scholarly journals Multi-label learning with missing and completely unobserved labels

2021 ◽  
Author(s):  
Jun Huang ◽  
Linchuan Xu ◽  
Kun Qian ◽  
Jing Wang ◽  
Kenji Yamanishi
Keyword(s):  
Training Data ◽  
Data Set ◽  
New Approach ◽  
Useful Knowledge ◽  
Unseen Data ◽  
Benchmark Datasets ◽  
Label Correlations ◽  
Missing Labels ◽  
Class Labels ◽  
Multiple Class

AbstractMulti-label learning deals with data examples which are associated with multiple class labels simultaneously. Despite the success of existing approaches to multi-label learning, there is still a problem neglected by researchers, i.e., not only are some of the values of observed labels missing, but also some of the labels are completely unobserved for the training data. We refer to the problem as multi-label learning with missing and completely unobserved labels, and argue that it is necessary to discover these completely unobserved labels in order to mine useful knowledge and make a deeper understanding of what is behind the data. In this paper, we propose a new approach named MCUL to solve multi-label learning with Missing and Completely Unobserved Labels. We try to discover the unobserved labels of a multi-label data set with a clustering based regularization term and describe the semantic meanings of them based on the label-specific features learned by MCUL, and overcome the problem of missing labels by exploiting label correlations. The proposed method MCUL can predict both the observed and newly discovered labels simultaneously for unseen data examples. Experimental results validated over ten benchmark datasets demonstrate that the proposed method can outperform other state-of-the-art approaches on observed labels and obtain an acceptable performance on the new discovered labels as well.

BiLabel-Specific Features for Multi-Label Classification

2021 ◽  
Vol 16 (1) ◽  
pp. 1-23
Author(s):  
Min-Ling Zhang ◽  
Jun-Peng Fang ◽  
Yi-Bo Wang
Keyword(s):  
Predictive Models ◽  
Comparative Studies ◽  
State Of The Art ◽  
Classification Model ◽  
Generation Process ◽  
Prototype Selection ◽  
Class Label ◽  
Benchmark Datasets ◽  
Label Correlations ◽  
Class Labels

In multi-label classification, the task is to induce predictive models which can assign a set of relevant labels for the unseen instance. The strategy of label-specific features has been widely employed in learning from multi-label examples, where the classification model for predicting the relevancy of each class label is induced based on its tailored features rather than the original features. Existing approaches work by generating a group of tailored features for each class label independently, where label correlations are not fully considered in the label-specific features generation process. In this article, we extend existing strategy by proposing a simple yet effective approach based on BiLabel-specific features. Specifically, a group of tailored features is generated for a pair of class labels with heuristic prototype selection and embedding. Thereafter, predictions of classifiers induced by BiLabel-specific features are ensembled to determine the relevancy of each class label for unseen instance. To thoroughly evaluate the BiLabel-specific features strategy, extensive experiments are conducted over a total of 35 benchmark datasets. Comparative studies against state-of-the-art label-specific features techniques clearly validate the superiority of utilizing BiLabel-specific features to yield stronger generalization performance for multi-label classification.

scholarly journals Multiscale Mask R-CNN–Based Lung Tumor Detection Using PET Imaging

2019 ◽  
Vol 18 ◽  
pp. 153601211986353 ◽  
Author(s):  
Rui Zhang ◽  
Chao Cheng ◽  
Xuehua Zhao ◽  
Xuechen Li
Keyword(s):  
Pet Imaging ◽  
Lung Tumor ◽  
Training Data ◽  
Fine Tuning ◽  
Weighted Voting ◽  
Data Set ◽  
New Approach ◽  
Positron Emission ◽  
Voting Strategy ◽  
Aided Diagnosis

Positron emission tomography (PET) imaging serves as one of the most competent methods for the diagnosis of various malignancies, such as lung tumor. However, with an elevation in the utilization of PET scan, radiologists are overburdened considerably. Consequently, a new approach of “computer-aided diagnosis” is being contemplated to curtail the heavy workloads. In this article, we propose a multiscale Mask Region–Based Convolutional Neural Network (Mask R-CNN)–based method that uses PET imaging for the detection of lung tumor. First, we produced 3 models of Mask R-CNN for lung tumor candidate detection. These 3 models were generated by fine-tuning the Mask R-CNN using certain training data that consisted of images from 3 different scales. Each of the training data set included 594 slices with lung tumor. These 3 models of Mask R-CNN models were then integrated using weighted voting strategy to diminish the false-positive outcomes. A total of 134 PET slices were employed as test set in this experiment. The precision, recall, and F score values of our proposed method were 0.90, 1, and 0.95, respectively. Experimental results exhibited strong conviction about the effectiveness of this method in detecting lung tumors, along with the capability of identifying a healthy chest pattern and reducing incorrect identification of tumors to a large extent.

scholarly journals ITERATIVE RE-WEIGHTED INSTANCE TRANSFER FOR DOMAIN ADAPTATION

2016 ◽  
Vol III-3 ◽  
pp. 339-346 ◽  
Author(s):  
A. Paul ◽  
F. Rottensteiner ◽  
C. Heipke
Keyword(s):  
Domain Adaptation ◽  
Training Data ◽  
Aerial Images ◽  
Target Domain ◽  
Data Set ◽  
Source Domain ◽  
The Difference ◽  
A New Technique ◽  
Class Labels ◽  
Decision Boundaries

Domain adaptation techniques in transfer learning try to reduce the amount of training data required for classification by adapting a classifier trained on samples from a source domain to a new data set (target domain) where the features may have different distributions. In this paper, we propose a new technique for domain adaptation based on logistic regression. Starting with a classifier trained on training data from the source domain, we iteratively include target domain samples for which class labels have been obtained from the current state of the classifier, while at the same time removing source domain samples. In each iteration the classifier is re-trained, so that the decision boundaries are slowly transferred to the distribution of the target features. To make the transfer procedure more robust we introduce weights as a function of distance from the decision boundary and a new way of regularisation. Our methodology is evaluated using a benchmark data set consisting of aerial images and digital surface models. The experimental results show that in the majority of cases our domain adaptation approach can lead to an improvement of the classification accuracy without additional training data, but also indicate remaining problems if the difference in the feature distributions becomes too large.

Classifying Lithofacies from Textural Features in Whole Core CT-Scan Images

2021 ◽  
pp. 1-17
Author(s):  
Kurdistan Chawshin ◽  
Carl F. Berg ◽  
Damiano Varagnolo ◽  
Andres Gonzalez ◽  
Zoya Heidari ◽  
...  
Keyword(s):  
Principal Component ◽  
Ct Images ◽  
Nondestructive Method ◽  
Training Data ◽  
Support Vector ◽  
Svm Classifier ◽  
Statistical Features ◽  
Data Set ◽  
Unseen Data ◽  
Core Description

Summary X-ray computerized tomography (CT) is a nondestructive method of providing information about the internal composition and structure of whole core reservoir samples. In this study we propose a method to classify lithology. The novelty of this method is that it uses statistical and textural information extracted from whole core CT images in a supervised learning environment. In the proposed approaches, first-order statistical features and textural grey-levelco-occurrence matrix (GLCM) features are extracted from whole core CT images. Here, two workflows are considered. In the first workflow, the extracted features are used to train a support vector machine (SVM) to classify lithofacies. In the second workflow, a principal component analysis (PCA) step is added before training with two purposes: first, to eliminate collinearity among the features and second, to investigate the amount of information needed to differentiate the analyzed images. Before extracting the statistical features, the images are preprocessed and decomposed using Haar mother wavelet decomposition schemes to enhance the texture and to acquire a set of detail images that are then used to compute the statistical features. The training data set includes lithological information obtained from core description. The approach is validated using the trained SVM and hybrid (PCA + SVM) classifiers to predict lithofacies in a set of unseen data. The obtained results show that the SVM classifier can predict some of the lithofacies with high accuracy (up to 91% recall), but it misclassifies, to some extent, similar lithofacies with similar grain size, texture, and transport properties. The SVM classifier captures the heterogeneity in the whole core CT images more accurately compared with the core description, indicating that the CT images provide additional high-resolution information not observed by manual core description. Further, the obtained prediction results add information on the similarity of the lithofacies classes. The prediction results using the hybrid classifier are worse than the SVM classifier, indicating that low-power components may contain information that is required to differentiate among various lithofacies.

scholarly journals A Comprehensive Machine Learning Framework for the Exact Prediction of the Age of Onset in Familial and Sporadic Alzheimer’s Disease

Diagnostics ◽  
2021 ◽  
Vol 11 (5) ◽  
pp. 887
Author(s):  
Jorge I. Vélez ◽  
Luiggi A. Samper ◽  
Mauricio Arcos-Holzinger ◽  
Lady G. Espinosa ◽  
Mario A. Isaza-Ruget ◽  
...  
Keyword(s):  
Machine Learning ◽  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
Mean Squared Error ◽  
Age Of Onset ◽  
Late Onset ◽  
Training Data ◽  
Therapeutic Interventions ◽  
Data Set ◽  
Unseen Data

Machine learning (ML) algorithms are widely used to develop predictive frameworks. Accurate prediction of Alzheimer’s disease (AD) age of onset (ADAOO) is crucial to investigate potential treatments, follow-up, and therapeutic interventions. Although genetic and non-genetic factors affecting ADAOO were elucidated by other research groups and ours, the comprehensive and sequential application of ML to provide an exact estimation of the actual ADAOO, instead of a high-confidence-interval ADAOO that may fall, remains to be explored. Here, we assessed the performance of ML algorithms for predicting ADAOO using two AD cohorts with early-onset familial AD and with late-onset sporadic AD, combining genetic and demographic variables. Performance of ML algorithms was assessed using the root mean squared error (RMSE), the R-squared (R2), and the mean absolute error (MAE) with a 10-fold cross-validation procedure. For predicting ADAOO in familial AD, boosting-based ML algorithms performed the best. In the sporadic cohort, boosting-based ML algorithms performed best in the training data set, while regularization methods best performed for unseen data. ML algorithms represent a feasible alternative to accurately predict ADAOO with little human intervention. Future studies may include predicting the speed of cognitive decline in our cohorts using ML.

scholarly journals ITERATIVE RE-WEIGHTED INSTANCE TRANSFER FOR DOMAIN ADAPTATION

2016 ◽  
Vol III-3 ◽  
pp. 339-346 ◽  
Author(s):  
A. Paul ◽  
F. Rottensteiner ◽  
C. Heipke
Keyword(s):  
Domain Adaptation ◽  
Training Data ◽  
Aerial Images ◽  
Target Domain ◽  
Data Set ◽  
Source Domain ◽  
The Difference ◽  
A New Technique ◽  
Class Labels ◽  
Decision Boundaries

Domain adaptation techniques in transfer learning try to reduce the amount of training data required for classification by adapting a classifier trained on samples from a source domain to a new data set (target domain) where the features may have different distributions. In this paper, we propose a new technique for domain adaptation based on logistic regression. Starting with a classifier trained on training data from the source domain, we iteratively include target domain samples for which class labels have been obtained from the current state of the classifier, while at the same time removing source domain samples. In each iteration the classifier is re-trained, so that the decision boundaries are slowly transferred to the distribution of the target features. To make the transfer procedure more robust we introduce weights as a function of distance from the decision boundary and a new way of regularisation. Our methodology is evaluated using a benchmark data set consisting of aerial images and digital surface models. The experimental results show that in the majority of cases our domain adaptation approach can lead to an improvement of the classification accuracy without additional training data, but also indicate remaining problems if the difference in the feature distributions becomes too large.

Classification Methods

2011 ◽  
pp. 144-149 ◽  
Author(s):  
Aijun An
Keyword(s):  
Unsupervised Learning ◽  
Supervised Learning ◽  
A Priori ◽  
Training Data ◽  
Classification Model ◽  
Data Set ◽  
Data Object ◽  
Unseen Data ◽  
Data Objects ◽  
Class Variable

Generally speaking, classification is the action of assigning an object to a category according to the characteristics of the object. In data mining, classification refers to the task of analyzing a set of pre-classified data objects to learn a model (or a function) that can be used to classify an unseen data object into one of several predefined classes. A data object, referred to as an example, is described by a set of attributes or variables. One of the attributes describes the class that an example belongs to and is thus called the class attribute or class variable. Other attributes are often called independent or predictor attributes (or variables). The set of examples used to learn the classification model is called the training data set. Tasks related to classification include regression, which builds a model from training data to predict numerical values, and clustering, which groups examples to form categories. Classification belongs to the category of supervised learning, distinguished from unsupervised learning. In supervised learning, the training data consists of pairs of input data (typically vectors), and desired outputs, while in unsupervised learning there is no a priori output.

IDENTIFYING OPTIMAL TRAINING DATA SET - A NEW APPROACH

2004 ◽  
pp. 1621-1628
Author(s):  
C. SIVAPRAGASAM ◽  
SHIE-YUI LIONG
Keyword(s):  
Training Data ◽  
Data Set ◽  
New Approach

scholarly journals Improving Classification Performance through an Advanced Ensemble Based Heterogeneous Extreme Learning Machines

2017 ◽  
Vol 2017 ◽  
pp. 1-11 ◽  
Author(s):  
Adnan O. M. Abuassba ◽  
Dezheng Zhang ◽  
Xiong Luo ◽  
Ahmad Shaheryar ◽  
Hazrat Ali
Keyword(s):  
Learning Algorithm ◽  
Majority Vote ◽  
Classification Performance ◽  
Training Data ◽  
Generalization Performance ◽  
Fast Learning ◽  
Unseen Data ◽  
Benchmark Datasets ◽  
Learning Machine ◽  
Hidden Layer

Extreme Learning Machine (ELM) is a fast-learning algorithm for a single-hidden layer feedforward neural network (SLFN). It often has good generalization performance. However, there are chances that it might overfit the training data due to having more hidden nodes than needed. To address the generalization performance, we use a heterogeneous ensemble approach. We propose an Advanced ELM Ensemble (AELME) for classification, which includes Regularized-ELM, L2-norm-optimized ELM (ELML2), and Kernel-ELM. The ensemble is constructed by training a randomly chosen ELM classifier on a subset of training data selected through random resampling. The proposed AELM-Ensemble is evolved by employing an objective function of increasing diversity and accuracy among the final ensemble. Finally, the class label of unseen data is predicted using majority vote approach. Splitting the training data into subsets and incorporation of heterogeneous ELM classifiers result in higher prediction accuracy, better generalization, and a lower number of base classifiers, as compared to other models (Adaboost, Bagging, Dynamic ELM ensemble, data splitting ELM ensemble, and ELM ensemble). The validity of AELME is confirmed through classification on several real-world benchmark datasets.

Sign in / Sign up

Export Citation Format

Share Document