AR-Tri-Training: Tri-Training with Assistant Strategy

2014 ◽  
Vol 513-517 ◽  
pp. 1840-1844 ◽  
Author(s):  
Long Jie Cui ◽  
Hong Li Wang ◽  
Rong Yi Cui
Keyword(s):  
Learning Strategy ◽  
Voice Recognition ◽  
Classification Performance ◽  
Experimental Results ◽  
Training Algorithm ◽  
Information Strategy ◽  
Testing Rate ◽  
Rich Information ◽  
Validation Set

The classification performance of the classifier is weakened because the noise samples are introduced for the use of unlabeled samples in Tri-training. In this paper a new Tri-training style algorithm named AR-Tri-training (Tri-training with assistant and rich strategy) is proposed. Firstly, the assistant learning strategy is posed. Then the supporting learner is designed by combining the assistant learning strategy with rich information strategy. The number of mislabeled samples produced in the iterations of three classifiers mutually labeling are reduced by use of the supporting learner, moreover the unlabeled samples and the misclassified samples of validation set can be fully used. The proposed algorithm is applied to voice recognition. The experimental results show that AR-Tri-training algorithm can compensate for the shortcomings of Tri-training algorithm, further improve the testing rate.

Abnormal Voice Detection Algorithm Based on Semi-Supervised Co-Training Algorithm

2012 ◽  
Vol 461 ◽  
pp. 117-122 ◽  
Author(s):  
Ya Hui Zhao ◽  
Hong Li Wang ◽  
Rong Yi Cui
Keyword(s):  
Learning Strategy ◽  
Detection Algorithm ◽  
Training Algorithm ◽  
Mixed Programming ◽  
Training Samples ◽  
Rich Information ◽  
Detection Software ◽  
Voice Detection ◽  
The Rich ◽  
Validation Set

The AR-Tri-training algorithm is proposed for applying to the abnormal voice detection, and voice detection software is designed by mixed programming used Matlab and VC in this paper. Firstly, training samples are collected and the features of each sample are extracted including centroid, spectral entropy, wavelet and MFCC. Secondly, the assistant learning strategy is proposed, AR-Tri-training algorithm is designed by combining the rich information strategy. Finally, Classifiers are trained by using AR-Tri-training algorithm, and the integrated classifier is applied to voice detection. As can be drawn from the experimental results, AR-Tri-training not only removes mislabeled examples in training process, but also takes full advantage of the unlabeled examples and wrong-learning examples on validation set

scholarly journals A Heterogeneous Graph Enhanced LSTM Network for Hog Price Prediction Using Online Discussion

Agriculture ◽  
2021 ◽  
Vol 11 (4) ◽  
pp. 359
Author(s):  
Kai Ye ◽  
Yangheran Piao ◽  
Kun Zhao ◽  
Xiaohui Cui
Keyword(s):  
Decision Making ◽  
Essential Role ◽  
Online Discussion ◽  
Professional Community ◽  
Experimental Results ◽  
Agricultural Products ◽  
Price Prediction ◽  
Price Data ◽  
Rich Information ◽  
Lstm Network

Forecasting the prices of hogs has always been a popular field of research. Such information has played an essential role in decision-making for farmers, consumers, corporations, and governments. It is hard to predict hog prices because too many factors can influence them. Some of the factors are easy to quantify, but some are not. Capturing the characteristics behind the price data is also tricky considering their non-linear and non-stationary nature. To address these difficulties, we propose Heterogeneous Graph-enhanced LSTM (HGLTSM), which is a method that predicts weekly hog price. In this paper, we first extract the historical prices of necessary agricultural products in recent years. Then, we utilize discussions from the online professional community to build heterogeneous graphs. These graphs have rich information of both discussions and the engaged users. Finally, we construct HGLSTM to make the prediction. The experimental results demonstrate that forum discussions are beneficial to hog price prediction. Moreover, our method exhibits a better performance than existing methods.

scholarly journals Part-Aware Mask-Guided Attention for Thorax Disease Classification

Entropy ◽  
2021 ◽  
Vol 23 (6) ◽  
pp. 653
Author(s):  
Ruihua Zhang ◽  
Fan Yang ◽  
Yan Luo ◽  
Jianyi Liu ◽  
Jinbin Li ◽  
...  
Keyword(s):  
Visual Cues ◽  
Learning Strategy ◽  
Feature Learning ◽  
Classification Performance ◽  
Disease Classification ◽  
Feature Representations ◽  
Chest X Ray ◽  
Global And Local ◽  
Aided Diagnosis ◽  
Single Organ

Thorax disease classification is a challenging task due to complex pathologies and subtle texture changes, etc. It has been extensively studied for years largely because of its wide application in computer-aided diagnosis. Most existing methods directly learn global feature representations from whole Chest X-ray (CXR) images, without considering in depth the richer visual cues lying around informative local regions. Thus, these methods often produce sub-optimal thorax disease classification performance because they ignore the very informative pathological changes around organs. In this paper, we propose a novel Part-Aware Mask-Guided Attention Network (PMGAN) that learns complementary global and local feature representations from all-organ region and multiple single-organ regions simultaneously for thorax disease classification. Specifically, multiple innovative soft attention modules are designed to progressively guide feature learning toward the global informative regions of whole CXR image. A mask-guided attention module is designed to further search for informative regions and visual cues within the all-organ or single-organ images, where attention is elegantly regularized by automatically generated organ masks and without introducing computation during the inference stage. In addition, a multi-task learning strategy is designed, which effectively maximizes the learning of complementary local and global representations. The proposed PMGAN has been evaluated on the ChestX-ray14 dataset and the experimental results demonstrate its superior thorax disease classification performance against the state-of-the-art methods.

A novel astrophysics-based framework for prediction of binding affinity of glucose binder

2020 ◽  
Vol 34 (31) ◽  
pp. 2050346
Author(s):  
Rajesh Kondabala ◽  
Vijay Kumar ◽  
Amjad Ali ◽  
Manjit Kaur
Keyword(s):  
Decision Tree ◽  
Binding Affinity ◽  
Cross Validation ◽  
Learning Strategy ◽  
Experimental Results ◽  
The Other ◽  
Computational Time ◽  
Glucose Binding ◽  
Regression Algorithms

In this paper, a novel astrophysics-based prediction framework is developed for estimating the binding affinity of a glucose binder. The proposed framework utilizes the molecule properties for predicting the binding affinity. It also uses the astrophysics-learning strategy that incorporates the concepts of Kepler’s law during the prediction process. The proposed framework is compared with 10 regression algorithms over ZINC dataset. Experimental results reveal that the proposed framework provides 99.30% accuracy of predicting binding affinity. However, decision tree provides the prediction with 97.14% accuracy. Cross-validation results show that the proposed framework provides better accuracy than the other existing models. The developed framework enables researchers to screen glucose binder rapidly. It also reduces computational time for designing small glucose binding molecule.

scholarly journals Scene Recognition Based on Recurrent Memorized Attention Network

Electronics ◽  
2020 ◽  
Vol 9 (12) ◽  
pp. 2038
Author(s):  
Xi Shao ◽  
Xuan Zhang ◽  
Guijin Tang ◽  
Bingkun Bao
Keyword(s):  
Recognition Accuracy ◽  
State Of The Art ◽  
Classification Performance ◽  
Scene Recognition ◽  
Experimental Results ◽  
Scene Classification ◽  
Attention Network ◽  
Attention Model ◽  
Scene Image ◽  
Object Based

We propose a new end-to-end scene recognition framework, called a Recurrent Memorized Attention Network (RMAN) model, which performs object-based scene classification by recurrently locating and memorizing objects in the image. Based on the proposed framework, we introduce a multi-task mechanism that contiguously attends on the different essential objects in a scene image and recurrently performs memory fusion of the features of object focused by an attention model to improve the scene recognition accuracy. The experimental results show that the RMAN model has achieved better classification performance on the constructed dataset and two public scene datasets, surpassing state-of-the-art image scene recognition approaches.

scholarly journals Atrial Fibrillation Detection Directly from Compressed ECG with the Prior of Measurement Matrix

Information ◽  
2020 ◽  
Vol 11 (9) ◽  
pp. 436
Author(s):  
Yunfei Cheng ◽  
Ying Hu ◽  
Mengshu Hou ◽  
Tongjie Pan ◽  
Wenwen He ◽  
...  
Keyword(s):  
Atrial Fibrillation ◽  
Compressed Sensing ◽  
Health Monitoring ◽  
Prior Information ◽  
Classification Performance ◽  
Experimental Results ◽  
Measurement Matrix ◽  
Ecg Signals ◽  
Proposed Model ◽  
Atrial Fibrillation Detection

In the wearable health monitoring based on compressed sensing, atrial fibrillation detection directly from the compressed ECG can effectively reduce the time cost of data processing rather than classification after reconstruction. However, the existing methods for atrial fibrillation detection from compressed ECG did not fully benefit from the existing prior information, resulting in unsatisfactory classification performance, especially in some applications that require high compression ratio (CR). In this paper, we propose a deep learning method to detect atrial fibrillation directly from compressed ECG without reconstruction. Specifically, we design a deep network model for one-dimensional ECG signals, and the measurement matrix is used to initialize the first layer of the model so that the proposed model can obtain more prior information which benefits improving the classification performance of atrial fibrillation detection from compressed ECG. The experimental results on the MIT-BIH Atrial Fibrillation Database show that when the CR is 10%, the accuracy and F1 score of the proposed method reach 97.52% and 98.02%, respectively. Compared with the atrial fibrillation detection from original ECG, the corresponding accuracy and F1 score are only reduced by 0.88% and 0.69%. Even at a high CR of 90%, the accuracy and F1 score are still only reduced by 6.77% and 5.31%, respectively. All of the experimental results demonstrate that the proposed method is superior to other existing methods for atrial fibrillation detection from compressed ECG. Therefore, the proposed method is promising for atrial fibrillation detection in wearable health monitoring based on compressed sensing.

scholarly journals Cooperative Carrying Control for Multi-Evolutionary Mobile Robots in Unknown Environments

Electronics ◽  
2019 ◽  
Vol 8 (3) ◽  
pp. 298 ◽  
Author(s):  
Jyun-Yu Jhang ◽  
Cheng-Jian Lin ◽  
Kuu-Young Young
Keyword(s):  
Differential Evolution ◽  
Mobile Robots ◽  
Target Location ◽  
Learning Strategy ◽  
Control Method ◽  
Behavioral Control ◽  
Differential Evolution Algorithm ◽  
Experimental Results ◽  
Unknown Environment ◽  
Navigation Control

This study provides an effective cooperative carrying and navigation control method for mobile robots in an unknown environment. The manager mode switches between two behavioral control modes—wall-following mode (WFM) and toward-goal mode (TGM)—based on the relationship between the mobile robot and the unknown environment. An interval type-2 fuzzy neural controller (IT2FNC) based on a dynamic group differential evolution (DGDE) is proposed to realize the carrying control and WFM control for mobile robots. The proposed DGDE uses a hybrid method that involves a group concept and an improved differential evolution to overcome the drawbacks of the traditional differential evolution algorithm. A reinforcement learning strategy was adopted to develop an adaptive WFM control and achieve cooperative carrying control for mobile robots. The experimental results demonstrated that the proposed DGDE is superior to other algorithms at using WFM control. Moreover, the experimental results demonstrate that the proposed method can complete the task of cooperative carrying, and can realize navigation control to enable the robot to reach the target location.

GENETIC ALGORITHMS FOR POLYGONAL APPROXIMATION OF DIGITAL CURVES

1999 ◽  
Vol 13 (07) ◽  
pp. 1061-1082 ◽  
Author(s):  
PENG-YENG YIN
Keyword(s):  
Genetic Algorithm ◽  
Genetic Algorithms ◽  
Learning Strategy ◽  
Approximation Error ◽  
Experimental Results ◽  
Polygonal Approximation ◽  
The Third ◽  
Digital Curve

In this paper, three polygonal approximation approaches using genetic algorithms are proposed. The first approach approximates the digital curve by minimizing the number of sides of the polygon and the approximation error should be less than a prespecified tolerance value. The second approach minimizes the approximation error by searching for a polygon with a given number of sides. The third approach, which is more practical, determines the approximating polygon automatically without any given condition. Moreover, a learning strategy for each of the proposed genetic algorithm is presented to improve the results. The experimental results show that the proposed approaches have better performances than those of existing methods.

A New SVM Multi-Class Classification Algorithm Based on Sample Scale and Distribution Area

2013 ◽  
Vol 712-715 ◽  
pp. 2529-2533
Author(s):  
Yu Ping Qin ◽  
Peng Da Qin ◽  
Shu Xian Lun ◽  
Yi Wang
Keyword(s):  
Binary Tree ◽  
Classification Performance ◽  
Classification Algorithm ◽  
Experimental Results ◽  
Leaf Node ◽  
Distribution Area ◽  
Root Node ◽  
Multi Class Classification

A new SVM multi-class classification algorithm is proposed. Firstly, the optimal binary tree is constructed by the scale and the distribution area of every class sample, and then the sub-classifiers are trained for every non-leaf node in the binary tree. For the sample to be classified, the classification is done from the root node until someone leaf node, and the corresponding class of the leaf node is the class of the sample. The experimental results show that the algorithm improves the classification precision and classification speed, especially in the situation that the sample scale is less but its distribution area is bigger, the algorithm can improve greatly the classification performance.

Combining Expert Strategies in Multimodal Classification

2014 ◽  
Vol 590 ◽  
pp. 693-697
Author(s):  
Jing Peng
Keyword(s):  
Real World ◽  
Classification Performance ◽  
Experimental Results ◽  
Robust Classification ◽  
Multimodal Classification ◽  
Novel Algorithm

In multimodal classification, we look for a set of strategies for mining and exploiting the most informative modalities for a given situation. These strategies are computations performed by the algorithms. In this paper, we propose to consider strategies as advice given to an algorithm by “expert.” There can be several classification strategies. Each strategy makes different assumptions regarding the fidelity of a sensor modality and uses different data to arrive at its estimates. Each strategy may place different trust in a sensor at different times, and each may be better in different situations. In this paper, we introduce a novel algorithm for combining expert strategies to achieve robust classification performance in a multimodal setting. We provide experimental results using real world examples to demonstrate the efficacy of the proposed algorithm.

Sign in / Sign up

Export Citation Format

Share Document