Support Vector machine (SVM) is a kind of machinelearning method based on the statistical learning theory, it has been applied in the fault diagnosis field. After analyzing SVM pattern classification theory, a hierar...
详细信息
ISBN:
(纸本)0769528759
Support Vector machine (SVM) is a kind of machinelearning method based on the statistical learning theory, it has been applied in the fault diagnosis field. After analyzing SVM pattern classification theory, a hierarchical structure Fault Detection and Identification (FDI) system is presented in this paper, and simulation results show that this method can effectively handle the complex process characteristic and improve FDI model performance.
data analysis methods and techniques are revisited in the case of biological data sets. Particular emphasis is given to clustering and mining issues. Clustering is still a subject of active research in several fields ...
详细信息
ISBN:
(纸本)9783540770459
data analysis methods and techniques are revisited in the case of biological data sets. Particular emphasis is given to clustering and mining issues. Clustering is still a subject of active research in several fields such as statistics, patternrecognition, and machinelearning. datamining adds to clustering the complications of very large data-sets with many attributes of different types. And this is a typical situation in biology. Some cases studies are also described.
Dimension reduction methods are often applied in machinelearning and datamining problems. Linear subspace methods are the commonly used ones, such as principal component analysis (PCA), Fisher's linear discrimin...
详细信息
ISBN:
(数字)9783540734994
ISBN:
(纸本)9783540734987
Dimension reduction methods are often applied in machinelearning and datamining problems. Linear subspace methods are the commonly used ones, such as principal component analysis (PCA), Fisher's linear discriminant analysis (FDA), et al. In this paper, we describe a novel feature extraction method for binary classification problems. Instead of finding linear subspaces, our method finds lower-dimensional affine subspaces for data observations. Our method can be understood as a generalization of the Fukunaga-Koontz Transformation. We show that the proposed method has a closed-form solution and thus can be solved very efficiently. Also we investigate the information-theoretical properties of the new method and study the relationship of our method with other methods. The experimental results show that our method, as PCA and FDA, can be used as another preliminary data-exploring tool to help solve machinelearning and datamining problems.
In this paper, we propose a hybrid activity recognition method for ballroom dance exercise using video and wearable sensor. The purpose of our research is to design a mechanism to support ballroom dance exercise, and ...
详细信息
ISBN:
(纸本)9781728107882
In this paper, we propose a hybrid activity recognition method for ballroom dance exercise using video and wearable sensor. The purpose of our research is to design a mechanism to support ballroom dance exercise, and this paper reports the first part to design a mechanism is to support ballroom dance exercise, and this paper reports the first outcome to achieve the purpose - recognizing ballroom dance exercise. There are two conceivable ways to recognize dance exercise: videos and wearable sensor. However, each of them has its disadvantages. Using video is a good way to recognize the movement of the body. However, it cannot provide us accurate timing or strength of foot actions because the number of their flames per seconds is too small to recognize the fast movements of dancers. On the other hand, while a wearable sensor is good at recognizing foot timing and strength, it is not good at recognizing the movement of the whole body. Therefore we propose a hybrid recognition method utilizing the merits of both video and wearable sensor. This paper focuses to recognize four different types of steps in Latin American, a kind of ballroom dance. For each step, we record wearable sensing data and videos. As a result, it is found that the accuracy of step recognition is improved by adding wearable sensing data to video data shot from two different angles.
In these recent years, kernel methods have gained a considerable interest in many areas of machinelearning. This work investigates the ability of kernel clustering methods to deal with one of the meaningful problem o...
详细信息
data augmentation has been a prevalent approach in improving the performance of deep learning models against slight variations in data. Adversarial learning is one such form of data augmentation. In this work, we aim ...
详细信息
ISBN:
(数字)9781665490627
ISBN:
(纸本)9781665490627
data augmentation has been a prevalent approach in improving the performance of deep learning models against slight variations in data. Adversarial learning is one such form of data augmentation. In this work, we aim to introduce a framework to generate harder examples for a specific object class and an adversarial attack for the object detection task. We have also presented our study on the effect of training against such generated harder examples and adversarial samples in object detection. We have applied this adversarial learning technique to a YOLOv3 model and due to the nature of the attack, we demonstrated a substantial improvement in average precision (AP) for a single class of the COCO dataset. As per the literature, we are the first to introduce this kind of class-specific data augmentation strategy in object detection. With our approach, we have shown an improvement of 23.34% in AP for Cat class and 3.1% on overall mAP of YOLOv3 model on clean validation data, while 43.5% improvement in AP for the Cat class on the composite images with class-specific adversarial samples.
This paper uses pulsar signal data for datamining, on the basis of exploratory analysis, constructs a variety of classification models, such as Random Forest, SVM, Logical Regression, K-Nearest Neighbor, Naive Bayes,...
详细信息
ISBN:
(纸本)9781728160788
This paper uses pulsar signal data for datamining, on the basis of exploratory analysis, constructs a variety of classification models, such as Random Forest, SVM, Logical Regression, K-Nearest Neighbor, Naive Bayes, Decision tree, AdaBoost classifier, GBDT and XGBoost, to classify pulsar candidate samples. It is hoped that valuable suspected pulsar samples can be effectively screened from massive data for further observation and confirmation.
Binary decision diagrams (BDD) is a compact and efficient representation of Boolean functions with extensions available for sets and finite-valued functions. The key feature of the BDD is an ability to employ internal...
详细信息
ISBN:
(纸本)9783319089799;9783319089782
Binary decision diagrams (BDD) is a compact and efficient representation of Boolean functions with extensions available for sets and finite-valued functions. The key feature of the BDD is an ability to employ internal structure (not necessary known upfront) of an object being modelled in order to provide a compact in-memory representation. In this paper we propose application of the BDD for machinelearning as a tool for fast general patternrecognition. Multiple BDDs are used to capture a sets of training samples (patterns) and to estimate the similarity of a given test sample with the memorized training sets. Then, having multiple similarity estimates further analysis is done using additional layer of BDDs or common machinelearning techniques. We describe training algorithms for BDDs (supervised, unsupervised and combined), an approach for constructing multi-layered networks combining BDDs with traditional artificial neurons and present experimental results for handwritten digits recognition on the MNIST dataset.
With the advancement of modern Internet technology, email is widely used in people39;s daily lives and has become one of the common communication tools. However, at the same time, email-based spam has also been wide...
详细信息
Bangla is one of the mostly spoken languages all over the world, nonetheless, efforts on Bangla handwritten character recognition is not adequate. The use of Deep Convolutional Neural Network (DCNN) based classifiers ...
详细信息
ISBN:
(纸本)9781728107882
Bangla is one of the mostly spoken languages all over the world, nonetheless, efforts on Bangla handwritten character recognition is not adequate. The use of Deep Convolutional Neural Network (DCNN) based classifiers has become a triumph over the state of art machinelearning techniques. Using DCNN model to classify Bangla isolated basic characters can provide better output because of its ability to detect many hidden features from an image. In this paper, a new DCNN model, namely, BBCNet-15 is proposed for Bangla handwritten basic character recognition. the proposed model consists of 6 convolution layers, 6 max pooling layers, 2 fully connected layers followed by the softmax output unit. To avoid overfitting dropout regularization technique is used. The implementation of the proposed DCNN model is evaluated on a benchmark dataset, CMATErdb 3.1.2 which contains 50 character classes including 39 consonants and 11 vowels. The experimental evaluation of BBCNet-15 on the dataset provides a recognition accuracy of 96.40% which outperforms some prominent techniques in existence.
暂无评论