for developing novel and effective statistical approaches. In this paper, we introduce a new approach for classifying multiple disease states associated with cancer based on gene expression profiles. Our method focuses on detecting small sets of genes in which the relative comparison of their expression values leads to class discrimination. For an m-class problem, the classification rule typically depends on a small number of m-gene sets, which provide transparent decision boundaries and allow for potential biological interpretations. We first test our approach on seven common gene expression datasets and compare it with popular classification methods including support vector machines and random forests. We then consider an extremely large cohort of leukemia cancer patients to further assess its effectiveness. In both experiments, our method yields comparable or even better results to benchmark classifiers. In addition, we demonstrate that our approach can integrate pathway analysis of gene expression to provide accurate and biological meaningful classification.
| S-EPMC4775275 | biostudies-literature
Project description:Genome-wide association studies (GWAS) are widely used to search for genetic loci that underlie human disease. Another goal is to predict disease risk for different individuals given their genetic sequence. Such predictions could either be used as a "black box" in order to promote changes in life-style and screening for early diagnosis, or as a model that can be studied to better understand the mechanism of the disease. Current methods for risk prediction typically rank single nucleotide polymorphisms (SNPs) by the p-value of their association with the disease, and use the top-associated SNPs as input to a classification algorithm. However, the predictive power of such methods is relatively poor. To improve the predictive power, we devised BootRank, which uses bootstrapping in order to obtain a robust prioritization of SNPs for use in predictive models. We show that BootRank improves the ability to predict disease risk of unseen individuals in the Wellcome Trust Case Control Consortium (WTCCC) data and results in a more robust set of SNPs and a larger number of enriched pathways being associated with the different diseases. Finally, we show that combining BootRank with seven different classification algorithms improves performance compared to previous studies that used the WTCCC data. Notably, diseases for which BootRank results in the largest improvements were recently shown to have more heritability than previously thought, likely due to contributions from variants with low minimum allele frequency (MAF), suggesting that BootRank can be beneficial in cases where SNPs affecting the disease are poorly tagged or have low MAF. Overall, our results show that improving disease risk prediction from genotypic information may be a tangible goal, with potential implications for personalized disease screening and treatment.
| S-EPMC3749941 | biostudies-literature
Project description:Breast cancer is a common malignancy and a leading cause of cancer-related deaths in women worldwide. Its early diagnosis can significantly reduce the morbidity and mortality rates in women. To this end, histopathological diagnosis is usually followed as the gold standard approach. However, this process is tedious, labor-intensive, and may be subject to inter-reader variability. Accordingly, an automatic diagnostic system can assist to improve the quality of diagnosis. This paper presents a deep learning approach to automatically classify hematoxylin-eosin-stained breast cancer microscopy images into normal tissue, benign lesion, in situ carcinoma, and invasive carcinoma using our collected dataset. Our proposed model exploited six intermediate layers of the Xception (Extreme Inception) network to retrieve robust and abstract features from input images. First, we optimized the proposed model on the original (unnormalized) dataset using 5-fold cross-validation. Then, we investigated its performance on four normalized datasets resulting from Reinhard, Ruifrok, Macenko, and Vahadane stain normalization. For original images, our proposed framework yielded an accuracy of 98% along with a kappa score of 0.969. Also, it achieved an average AUC-ROC score of 0.998 as well as a mean AUC-PR value of 0.995. Specifically, for in situ carcinoma and invasive carcinoma, it offered sensitivity of 96% and 99%, respectively. For normalized images, the proposed architecture performed better for Makenko normalization compared to the other three techniques. In this case, the proposed model achieved an accuracy of 97.79% together with a kappa score of 0.965. Also, it attained an average AUC-ROC score of 0.997 and a mean AUC-PR value of 0.991. Especially, for in situ carcinoma and invasive carcinoma, it offered sensitivity of 96% and 99%, respectively. These results demonstrate that our proposed model outperformed the baseline AlexNet as well as state-of-the-art VGG16, VGG19, Inception-v3, and Xception models with their default settings. Furthermore, it can be inferred that although stain normalization techniques offered competitive performance, they could not surpass the results of the original dataset.
| S-EPMC9649689 | biostudies-literature