Intensive Feature Engineering Research Articles

A novel deep learning-based approach for malware detection

Malware detection approaches can be classified into two classes, including static analysis and dynamic analysis. Conventional approaches of the two classes have their respective advantages and disadvantages. For example, static analysis is faster but cannot detect the malware variants generated through code obfuscation, whereas dynamic analysis can effectively detect variants generated through code obfuscation but is slower and requires intensive resources. This paper proposes a novel deep learning-based approach for malware detection. It delivers better performance than conventional approaches by combining static and dynamic analysis advantages. First, it visualises a portable executable (PE) file as a coloured image. Second, it extracts deep features from the colour image using fine-tuned deep learning model. Third, it detects malware based on the deep features using support vector machines (SVM). The proposed method combines deep learning with machine learning and eliminates the need for intensive feature engineering tasks and domain knowledge. The proposed approach is scalable, cost-effective, and efficient. The detection effectiveness of the proposed method is validated through 12 machine learning models and 15 deep learning models. The generalisability of the proposed framework is validated on various benchmark datasets. The proposed approach outperformed with an accuracy of 99.06% on the Malimg dataset. The Wilcoxon signed-rank test is used to show the statistical significance of the proposed framework. The detailed experimental results demonstrate the superiority of the proposed method over the other state-of-the-art approaches, with an average increase in accuracy of 16.56%. Finally, to tackle the problems of imbalanced data and the shortage of publicly available datasets for malware detection, various data augmentation techniques are proposed, which lead to improved performance. It is evident from the results that the proposed framework can be useful to the defence industry, which will be helpful in devising more efficient malware detection solutions.

Multi-view learning with distinguishable feature fusion for rumor detection

Researchers, enterprises, and governments have made great efforts to detect misinformation promptly and accurately. Traditional solutions either examine complicated hand-crafted features or rely heavily on the constructed credibility networks to extract useful indicators for discerning false information. However, such approaches require insightful domain expert knowledge and intensive feature engineering that are often non-generalizable. Recent advances in deep learning techniques have spurred learning high-level representations from textual and image content and discovering diffusion patterns with various neural networks. Despite the progress made by these methods, they still face the problem of overdependence on the content features and fail to discriminate against the influence of each user involved in the process of rumor spreading. Different user-aspect information plays different roles in various stages of rumor diffusion, effectively extract features from each aspect, and aggregate the learned features into a unique representation, which has not been well investigated. To address these limitations, we propose a novel model, UMLARD (User-aspect Multi-view Learning with Attention for Rumor Detection), to effectively learn the representation of different views of the users who engaged in spreading the tweet, and fuse the learned features through the distinguishable fusion mechanism. Finally, we concatenate the learned user-aspect features with content features to form a unique representation and feed it into a fully connected layer to predict the label of rumors. Our experiments conducted on real-world datasets demonstrate that UMLARD significantly improves the rumor detection performance compared to state-of-the-art baselines. It also allows explainability of the model behavior and the predicted results.

Discovering New Intents via Constrained Deep Adaptive Clustering with Cluster Refinement

Identifying new user intents is an essential task in the dialogue system. However, it is hard to get satisfying clustering results since the definition of intents is strongly guided by prior knowledge. Existing methods incorporate prior knowledge by intensive feature engineering, which not only leads to overfitting but also makes it sensitive to the number of clusters. In this paper, we propose constrained deep adaptive clustering with cluster refinement (CDAC+), an end-to-end clustering method that can naturally incorporate pairwise constraints as prior knowledge to guide the clustering process. Moreover, we refine the clusters by forcing the model to learn from the high confidence assignments. After eliminating low confidence assignments, our approach is surprisingly insensitive to the number of clusters. Experimental results on the three benchmark datasets show that our method can yield significant improvements over strong baselines. 1

Constrained Self-Supervised Clustering for Discovering New Intents (Student Abstract)

Discovering new user intents is an emerging task in the dialogue system. In this paper, we propose a self-supervised clustering method that can naturally incorporate pairwise constraints as prior knowledge to guide the clustering process and does not require intensive feature engineering. Extensive experiments on three benchmark datasets show that our method can yield significant improvements over strong baselines.

Clinical Inference From Cardiovascular Imaging: Paradigm Shift Towards Machine-Based Intelligent Platform

Diagnostic imaging plays an indispensable role in the modern management of cardiovascular diseases. Despite significant advances in imaging modalities, the workflow in clinical imaging requires a great deal of human input. With increasing demands in patient care, current workflows in cardiovascular imaging face a number of challenges which include insufficient number of specialists, time constraints, learning curves, and inter-observer variability. Automated techniques using revolutionary approaches with machine learning can fundamentally alter our interpretation process and produce smarter, faster, and efficient results. With the emergence of big data and deep learning techniques which enable direct inference of image data without intensive feature engineering or can process raw images directly. Machine learning has been rapidly adopted in the field of medicine and has shown its fast and reproducible automatic inference in various cardiovascular imaging modalities, including echocardiography, nuclear imaging, cardiac computed tomography, and cardiac magnetic resonance. While deep domain knowledge of cardiovascular disease remains the cornerstone throughout the process, physicians may be relieved from mundane tasks using automation and be able to spend more time on decision-making and communication with patients and other medical staff in the new paradigm brought by machine learning. In this review article, we explore the role of machine in the clinical inference from cardiac imaging data recently.

Deep Neural Network Based Predictions of Protein Interactions Using Primary Sequences.

Machine learning based predictions of protein–protein interactions (PPIs) could provide valuable insights into protein functions, disease occurrence, and therapy design on a large scale. The intensive feature engineering in most of these methods makes the prediction task more tedious and trivial. The emerging deep learning technology enabling automatic feature engineering is gaining great success in various fields. However, the over-fitting and generalization of its models are not yet well investigated in most scenarios. Here, we present a deep neural network framework (DNN-PPI) for predicting PPIs using features learned automatically only from protein primary sequences. Within the framework, the sequences of two interacting proteins are sequentially fed into the encoding, embedding, convolution neural network (CNN), and long short-term memory (LSTM) neural network layers. Then, a concatenated vector of the two outputs from the previous layer is wired as the input of the fully connected neural network. Finally, the Adam optimizer is applied to learn the network weights in a back-propagation fashion. The different types of features, including semantic associations between amino acids, position-related sequence segments (motif), and their long- and short-term dependencies, are captured in the embedding, CNN and LSTM layers, respectively. When the model was trained on Pan’s human PPI dataset, it achieved a prediction accuracy of 98.78% at the Matthew’s correlation coefficient (MCC) of 97.57%. The prediction accuracies for six external datasets ranged from 92.80% to 97.89%, making them superior to those achieved with previous methods. When performed on Escherichia coli, Drosophila, and Caenorhabditis elegans datasets, DNN-PPI obtained prediction accuracies of 95.949%, 98.389%, and 98.669%, respectively. The performances in cross-species testing among the four species above coincided in their evolutionary distances. However, when testing Mus Musculus using the models from those species, they all obtained prediction accuracies of over 92.43%, which is difficult to achieve and worthy of note for further study. These results suggest that DNN-PPI has remarkable generalization and is a promising tool for identifying protein interactions.

Deep learning approaches for mining structure-property linkages in high contrast composites from simulation datasets

Data-driven methods are emerging as an important toolset in the studies of multiscale, multiphysics, materials phenomena. More specifically, data mining and machine learning methods offer an efficient toolset for extracting and curating the important correlations controlling these multiscale materials phenomena in high-value reduced-order forms called process-structure-property (PSP) linkages. Traditional machine learning methods usually depend on intensive feature engineering, and have enjoyed some success in establishing the desired PSP linkages. In contrast, deep learning approaches provide a feature-engineering-free framework with high learning capability. In this work, a deep learning approach is designed and implemented to model an elastic homogenization structure-property linkage in a high contrast composite material system. More specifically, the proposed deep learning model is employed to capture the nonlinear mapping between the three-dimensional material microstructure and its macroscale (effective) stiffness. It is demonstrated that this end-to-end framework can predict the effective stiffness of high contrast elastic composites with a wide of range of microstructures, while exhibiting high accuracy and low computational cost for new evaluations.

Identifying sleep spindles with multichannel EEG and classification optimization

Researchers classify critical neural events during sleep called spindles that are related to memory consolidation using the method of scalp electroencephalography (EEG). Manual classification is time consuming and is susceptible to low inter-rater agreement. This could be improved using an automated approach. This study presents an optimized filter based and thresholding (FBT) model to set up a baseline for comparison to evaluate machine learning models using naïve features, such as raw signals, peak frequency, and dominant power. The FBT model allows us to formally define sleep spindles using signal processing but may miss examples most human scorers would agree are spindles. Machine learning methods in theory should be able to approach performance of human raters but they require a large quantity of scored data, proper feature representation, intensive feature engineering, and model selection. We evaluate both the FBT model and machine learning models with naïve features. We show that the machine learning models derived from the FBT model improve classification performance. An automated approach designed for the current data was applied to the DREAMS dataset [1]. With one of the expert's annotation as a gold standard, our pipeline yields an excellent sensitivity that is close to a second expert's scores and with the advantage that it can classify spindles based on multiple channels if more channels are available. More importantly, our pipeline could be modified as a guide to aid manual annotation of sleep spindles based on multiple channels quickly (6–10 s for processing a 40-min EEG recording), making spindle detection faster and more objective.