Traditional Classification Problem Research Articles

Scalable deep learning artificial intelligence histopathology slide analysis and validation.

Deep learning involves an artificial intelligence (AI) approach and has been shown to provide superior performance for automating image recognition tasks, as well as exceeding human capabilities in both time and accuracy. Histopathology diagnostics is one of the more popular challenges at the intersection of artificial intelligence, computer vision, and medicine. Developing methods to automatically detect and identify pathologies in digitized histology slides imposes unique challenges due to the large size of these images and the complexity of the features present in biological tissue. Most methods that are capable of human-level recognition in histopathology are tuned to a specific problem since the computational complexity exceeds that of traditional image classification problems. In the current study, a deep learning approach is developed and presented that can be trained to locate and accurately classify different types of pathologies in gigapixel digitized histology slides along with completing the binary disease classification for the entire image. The approach uses a novel pyramid tiling approach to take advantage of spatial awareness around the area to be classified, while maintaining efficiency and scalability for gigapixel images. The approach is trained and validated on a wide variety of tissue types (i.e., testis, ovary, prostate, kidney) and pathologies taken from an epigenetically altered histology study at Washington State University. The newly developed procedure was optimized and validated along with comparison and validation on public histology datasets. The current developed procedure was found to be optimal and more reproducible when compared to manual procedures, and optimal to previous protocols that used fragmented tissue or slide analysis. Observations demonstrate that the deep learning histopathology analysis is significantly more efficient and accurate than standard manual histopathology analysis.

Dense Projection for Anomaly Detection

This work presents a novel method called dense projection for unsupervised anomaly detection (DPAD). The main idea is maximizing the local density of (normal) training data and then determining whether a test data is anomalous or not by evaluating its density. Specifically, DPAD uses a deep neural network to learn locally dense representations of normal data. Since density estimation is computationally expensive, we minimize the local distances of the representations in an iteratively reweighting manner, where the weights are updated adaptively and the parameters are regularized to avoid model collapse (all representations collapse to a single point). Compared with many state-of-the-art methods of anomaly detection, our DPAD does not rely on any assumption about the distribution or spatial structure of the normal data and representations. Moreover, we provide theoretical guarantees for the effectiveness of DPAD. The experiments show that our method DPAD is effective not only in traditional one-class classification problems but also in scenarios with complex normal data composed of multiple classes.

A lightweight deep learning based bowel sounds segmentation algorithm for gastrointestinal (GI) monitoring

Segmentation of bowel sounds (BS) events is a significant task of automatic BS monitoring. Recently, deep learning (DL) has been utilized to realize the segmentation of BS events. However, most researchers treat BS segmentation as a traditional classification problem, which causes that more precise locations for the occurrence of BS events are unable to obtained. Besides, the performance of segmentation of BS events is easily affected by thresholds of model output in practical applications. To tackle these issues, in this paper, a lightweight DL-based BS segmentation algorithm is proposed. The one-dimensional convolution layers and bidirectional gate recurrent unit (GRU) layers are adopted to enhance the ability of feature extraction. A loss function named shape loss is proposed to reduce the sensitivity of model to thresholds. Moreover, a portable BS monitoring device is developed to realize data acquisition, display and algorithm deployment. Based on this device, experiments on human and rats are conducted to verify the effectiveness of our proposed approach. Experimental results shows that the proposed method outperforms other comparison methods with the higher f1-scores and lower sensitivity to thresholds.

Noisy Label Learning With Provable Consistency for a Wider Family of Losses.

Deep models have achieved state-of-the-art performance on a broad range of visual recognition tasks. Nevertheless, the generalization ability of deep models is seriously affected by noisy labels. Though deep learning packages have different losses, this is not transparent for users to choose consistent losses. This paper addresses the problem of how to use abundant loss functions designed for the traditional classification problem in the presence of label noise. We present a dynamic label learning (DLL) algorithm for noisy label learning and then prove that any surrogate loss function can be used for classification with noisy labels by using our proposed algorithm, with a consistency guarantee that the label noise does not ultimately hinder the search for the optimal classifier of the noise-free sample. In addition, we provide a depth theoretical analysis of our algorithm to verify the justifies' correctness and explain the powerful robustness. Finally, experimental results on synthetic and real datasets confirm the efficiency of our algorithm and the correctness of our justifies and show that our proposed algorithm significantly outperforms or is comparable to current state-of-the-art counterparts.

An Escalated Eavesdropping Attack on Mobile Devices via Low-Resolution Vibration Signals

With the global prevalence of mobile devices, concerns about mobile devices regarding privacy breaches and data leakage are rising. Although sensor permissions are required for mobile applications to access outputs of built-in sensors, motion sensors (e.g., accelerometer and gyroscope) can be visited directly without permission requirement. Extant studies have shown that motion sensors may cause breaches of confidential information, such as passwords, digits, and voice-based commands, but whether it is possible to synthesize intelligible speech waveforms from low-resolution motion sensors has been understudied. In this paper, we present an escalated side-channel attack of built-in speakers by synthesizing intelligible speech waveforms from low-resolution vibration signals. Opposite to traditional classification problems, we formulate this task as a generative problem and introduce an end-to-end synthesis framework dubbed as <i>AccMyrinx</i> to eavesdrop on the speaker via the low-resolution vibration signals. In <i>AccMyrinx</i>, we introduce the data alignment solution to provide the pair-wise voice-vibration sequences and present wavelet-based MelGAN (WMelGAN) with multi-scale time-frequency domain discriminators to generate intelligible acoustic waveforms. We conducted intensive experiments and demonstrated the feasibility of synthesizing the intelligible acoustic signals from low-resolution solid-borne vibration signals. Compared with existing synthesis solutions, our proposed solution outperforms the baselines in both subject and object metrics with the smoothed word error rate of 42.67&#x0025; and the Mel-Cepstral distortion of 0.298. In addition, the quality of synthetic speeches could be impacted by several factors, including gender, speech rate, volume, and sampling frequency.

A Regression-Based Approach to Diabetic Retinopathy Diagnosis Using Efficientnet.

The aim of this study is to develop a computer-assisted solution for the efficient and effective detection of diabetic retinopathy (DR), a complication of diabetes that can damage the retina and cause vision loss if not treated in a timely manner. Manually diagnosing DR through color fundus images requires a skilled clinician to spot lesions, but this can be challenging, especially in areas with a shortage of trained experts. As a result, there is a push to create computer-aided diagnosis systems for DR to help reduce the time it takes to diagnose the condition. The detection of diabetic retinopathy through automation is challenging, but convolutional neural networks (CNNs) play a vital role in achieving success. CNNs have been proven to be more effective in image classification than methods based on handcrafted features. This study proposes a CNN-based approach for the automated detection of DR using Efficientnet-B0 as the backbone network. The authors of this study take a unique approach by viewing the detection of diabetic retinopathy as a regression problem rather than a traditional multi-class classification problem. This is because the severity of DR is often rated on a continuous scale, such as the international clinical diabetic retinopathy (ICDR) scale. This continuous representation provides a more nuanced understanding of the condition, making regression a more suitable approach for DR detection compared to multi-class classification. This approach has several benefits. Firstly, it allows for more fine-grained predictions as the model can assign a value that falls between the traditional discrete labels. Secondly, it allows for better generalization. The model was tested on the APTOS and DDR datasets. The proposed model demonstrated improved efficiency and accuracy in detecting DR compared to traditional methods. This method has the potential to enhance the efficiency and accuracy of DR diagnosis, making it a valuable tool for healthcare professionals. The model has the potential to aid in the rapid and accurate diagnosis of DR, leading to the improved early detection, and management, of the disease.

Traditional Music Regional Classification using Convolutional Neural Network (CNN)

Traditional Indonesian music is an Indonesian cultural heritage that is often forgotten by modern society. Many people do not know which area the traditional music came from. This is a problem because of the large amount of traditional music that loses its identity. Deep Learning technology can be a solution to this traditional music classification problem. The topic of traditional music classification was chosen because there has been no research using this topic before.This research will classify traditional music based on the area of origin using data from Youtube with the extraction method of the Mel-Frequency Cepstral Coefficients (MFCC) feature and the Convolutional Neural Network (CNN) classification model. There are 7 provinces that will be used as classification labels, namely Riau, Papua, Special Capital District of Jakarta, Special Region of Yogyakarta , North Sumatra, West Java, and South Sulawesi.The classification system produced in this study produced good classification accuracy with a value of 74.03%.

Research on Model and Algorithm of Multiview and Multilabel Classification Based on Nearest-Neighbor Model

In the field of machine learning, multilabel learning gradually evolved from the traditional text classification problem. In this study, a multiview multilabel via optimal classifier chain (MVMLOCC) algorithm based on the nearest-neighbor model is proposed. The algorithm model establishes a multilabel chain classifier for each view of the dataset and predicts unknown data samples by dynamically adjusting the weights of chain learners. When an unknown example is input, multiple chain classifier models are multiplied by corresponding weights to obtain the final tag set situation. The model makes full use of the relevance between multiple tags and the complementarity and integrity between multiple perspectives and can achieve better learning results. The final experimental results show that our proposed model can be applied to different multilabel classification tasks and has achieved excellent performance under different evaluation indicators. This study mainly studies the multiview and multilabel classification method and uses active learning technology to solve the problem of high label components in data collection.

Multi-label classification approach for quranic verses labeling

Machine learning involves the task of training systems to be able to make decisions without being explicitly programmed. Important among machine learning tasks is classification involving the process of training machines to make predictions from predefined labels. Classification is broadly categorized into three distinct groups: single-label (SL), multi-class, and multi-label (ML) classification. This research work presents an application of a multi-label classification (MLC) technique in automating Quranic verses labeling. MLC has been gaining attention in recent years. This is due to the increasing amount of works based on real-world classification problems of multi-label data. In traditional classification problems, patterns are associated with a single-label from a set of disjoint labels. However, in MLC, an instance of data is associated with a set of labels. In this paper, three standard &lt;em&gt;MLC&lt;/em&gt; methods: &lt;span&gt;binary relevance (BR), classifier chain (CC), and label powerset (LP) algorithms are implemented with four baseline classifiers: support vector machine (SVM), naïve Bayes (NB), k-nearest neighbors (k-NN), and J48. The research methodology adopts the multi-label problem transformation (PT) approach. The results are validated using six conventional performance metrics. These include: hamming loss, accuracy, one error, micro-F1, macro-F1, and avg. precision. From the results, the classifiers effectively achieved above 70% accuracy mark. Overall, SVM achieved the best results with CC and LP algorithms.&lt;/span&gt;

Target unbiased meta-learning for graph classification

Abstract Even though numerous works focus on the few-shot learning issue by combining meta-learning, there are still limits to traditional graph classification problems. The antecedent algorithms directly extract features from the samples, and do not take into account the preference of the trained model to the previously “seen” targets. In order to overcome the aforementioned issues, an effective strategy with training an unbiased meta-learning algorithm was developed in this paper, which sorted out problems of target preference and few-shot under the meta-learning paradigm. First, the interactive attention extraction module as a supplement to feature extraction was employed, which improved the separability of feature vectors, reduced the preference of the model for a certain target, and remarkably improved the generalization ability of the model on the new task. Second, the graph neural network was used to fully mine the relationship between samples to constitute graph structures and complete image classification tasks at a node level, which greatly enhanced the accuracy of classification. A series of experimental studies were conducted to validate the proposed methodology, where the few-shot and semisupervised learning problem has been effectively solved. It also proved that our model has better accuracy than traditional classification methods on real-world datasets.

Auto-adaptive Grammar-Guided Genetic Programming algorithm to build Ensembles of Multi-Label Classifiers

Multi-label classification has been used to solve a wide range of problems where each example in the dataset may be related either to one class (as in traditional classification problems) or to several class labels at the same time. Many ensemble-based approaches have been proposed in the literature, aiming to improve the performance of traditional multi-label classification algorithms. However, most of them do not consider the data characteristics to build the ensemble, and those that consider them need to tune many parameters to maximize their performance.In this paper, we propose an Auto-adaptive algorithm based on Grammar-Guided Genetic Programming to generate Ensembles of Multi-Label Classifiers based on projections of k labels (AG3P-kEMLC). It creates a tree-shaped ensemble, where each leaf is a multi-label classifier focused on a subset of k labels. Unlike other methods in the literature, our proposal can deal with different values of k in the same ensemble, instead of fixing one specific value. It also includes an auto-adaptive process to reduce the number of hyper-parameters to tune, prevent overfitting and reduce the runtime required to execute it. Three versions of the algorithm are proposed. The first, fixed, uses the same value of k for all multi-label classifiers in the ensemble. The remaining two deal with different k values in the ensemble: uniform gives the same probability to choose each available value of k, and gaussian favors the selection of smaller values of k.The experimental study carried out considering twenty reference datasets and five evaluation metrics, compared with eleven ensemble methods demonstrates that our proposal performs significantly better than the state-of-the-art methods.

Generative adversarial network-based rogue device identification using differential constellation trace figure

With the dramatic development of the internet of things (IoT), security issues such as identity authentication have received serious attention. The radio frequency (RF) fingerprint of IoT device is an inherent feature, which can hardly be imitated. In this paper, we propose a rogue device identification technique via RF fingerprinting using deep learning-based generative adversarial network (GAN). Being different from traditional classification problems in RF fingerprint identifications, this work focuses on unknown accessing device recognition without prior information. A differential constellation trace figure generation process is initially employed to transform RF fingerprint features from time-domain waveforms to two-dimensional figures. Then, by using GAN, which is a kind of unsupervised learning algorithm, we can discriminate rogue devices without any prior information. An experimental verification system is built with 54 ZigBee devices regarded as recognized devices and accessing devices. A universal software radio peripheral receiver is used to capture the signal and identify the accessing devices. Experimental results show that the proposed rogue device identification method can achieve 95% identification accuracy in a real environment.

Data quality measures based on granular computing for multi-label classification

Rough set theory is a granular computing formalism that allows analyzing a given dataset through well-defined measures. Some of these measures aim to characterize datasets used to discover knowledge, mostly in traditional classification problems. Measuring the data quality is pivotal to estimate beforehand the problem’s difficulty since a classification model’s accuracy heavily depends on the data quality. However, to the best of our knowledge, there are no measures devoted to analyzing the quality of multi-label datasets. In this paper, we propose six data quality measures for multi-label problems, which are based on different granular approaches. Some of these measures redefine the decision class concept, while others redefine the consistency concept. Moreover, we study the impact of the similarity threshold parameters and the distance functions on the behavior of these measures. The numerical simulations show a statistical correlation between the measures that redefine the consistency concept and the performance of the ML-kNN classifier.

Multi-View Attention-Guided Multiple Instance Detection Network for Interpretable Breast Cancer Histopathological Image Diagnosis

Deep learning approaches have demonstrated significant progress in breast cancer histopathological image diagnosis. Training an interpretable diagnosis model using high-resolution histopathological image is still challenging. To alleviate this problem, a novel multi-view attention-guided multiple instance detection network (MA-MIDN) is proposed. The traditional image classification problem is framed as a weakly supervised multiple instance learning (MIL) problem. We first divide each histopathology image into instances and form a corresponding bag to fully utilize high-resolution information through MIL. Then a new multiple-view attention (MVA) algorithm is proposed to learn attention on the instances from the image to localize the lesion regions in this image. A MVA-guided MIL pooling strategy is designed for aggregating instance-level features to obtain bag-level features for the final classification. The proposed MA-MIDN model performs lesion localization and image classification, simultaneously. Particularly, we train the MA-MIDN model under the deep mutual learning (DML) schema. This transfers DML to a weakly supervised learning problem. Three public breast cancer histopathological image datasets are chosen to evaluate classification and localization results. The experimental results demonstrate that the MA-MIDN model is superior to the latest baselines in terms of diagnosis accuracy, AUC, Precision, Recall, and F1. Notably, it achieves better localization results without compromising classification performance, thereby proving its higher practicality. The code for the MA-MIDN model is available at https://github.com/lcxlcx/MA-MIDN.

Multivariate and Online Prediction of Closing Price Using Kernel Adaptive Filtering.

This paper proposes a multivariate and online prediction of stock prices via the paradigm of kernel adaptive filtering (KAF). The prediction of stock prices in traditional classification and regression problems needs independent and batch-oriented nature of training. In this article, we challenge this existing notion of the literature and propose an online kernel adaptive filtering-based approach to predict stock prices. We experiment with ten different KAF algorithms to analyze stocks' performance and show the efficacy of the work presented here. In addition to this, and in contrast to the current literature, we look at granular level data. The experiments are performed with quotes gathered at the window of one minute, five minutes, ten minutes, fifteen minutes, twenty minutes, thirty minutes, one hour, and one day. These time windows represent some of the common windows frequently used by traders. The proposed framework is tested on 50 different stocks making up the Indian stock index: Nifty-50. The experimental results show that online learning and KAF is not only a good option, but practically speaking, they can be deployed in high-frequency trading as well.

ZJU-Leaper: A Benchmark Dataset for Fabric Defect Detection and a Comparative Study

Fabric inspection plays an important role in the process of quality control in textile manufacturing. There is a growing demand in the textile industry to leverage computer vision technology for more efficient quality control in the hope that it will replace the traditional labor-intensive inspection by naked eyes. However, there is an underlying viewpoint in most existing fabric datasets that automatic defect detection is a traditional image classification problem, thus more samples help better, which lacks enough consideration about the problem itself and real application environments. After deep communication with users, we find these facts that an assembly line usually has only a few fixed texture fabrics for a long period, users prefer fast deployment and easily upgradable model to a general model and long-time tuning, and users hope the process of collecting samples, annotating, and deployment affects assembly lines as little as possible. This implies that defect detection is different from popular deep learning problems. Multiple-stage models and fast training become more attractive since users are able to train and deploy models by themselves according to the real conditions of samples that can be obtained. Based on this analysis, we propose a new fabric dataset “ZJU-Leaper”. It provides a series of task settings in accordance with the progressive strategy dealing with the problem, from only normal samples to many defective samples with precise annotations, to facilitate real-world application. To avoid misleading information and inconsistency issues associated with the prior evaluation metrics, we propose a new evaluation protocol by experimental analysis of task-specific indexes, which can tell a truthful comparison between different inspection methods. We also offer some novel solutions to address the new challenges of our dataset, as part of the baseline experiments. It is our hope that ZJU-Leaper can accelerate the research of automated visual inspection and empower the practitioners with more opportunities for manufacturing automation in the textile industry. <p xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><i>Impact Statement</i>—Automatic defect inspection is very important in quality control of the fabric industry by helping manufacturers to identify production problems early, hence improving product quality and production efficiency. Meanwhile, it is able to reduce the high labor cost of manual inspection and boost the productivity of the textile industry. To develop effective mathematical inspection algorithms, the fabric dataset serves as an indispensable component to present a practical application environment and enable fair evaluation for algorithms. This paper proposes a new dataset, called “ZJU-Leaper” designed from a viewpoint of multiple-stage models and fast training, containing threefold novelty: 1) the data collection and organization consider the actual requirements and special characteristics of assembly lines in textile factories; 2) it has several designed task settings in order to meet the different levels of requirements in the practical inspection task; 3) it provides a reasonable evaluation protocol for comprehensive comparisons between different inspection algorithms. The preliminary experiments show that some existing algorithms still cannot reach the satisfying performance by this benchmark, which implies more effort should be made to develop new methods for the real use of automatic defect inspection.

Text classification based on graph convolutional network with attention

The paper proposes a text classification method based on the graph convolutional network, the traditional text classification problem can be transformed into the node classification problem in the graph, and then proposes a attention structure to increase the weight of certain words and phrases. This paper verifies that the above structure has a positive effect on text classification accuracy. In particular, the structure can handle irregular data, such as citation network. We obtain very competitive results on 5 commonly used text classification datasets and achieve state-of-the-art results on 4 datasets, which include one dataset of citation network. Experiments show that combination of both structures can significantly reduce the classification error.

Vegetation extraction in the field using multi-level features

Accurate and automatic vegetation extraction from digital plant images in the field is a widely studied topic in precision agriculture. Many techniques focus on pixels or regions to be segmented as plants or back-grounds, such as colour index-based and learning-based methods. Different from a traditional two-class classification problem, the proposed method regarded vegetation extraction as a multi-class task. In consideration of manually annotated errors at the edge of a plant image, the original marked mask was re-labelled using a Gaussian probability function. To capture more adequate information in the process of feature extraction, 9 pixel-level colour features and 18 region-level statistical characteristics of neighbourhood pixels were computed from three colour spaces. The extracted 27-dimensional features were inputs of a classification model, which output multi-class labels. A suitable threshold was finally selected to obtain the segmented image. Experimental results showed that the proposed multi-class and multi-level features (MCMLF) method achieved better performance than the other approaches. Through the quantitative and qualitative analysis of segmentation results, it was also found that the suggested method had high computation efficiency as well as strong adaptation ability to solve the outdoor challenges, including various lighting conditions, shadow regions, and complex backgrounds. • Vegetation extraction is regarded as a multi-class recognition task. • Pixel-level and region-level features are extracted from three colour spaces. • The method has high computation efficiency as well as strong adaptation ability.

Hierarchical Transformer Network for Utterance-Level Emotion Recognition

While there have been significant advances in detecting emotions in text, in the field of utterance-level emotion recognition (ULER), there are still many problems to be solved. In this paper, we address some challenges in ULER in dialog systems. (1) The same utterance can deliver different emotions when it is in different contexts. (2) Long-range contextual information is hard to effectively capture. (3) Unlike the traditional text classification problem, for most datasets of this task, they contain inadequate conversations or speech. (4) To better model the emotional interaction between speakers, speaker information is necessary. To address the problems of (1) and (2), we propose a hierarchical transformer framework (apart from the description of other studies, the “transformer” in this paper usually refers to the encoder part of the transformer) with a lower-level transformer to model the word-level input and an upper-level transformer to capture the context of utterance-level embeddings. For problem (3), we use bidirectional encoder representations from transformers (BERT), a pretrained language model, as the lower-level transformer, which is equivalent to introducing external data into the model and solves the problem of data shortage to some extent. For problem (4), we add speaker embeddings to the model for the first time, which enables our model to capture the interaction between speakers. Experiments on three dialog emotion datasets, Friends, EmotionPush, and EmoryNLP, demonstrate that our proposed hierarchical transformer network models obtain competitive results compared with the state-of-the-art methods in terms of the macro-averaged F1-score (macro-F1).

Sentiment classification with adversarial learning and attention mechanism

AbstractSentiment classification is a key task in sentiment analysis, reviews mining, and other text mining applications. Various models have been proposed to build sentiment classifiers, but the classification performances of some existing methods are not good enough. Meanwhile, as a subproblem of sentiment classification, positive and unlabeled learning (PU learning) problem widely exists in real‐world cases, but it has not been given enough attention. In this article, we aim to solve the two problems in one framework. We first build a model for traditional sentiment classification based on adversarial learning, attention mechanism, and long short‐term memory (LSTM) network. We further propose an enhanced adversarial learning method to tackle PU learning problem. We conducted extensive experiments in three real‐world datasets. The experimental results demonstrate that our models outperform the compared methods in both traditional sentiment classification problem and PU learning problem. Furthermore, we study the effect of our models on word embedding. Finally, we report and discuss the sensitivity of our models to parameters.