Tensor-Based Classification Models for Hyperspectral Data Analysis

Most of the existing machine learning methods for charge prediction adopt the training mechanism of supervised learning. These algorithms have high requirements for the number of training samples corresponding to each crime. However, few or no cases are corresponding to some crimes in real scenarios, which leads to the poor performance of these models in practice. To alleviate this problem, we propose a novel Zero-Shot Learning (ZSL) based method for legal charge prediction tasks. Specifically, we define a set of semantic attributes to represent the domain knowledge of charges, which enables the model to migrate knowledge from seen charges to unseen charges. In this way, with the help of the ZSL mechanism, unseen charges and charges with a small number of training samples could be relatively predicted accurately. We evaluate the performance of the proposed method on a dataset collected from China Judgements Online, and the experimental results show that our method obtains \(32.4\%\) accuracy for the unseen charges and can largely retain the predictive power for the seen charges.

Practical applications of online handwritten character recognition demand robust and highly accurate recognition along with low memory requirements. The Active-DTW~\cite{activedtw} classifier proposed by Sridhar {\it et al}. combines the advantages of generative and discriminative classifiers to address the similarity of between-class samples, while taking into account the variability of writing styles within the same character class. Active-DTW uses Active Shape Models to model the significant writing styles in a memory-efficient manner.However, in order to create accurate models, a large number of training samples is needed up front, which is not desirable or available in many practical applications. In this paper, we propose a supervised adaptation framework for the Active-DTW classifier which allows recognition to begin with a small number of training samples, and adapts the classifier to the new samples presented to the system during recognition. We compare the performance of Active-DTW using the proposed adaptation framework, with a nearest-neighbor classifier using an LVQ-based adaptation scheme, on the online handwritten Tamil character dataset.

Hyperspectral images (HSIs) obtained from remote sensing contain abundant information of ground objects, and precise analysis of landcover depends on effective and efficient classification of HSIs into homogeneous ground regions. While many advanced algorithms have been developed for HSI classification, it is a challenge for an algorithm to achieve a good balance between its effectiveness and efficiency due to the high dimensionality of HSIs and insufficient labeled training samples. By taking both the rich spectral features and the spatially homogeneous property of land cover distributions, in this paper, we propose a simple and efficient yet effective method for HSI classification. First, features are extracted by using a weighted spatial-spectral and global-local discriminant analysis algorithm, which is proposed to reduce the feature dimension. Then, combining the discriminant information of neighboring pixels, we propose a spatial collaboration nearest neighbor (SC-NN) classifier to make reliable class judgment for the central one. In the SC-NN classifier, the spatially homogeneous property of landcover distribution of HSIs is utilized to effectively reduce the probability of misclassification when only a small number of training samples are available. To further address the issue of a small number of training samples, we adopt an incremental learning strategy by adding the samples with high classification confidence to the training set. Experimental results on four public datasets show that our proposed method outperforms several state-of-the-art methods with high classification accuracy.

With the wide deployment of wireless networks, channel state information (CSI)-based joint recognition of location and activity boasts a wide range of application prospects in the fields of smart homes, remote health monitoring, and security alarm. However, the existing algorithms for joint recognition of location and activity can achieve high recognition accuracy only when trained on a large number of samples. To address this problem, this study proposes a novel, small-sample joint “location and activity” recognition algorithm based on residual networks (ResNets) with multiattention layers, referred to as the SSJR algorithm. In the algorithm, a small number of training samples are expanded via subtraction between CSI amplitude data of multiple links and horizontal flipping of images. Furthermore, a peak-seeking algorithm is proposed to decompose CSI data into location and activity recognition components, both of which are input into the constructed ResNets with multiattention layers for joint recognition. The algorithm can not only expand number of samples but also accurately extract and enhance the features of each category of samples. Experimental results indicate that the SSJR algorithm achieves an average recognition accuracy of 96.83% with only five training samples per category, demonstrating both its practicality and feasibility for human-computer interaction applications.

This paper reports the scaling behavior of the fuzzy c-means based classifier (FCMC) with many parameters. FCMC is a classifier based on clustering approaches. The classification accuracy on test sets (i.e., the generalization capability) is not necessarily improved by increasing the number of clusters. Especially when the number of training samples is relatively small, not only the classification boundary over-fits the data, but also covariance matrices and cluster centers are computed incorrectly, since the number of samples in each cluster becomes smaller. Hence, the test set accuracy deteriorates. The performance of FCMC with two clusters in each class and the number of training samples less than 1000, was reported in the literature. This paper reports the scaling behavior of FCMC by testing with variously-sized training samples. The number of clusters of FCMC is increased up to eight. The number of clusters used in this paper is not very large but the number of parameters is relatively large. So, the parameters are optimized to training sets. LibSVM is one of the widely known state of the art tools for support vector machines (SVM). The test set accuracy, training time and testing time (i.e., the detection time) of FCMC are compared with LibSVM by varying the size of training sets. FCMC shows a good generalization capability, though the parameters are optimized to training sets. When the number of training samples is increased by 10 times, the training time of FCMC increases by 10 times, but that of LibSVM increases by a factor of 100. The testing time is also much shorter than LibSVM when the size of the training set is large.

Linear classifier design in the weight space

This paper tackles the challenging task of automatically predicting the floor of a balcony based on a sound bound of two minutes recorded on that balcony. The sound fragments are typical of the environment, as nothing out of the usual can be heard in them. However, there is a good probability that, when hearing a fragment in quiet surroundings with consistent metropolitan background noise, it would not be straightforward to determine or estimate the floor height. In our experiments, it was found that sound chunks lasting 5 s can be identified with high accuracy even with a small number of training samples. In addition, when using a late fusion strategy to combine the outputs of classifiers trained on two modalities of the sound tracks, the floors of these bands are perfectly correctly classified. This result was consistent throughout all twenty tests when training and test sets were chosen at random, supporting the viability of the suggested method.

How Can a Massive Training Artificial Neural Network (MTANN) Be Trained With a Small Number of Cases in the Distinction Between Nodules and Vessels in Thoracic CT? 1

Multimodal Earth observation data fusion: Graph-based approach in shared latent space

PurposeThe widespread acceptance of various social platforms has increased the number of users posting about various services based on their experiences about the services. Finding out the intended ratings of social media (SM) posts is important for both organizations and prospective users since these posts can help in capturing the user’s perspectives. However, unlike merchant websites, the SM posts related to the service-experience cannot be rated unless explicitly mentioned in the comments. Additionally, predicting ratings can also help to build a database using recent comments for testing recommender algorithms in various scenarios.Design/methodology/approachIn this study, the authors have predicted the ratings of SM posts using linear (Naïve Bayes, max-entropy) and non-linear (k-nearest neighbor, k-NN) classifiers utilizing combinations of different features, sentiment scores and emotion scores.FindingsOverall, the results of this study reveal that the non-linear classifier (k-NN classifier) performed better than the linear classifiers (Naïve Bayes, Max-entropy classifier). Results also show an improvement of performance where the classifier was combined with sentiment and emotion scores. Introduction of the feature “factors of importance” or “the latent factors” also show an improvement of the classifier performance.Originality/valueThis study provides a new avenue of predicting ratings of SM feeds by the use of machine learning algorithms along with a combination of different features like emotional aspects and latent factors.

Class imbalance is a challenging problem in many classification tasks. It induces biased classification results for minority classes that contain less training samples than others. Most existing approaches aim to remedy the imbalanced number of instances among categories by resampling the majority and minority classes accordingly. However, the imbalanced level of difficulty of recognizing different categories is also crucial, especially for distinguishing samples with many classes. For example, in the task of clinical skin disease recognition, several rare diseases have a small number of training samples, but they are easy to diagnose because of their distinct visual properties. On the other hand, some common skin diseases, e.g., eczema, are hard to recognize due to the lack of special symptoms. To address this problem, we propose a self-paced balance learning (SPBL) algorithm in this paper. Specifically, we introduce a comprehensive metric termed the complexity of image category that is a combination of both sample number and recognition difficulty. First, the complexity is initialized using the model of the first pace, where the pace indicates one iteration in the self-paced learning paradigm. We then assign each class a penalty weight that is larger for more complex categories and smaller for easier ones, after which the curriculum is reconstructed by rearranging the training samples. Consequently, the model can iteratively learn discriminative representations via balancing the complexity in each pace. Experimental results on the SD-198 and SD-260 benchmark data sets demonstrate that the proposed SPBL algorithm performs favorably against the state-of-the-art methods. We also demonstrate the effectiveness of the SPBL algorithm's generalization capacity on various tasks, such as indoor scene image recognition and object classification.

In this paper, a graph regularized nonlinear ridge regression (RR) model is proposed for remote sensing data analysis, including hyper-spectral image classification and atmospheric aerosol retrieval. The RR is an efficient linear regression method, especially in handling cases with a small number of training samples or with correlated features. However, large amounts of unlabeled samples exist in remote sensing data analysis. To sufficiently explore the information in unlabeled samples, we propose a graph regularized RR (GRR) method, where the vertices denote labeled or unlabeled samples and the edges represent the similarities among different samples. A natural assumption is that the predict values of neighboring samples are close to each other. To further address the nonlinearly separable problem in remote sensing data caused by the complex acquisition process as well as the impacts of atmospheric and geometric distortions, we extend the proposed GRR into a kernelized nonlinear regression method, namely KGRR. To evaluate the proposed method, we apply it to both classification and regression tasks and compare with representative methods. The experimental results show that KGRR can achieve favorable performance in terms of predictability and computation time.

Imaging spectroscopy as a tool to study sediment characteristics on a tidal sandbank in the Westerschelde

Markov random field (MRF) based methods have been widely used in high spatial resolution (HSR) image classification. However, many existing MRF-based methods put more emphasis on pixel level contexts while less on superpixel level contextual information. To cope with this issue, this article presents a novel bilevel contextual MRF framework, named BLC-MRF, for HSR imagery classification. Specifically, pixel and superpixel level dependence are incorporated into the proposed MRF model to fully exploit spectral–spatial contextual information and preserve object boundaries in HSR images. In BLC-MRF, a pixel level MRF model is first performed and then cascaded as an input of a superpixel level MRF. In superpixel level, unary and pairwise potential terms are constructed by using the superpixel probability estimation method and spectral histogram distance, respectively. At last, a contextual MRF model is conducted and the final classification map can be computed by using $\alpha$ -expansion algorithm. The benefits of BLC-MRF are twofold: first, the pixel and superpixel level contextual information can be exploited under MRF framework to preserve object boundaries for improving the classification performance, and, second, the algorithm can provide promising results with a small number of training samples. Experimental results on three HSR datasets demonstrate that the proposed approach outperforms several state-of-the-art methods in terms of the classification performance.

Exploring visual representations of computer mouse movements for bot detection using deep learning approaches