The ability to automatically determine the age audience of a novel provides many opportunities for the development of information retrieval tools. Firstly, developers of book recommendation systems and electronic libraries may be interested in filtering texts by the age of the most likely readers. Further, parents may want to select literature for children. Finally, it will be useful for writers and publishers to determine which features influence whether the texts are suitable for children. In this article, we compare the empirical effectiveness of various types of linguistic features for the task of age-based classification of fiction texts. For this purpose, we collected a text corpus of book previews labeled with one of two categories -- children's or adult. We evaluated the following types of features: readability indices, sentiment, lexical, grammatical and general features, and publishing attributes. The results obtained show that the features describing the text at the document level can significantly increase the quality of machine learning models.
Deep learning has gained great success in various classification tasks. Typically, deep learning models learn underlying features directly from data, and no underlying relationship between classes are included. Similarity between classes can influence the performance of classification. In this article, we propose a method that incorporates class similarity knowledge into convolutional neural networks models using a graph convolution layer. We evaluate our method on two benchmark image datasets: MNIST and CIFAR10, and analyze the results on different data and model sizes. Experimental results show that our model can improve classification accuracy, especially when the amount of available data is small.
Recently, multiscale spatial features have been widely utilized to improve the hyperspectral image (HSI) classification performance. However, fixed-size neighborhood involving the contextual information probably leads to misclassifications, especially for the boundary pixels. Additionally, it has been demonstrated that deep neural network (DNN) is practical to extract representative features for the classification tasks. Nevertheless, under the condition of high dimensionality versus small sample sizes, DNN tends to be over-fitting and it is generally time-consuming due to the deep-level feature learning process. To alleviate the aforementioned issues, we propose a multiscale context-aware ensemble deep kernel extreme learning machine (MSC-EDKELM) for efficient HSI classification. First, the scene of the HSI data set is over-segmented in multiscale via using the adaptive superpixel segmentation technique. Second, superpixel pattern (SP) and attentional neighboring superpixel pattern (ANSP) are generated by leveraging the superpixel maps, which can automatically comprise local and global contextual information, respectively. Afterward, an ensemble deep kernel extreme learning machine (EDKELM) is presented to investigate the deep-level characteristics in the SP and ANSP. Finally, the category of each pixel is accurately determined by the decision fusion and weighted output layer fusion strategy. Experimental results on four real-world HSI data sets demonstrate that the proposed frameworks outperform some classic and state-of-theart methods with high computational efficiency, which can be employed to serve real-time applications.
Text classification has become a key operation in various natural language processing tasks. The efficiency of most classification algorithms predominantly confide in the quality of input features. In this work, we propose a novel multi-class text classification technique that harvests features from two distinct feature extraction methods. Firstly, a structured heterogeneous text graph built based on document-word relations and word co-occurrences is leveraged using a Graph Convolution Network (GCN). Secondly, the documents are topic modeled to use the document-topic score as features into the classification model. The concerned graph is constructed using Point-Wise Mutual Information (PMI) between pair of word co-occurrences and Term Frequency-Inverse Document Frequency (TF-IDF) score for words in the documents for word co-occurrences. Experimentation reveals that our text classification model outperforms the existing techniques for five benchmark text classification data sets.
Recently, convolutional neural networks (CNNs) have attracted enormous attention in pattern recognition and demonstrated excellent performance in hyperspectral image (HSI) classification. However, high-dimensional HSI dataset versus limited training samples is easy to cause the overfitting phenomenon in deep neural networks. Additionally, the intraclass distance of the embedding features extracted through the softmax-based CNNs may be greater than that of the interclass, which makes it difficult to further improve the classification accuracy. To address these issues, this article proposes a deep prototypical network with hybrid residual attention, which can effectively investigate the spectral-spatial information in the HSI. Specifically, in order to improve the generalization capability of the model, feature extraction with a hybrid residual attention module is presented to enhance the critical spectral-spatial features and suppress the useless ones in the classification task. Furthermore, a novel discriminant distance-based cross-entropy loss is proposed to increase the intraclass compactness, to obtain more superior results. Extensive experiments on three benchmark datasets are carried out to convincingly evaluate the proposed framework. With the generation of optimal prototypes representing each class and more discriminative embedding features, encouraging classification results are achieved compared with state-of-the-art methods.
Network embedding is a highly effective method to learn low-dimensional node vector representations with original network structures being well preserved. However, existing network embedding algorithms are mostly developed for a single network, which fail to learn generalized feature representations across different networks. In this paper, we study a cross-network node classification problem, which aims at leveraging the abundant labeled information from a source network to help classify the unlabeled nodes in a target network. To succeed in such a task, transferable features should be learned for nodes across different networks. To this end, a novel cross-network deep network embedding (CDNE) model is proposed to incorporate domain adaptation into deep network embedding so as to learn label-discriminative and network-invariant node vector representations. On one hand, CDNE leverages network structures to capture the proximities between nodes within a network, by mapping more strongly connected nodes to have more similar latent vector representations. On the other hand, node attributes and labels are leveraged to capture the proximities between nodes across different networks by making the same labeled nodes across networks have aligned latent vector representations. Extensive experiments have been conducted, demonstrating that the proposed CDNE model significantly outperforms the state-of-the-art network embedding algorithms in cross-network node classification.
The latest research in expertise assessment of soccer players has affirmed the importance of perceptual skills (especially for decision making) by focusing either on high experimental control or on a realistic presentation. To assess the perceptual skills of athletes in an optimized manner, we captured omnidirectional in-field scenes and showed these to 12 expert, 10 intermediate and 13 novice soccer goalkeepers on virtual reality glasses. All scenes were shown from the same natural goalkeeper perspective and ended after the return pass to the goalkeeper. Based on their gaze behavior we classified their expertise with common machine learning techniques. This pilot study shows promising results for objective classification of goalkeepers expertise based on their gaze behaviour and provided valuable insight to inform the design of training systems to enhance perceptual skills of athletes.
Graph embedding is a powerful method to represent graph neurological data (e.g., brain connectomes) in a low dimensional space for brain connectivity mapping, prediction and classification. However, existing embedding algorithms have two major limitations. First, they primarily focus on preserving one-to-one topological relationships between nodes (i.e., regions of interest (ROIs) in a connectome), but they have mostly ignored many-to-many relationships (i.e., set to set), which can be captured using a hyperconnectome structure. Second, existing graph embedding techniques cannot be easily adapted to multi-view graph data with heterogeneous distributions. In this paper, while cross-pollinating adversarial deep learning with hypergraph theory, we aim to jointly learn deep latent embeddings of subject0specific multi-view brain graphs to eventually disentangle different brain states. First, we propose a new simple strategy to build a hyperconnectome for each brain view based on nearest neighbour algorithm to preserve the connectivities across pairs of ROIs. Second, we design a hyperconnectome autoencoder (HCAE) framework which operates directly on the multi-view hyperconnectomes based on hypergraph convolutional layers to better capture the many-to-many relationships between brain regions (i.e., nodes). For each subject, we further regularize the hypergraph autoencoding by adversarial regularization to align the distribution of the learned hyperconnectome embeddings with that of the input hyperconnectomes. We formalize our hyperconnectome embedding within a geometric deep learning framework to optimize for a given subject, thereby designing an individual-based learning framework. Our experiments showed that the learned embeddings by HCAE yield to better results for brain state classification compared with other deep graph embedding methods methods.
Automated gender classification has important applications in many domains, such as demographic research, law enforcement, online advertising, as well as human-computer interaction. Recent research has questioned the fairness of this technology across gender and race. Specifically, the majority of the studies raised the concern of higher error rates of the face-based gender classification system for darker-skinned people like African-American and for women. However, to date, the majority of existing studies were limited to African-American and Caucasian only. The aim of this paper is to investigate the differential performance of the gender classification algorithms across gender-race groups. To this aim, we investigate the impact of (a) architectural differences in the deep learning algorithms and (b) training set imbalance, as a potential source of bias causing differential performance across gender and race. Experimental investigations are conducted on two latest large-scale publicly available facial attribute datasets, namely, UTKFace and FairFace. The experimental results suggested that the algorithms with architectural differences varied in performance with consistency towards specific gender-race groups. For instance, for all the algorithms used, Black females (Black race in general) always obtained the least accuracy rates. Middle Eastern males and Latino females obtained higher accuracy rates most of the time. Training set imbalance further widens the gap in the unequal accuracy rates across all gender-race groups. Further investigations using facial landmarks suggested that facial morphological differences due to the bone structure influenced by genetic and environmental factors could be the cause of the least performance of Black females and Black race, in general.
Graph convolutional networks (GCNs) have achieved promising performance on various graph-based tasks. However they suffer from over-smoothing when stacking more layers. In this paper, we present a quantitative study on this observation and develop novel insights towards the deeper GCN. First, we interpret the current graph convolutional operations from an optimization perspective and argue that over-smoothing is mainly caused by the naive first-order approximation of the solution to the optimization problem. Subsequently, we introduce two metrics to measure the over-smoothing on node-level tasks. Specifically, we calculate the fraction of the pairwise distance between connected and disconnected nodes to the overall distance respectively. Based on our theoretical and empirical analysis, we establish a universal theoretical framework of GCN from an optimization perspective and derive a novel convolutional kernel named GCN+ which has lower parameter amount while relieving the over-smoothing inherently. Extensive experiments on real-world datasets demonstrate the superior performance of GCN+ over state-of-the-art baseline methods on the node classification tasks.
Wildlife monitoring is crucial to nature conservation and has been done by manual observations from motion-triggered camera traps deployed in the field. Widespread adoption of such in-situ sensors has resulted in unprecedented data volumes being collected over the last decade. A significant challenge exists to process and reliably identify what is in these images efficiently. Advances in computer vision are poised to provide effective solutions with custom AI models built to automatically identify images of interest and label the species in them. Here we outline the data unification effort for the Wildlife Insights platform from various conservation partners, and the challenges involved. Then we present an initial deep convolutional neural network model, trained on 2.9M images across 465 fine-grained species, with a goal to reduce the load on human experts to classify species in images manually. The long-term goal is to enable scientists to make conservation recommendations from near real-time analysis of species abundance and population health.
Fuzzy string matching and language classification are important tools in Natural Language Processing pipelines, this paper provides advances in both areas. We propose a fast novel approach to string tokenisation for fuzzy language matching and experimentally demonstrate an 83.6% decrease in processing time with an estimated improvement in recall of 3.1% at the cost of a 2.6% decrease in precision. This approach is able to work even where keywords are subdivided into multiple words, without needing to scan character-to-character. So far there has been little work considering using metadata to enhance language classification algorithms. We provide observational data and find the Accept-Language header is 14% more likely to match the classification than the IP Address.
Sequential sensor data is generated in a wide variety of practical applications. A fundamental challenge involves learning effective classifiers for such sequential data. While deep learning has led to impressive performance gains in recent years in domains such as speech, this has relied on the availability of large datasets of sequences with high-quality labels. In many applications, however, the associated class labels are often extremely limited, with precise labelling/segmentation being too expensive to perform at a high volume. However, large amounts of unlabeled data may still be available. In this paper we propose a novel framework for semi-supervised learning in such contexts. In an unsupervised manner, change point detection methods can be used to identify points within a sequence corresponding to likely class changes. We show that change points provide examples of similar/dissimilar pairs of sequences which, when coupled with labeled, can be used in a semi-supervised classification setting. Leveraging the change points and labeled data, we form examples of similar/dissimilar sequences to train a neural network to learn improved representations for classification. We provide extensive synthetic simulations and show that the learned representations are superior to those learned through an autoencoder and obtain improved results on both simulated and real-world human activity recognition datasets.
Mammography and ultrasound are extensively used by radiologists as complementary modalities to achieve better performance in breast cancer diagnosis. However, existing computer-aided diagnosis (CAD) systems for the breast are generally based on a single modality. In this work, we propose a deep-learning based method for classifying breast cancer lesions from their respective mammography and ultrasound images. We present various approaches and show a consistent improvement in performance when utilizing both modalities. The proposed approach is based on a GoogleNet architecture, fine-tuned for our data in two training steps. First, a distinct neural network is trained separately for each modality, generating high-level features. Then, the aggregated features originating from each modality are used to train a multimodal network to provide the final classification. In quantitative experiments, the proposed approach achieves an AUC of 0.94, outperforming state-of-the-art models trained over a single modality. Moreover, it performs similarly to an average radiologist, surpassing two out of four radiologists participating in a reader study. The promising results suggest that the proposed method may become a valuable decision support tool for breast radiologists.
In this paper, we focus on data augmentation for the extreme multi-label classification (XMC) problem. One of the most challenging issues of XMC is the long tail label distribution where even strong models suffer from insufficient supervision. To mitigate such label bias, we propose a simple and effective augmentation framework and a new state-of-the-art classifier. Our augmentation framework takes advantage of the pre-trained GPT-2 model to generate label-invariant perturbations of the input texts to augment the existing training data. As a result, it present substantial improvements over baseline models. Our contributions are two-factored: (1) we introduce a new state-of-the-art classifier that uses label attention with RoBERTa and combine it with our augmentation framework for further improvement; (2) we present a broad study on how effective are different augmentation methods in the XMC task.
Standard deep learning models that employ the categorical cross-entropy loss are known to perform well at image classification tasks. However, many standard models thus obtained often exhibit issues like feature redundancy, low interpretability, and poor calibration. A body of recent work has emerged that has tried addressing some of these challenges by proposing the use of new regularization functions in addition to the cross-entropy loss. In this paper, we present some surprising findings that emerge from exploring the role of simple orthogonality constraints as a means of imposing physics-motivated constraints common in imaging. We propose an Orthogonal Sphere (OS) regularizer that emerges from physics-based latent-representations under simplifying assumptions. Under further simplifying assumptions, the OS constraint can be written in closed-form as a simple orthonormality term and be used along with the cross-entropy loss function. The findings indicate that orthonormality loss function results in a) rich and diverse feature representations, b) robustness to feature sub-selection, c) better semantic localization in the class activation maps, and d) reduction in model calibration error. We demonstrate the effectiveness of the proposed OS regularization by providing quantitative and qualitative results on four benchmark datasets - CIFAR10, CIFAR100, SVHN and tiny ImageNet.
Extreme multi-label classification (XMLC) is a learning task of tagging instances with a small subset of relevant labels chosen from an extremely large pool of possible labels. Problems of this scale can be efficiently handled by organizing labels as a tree, like in hierarchical softmax used for multi-class problems. In this paper, we thoroughly investigate probabilistic label trees (PLTs) which can be treated as a generalization of hierarchical softmax for multi-label problems. We first introduce the PLT model and discuss training and inference procedures and their computational costs. Next, we prove the consistency of PLTs for a wide spectrum of performance metrics. To this end, we upperbound their regret by a function of surrogate-loss regrets of node classifiers. Furthermore, we consider a problem of training PLTs in a fully online setting, without any prior knowledge of training instances, their features, or labels. In this case, both node classifiers and the tree structure are trained online. We prove a specific equivalence between the fully online algorithm and an algorithm with a tree structure given in advance. Finally, we discuss several implementations of PLTs and introduce a new one, napkinXC, which we empirically evaluate and compare with state-of-the-art algorithms.
Multi-modal fusion has been proved to help enhance the performance of scene classification tasks. This paper presents a 2D-3D Fusion stage that combines 3D Geometric Features with 2D Texture Features obtained by 2D Convolutional Neural Networks. To get a robust 3D Geometric embedding, a network that uses two novel layers is proposed. The first layer, Multi-Neighbourhood Graph Convolution, aims to learn a more robust geometric descriptor of the scene combining two different neighbourhoods: one in the Euclidean space and the other in the Feature space. The second proposed layer, Nearest Voxel Pooling, improves the performance of the well-known Voxel Pooling. Experimental results, using NYU-Depth-V2 and SUN RGB-D datasets, show that the proposed method outperforms the current state-of-the-art in RGB-D indoor scene classification task.
This paper studies the problem of detecting novel or unexpected instances in text classification. In traditional text classification, the classes appeared in testing must have been seen in training. However, in many applications, this is not the case because in testing, we may see unexpected instances that are not from any of the training classes. In this paper, we propose a significantly more effective approach that converts the original problem to a pair-wise matching problem and then outputs how probable two instances belong to the same class. Under this approach, we present two models. The more effective model uses two embedding matrices of a pair of instances as two channels of a CNN. The output probabilities from such pairs are used to judge whether a test instance is from a seen class or is novel/unexpected. Experimental results show that the proposed method substantially outperforms the state-of-the-art baselines.
Distinguishing visually similar objects like forged/authentic bills and healthy/unhealthy plants is beyond the capabilities of even the most sophisticated classifiers. We propose the use of multiplexed illumination to extend the range of objects that can be successfully classified. We construct a compact RGB-IR light stage that images samples under different combinations of illuminant position and colour. We then develop a methodology for selecting illumination patterns and training a classifier using the resulting imagery. We use the light stage to model and synthetically relight training samples, and propose a greedy pattern selection scheme that exploits this ability to train in simulation. We then apply the trained patterns to carry out fast classification of new objects. We demonstrate the approach on visually similar artificial and real fruit samples, showing a marked improvement compared with fixed-illuminant approaches as well as a more conventional code selection scheme. This work allows fast classification of previously indistinguishable objects, with potential applications in forgery detection, quality control in agriculture and manufacturing, and skin lesion classification.