A Survey on Deep Learning Pouyanfar, Samira; Sadiq, Saad; Yan, Yilin ...
ACM computing surveys,
09/2019, Letnik:
51, Številka:
5
Journal Article
Recenzirano
The field of machine learning is witnessing its golden era as deep learning slowly becomes the leader in this domain. Deep learning uses multiple layers to represent the abstractions of data to build ...computational models. Some key enabler deep learning algorithms such as generative adversarial networks, convolutional neural networks, and model transfers have completely changed our perception of information processing. However, there exists an aperture of understanding behind this tremendously fast-paced domain, because it was never previously represented from a multiscope perspective. The lack of core understanding renders these powerful methods as black-box machines that inhibit development at a fundamental level. Moreover, deep learning has repeatedly been perceived as a silver bullet to all stumbling blocks in machine learning, which is far from the truth. This article presents a comprehensive review of historical and recent state-of-the-art approaches in visual, audio, and text processing; social network analysis; and natural language processing, followed by the in-depth analysis on pivoting and groundbreaking advances in deep learning applications. It was also undertaken to review the issues faced in deep learning such as unsupervised learning, black-box models, and online learning and to illustrate how these challenges can be transformed into prolific future research avenues.
The explosive growth and widespread accessibility of digital health data have led to a surge of research activity in the healthcare and data sciences fields. The conventional approaches for health ...data management have achieved limited success as they are incapable of handling the huge amount of complex data with high volume, high velocity, and high variety. This article presents a comprehensive overview of the existing challenges, techniques, and future directions for computational health informatics in the big data age, with a structured analysis of the historical and state-of-the-art methods. We have summarized the challenges into four Vs (i.e., volume, velocity, variety, and veracity) and proposed a systematic data-processing pipeline for generic big data in health informatics, covering data capturing, storing, sharing, analyzing, searching, and decision support. Specifically, numerous techniques and algorithms in machine learning are categorized and compared. On the basis of this material, we identify and discuss the essential prospects lying ahead for computational health informatics in this big data age.
With the proliferation of online services and mobile technologies, the world has stepped into a multimedia big data era. A vast amount of research work has been done in the multimedia area, targeting ...different aspects of big data analytics, such as the capture, storage, indexing, mining, and retrieval of multimedia big data. However, very few research work provides a complete survey of the whole pine-line of the multimedia big data analytics, including the management and analysis of the large amount of data, the challenges and opportunities, and the promising research directions. To serve this purpose, we present this survey, which conducts a comprehensive overview of the state-of-the-art research work on multimedia big data analytics. It also aims to bridge the gap between multimedia challenges and big data solutions by providing the current big data frameworks, their applications in multimedia analyses, the strengths and limitations of the existing methods, and the potential future directions in multimedia big data analytics. To the best of our knowledge, this is the first survey that targets the most recent multimedia management techniques for very large-scale data and also provides the research studies and technologies advancing the multimedia analyses in this big data era.
Casting concrete at different ages for new construction and repairing or retrofitting concrete structures requires a sufficient bond between concrete casts. The bond strength between different casts ...is attributed to surface roughness. Surface roughness can be achieved in many ways, such as water‐jetting or sandblasting. To evaluate the degree of surface roughness, qualitative and quantitative methods are introduced by many researchers; however, several drawbacks are associated with most of these methods, including cost, availability, human errors, and inability to assess old structures from prior inspection records. Two novel industrial implementation methods are introduced in this paper to estimate, quantitatively, the concrete surface roughness from images with sufficient resolution. In the first application method, a digital image processing method is proposed to distinguish the coarse aggregate from cement paste, and a new index is presented as a function of aggregate proportional area to the surface area. In the second application method, data augmentation and transfer learning techniques in computer vision and machine learning are utilized to classify new images based on predefined images during the learning process. Both application methods were related to a well‐established method of 3D laser scanning from sandblasted concrete surfaces. Finally, a brand new set of images of sandblasted surfaces was used to test and validate both methods. The results show that both methods successfully estimate the concrete surface roughness with an accuracy of more than 93%.
Real-world applications usually encounter data with various modalities, each containing valuable information. To enhance these applications, it is essential to effectively analyze all information ...extracted from different data modalities, while most existing learning models ignore some data types and only focus on a single modality. This paper presents a new multimodal deep learning framework for event detection from videos by leveraging recent advances in deep neural networks. First, several deep learning models are utilized to extract useful information from multiple modalities. Among these are pre-trained Convolutional Neural Networks (CNNs) for visual and audio feature extraction and a word embedding model for textual analysis. Then, a novel fusion technique is proposed that integrates different data representations in two levels, namely frame-level and video-level. Different from the existing multimodal learning algorithms, the proposed framework can reason about a missing data type using other available data modalities. The proposed framework is applied to a new video dataset containing natural disaster classes. The experimental results illustrate the effectiveness of the proposed framework compared to some single modal deep learning models as well as conventional fusion techniques. Specifically, the final accuracy is improved more than 16% and 7% compared to the best results from single modality and fusion models, respectively.
From the start, the airline industry has remarkably connected countries all over the world through rapid long-distance transportation, helping people overcome geographic barriers. Consequently, this ...has ushered in substantial economic growth, both nationally and internationally. The airline industry produces vast amounts of data, capturing a diverse set of information about their operations, including data related to passengers, freight, flights, and much more. Analyzing air travel data can advance the understanding of airline market dynamics, allowing companies to provide customized, efficient, and safe transportation services. Due to big data challenges in such a complex environment, the benefits of drawing insights from the air travel data in the airline industry have not yet been fully explored. This article aims to survey various components and corresponding proposed data analysis methodologies that have been identified as essential to the inner workings of the airline industry. We introduce existing data sources commonly used in the papers surveyed and summarize their availability. Finally, we discuss several potential research directions to better harness airline data in the future. We anticipate this study to be used as a comprehensive reference for both members of the airline industry and academic scholars with an interest in airline research.
The fast and explosive growth of digital data in social media and World Wide Web has led to numerous opportunities and research activities in multimedia big data. Among them, disaster management ...applications have attracted a lot of attention in recent years due to its impacts on society and government. This study targets content analysis and mining for disaster management. Specifically, a multimedia big data framework based on the advanced deep learning techniques is proposed. First, a video dataset of natural disasters is collected from YouTube. Then, two separate deep networks including a temporal audio model and a spatio-temporal visual model are presented to analyze the audio-visual modalities in video clips effectively. Thereafter, the results of both models are integrated using the proposed fusion model based on the Multiple Correspondence Analysis (MCA) algorithm which considers the correlations between data modalities and final classes. The proposed multimodal framework is evaluated on the collected disaster dataset and compared with several state-of-the-art single modality and fusion techniques. The results demonstrate the effectiveness of both visual model and fusion model compared to the baseline approaches. Specifically, the accuracy of the final multi-class classification using the proposed MCA-based fusion reaches to 73% on this challenging dataset.
With the proliferation of online services and mobile technologies, the world has stepped into a multimedia big data era, where new opportunities and challenges appear with the high diversity ...multimedia data together with the huge amount of social data. Nowadays, multimedia data consisting of audio, text, image, and video has grown tremendously. With such an increase in the amount of multimedia data, the main question raised is how one can analyze this high volume and variety of data in an efficient and effective way. A vast amount of research work has been done in the multimedia area, targeting different aspects of big data analytics, such as the capture, storage, indexing, mining, and retrieval of multimedia big data. However, there is insufficient research that provides a comprehensive framework for multimedia big data analytics and management. To address the major challenges in this area, a new framework is proposed based on deep neural networks for multimedia semantic concept detection with a focus on spatio-temporal information analysis and rare event detection. The proposed framework is able to discover the pattern and knowledge of multimedia data using both static deep data representation and temporal semantics. Specifically, it is designed to handle data with skewed distributions. The proposed framework includes the following components: (1) a synthetic data generation component based on simulation and adversarial networks for data augmentation and deep learning training, (2) an automatic sampling model to overcome the imbalanced data issue in multimedia data, (3) a deep representation learning model leveraging novel deep learning techniques to generate the most discriminative static features from multimedia data, (4) an automatic hyper-parameter learning component for faster training and convergence of the learning models, (5) a spatio-temporal deep learning model to analyze dynamic features from multimedia data, and finally (6) a multimodal deep learning fusion model to integrate different data modalities. The whole framework has been evaluated using various large-scale multimedia datasets that include the newly collected disaster-events video dataset and other public datasets.
Multimedia concept detection is a challenging topic due to the well-known class imbalance issue, where the data instances are distributed unevenly across different classes. This problem becomes even ...more prominent when the minority class that contains an extremely small proportion of the data represents the concept of interest as has occurred in many real-world applications such as frauds in banking transactions and goal events in soccer videos. Traditional data mining approaches often have difficulty handling largely skewed data distributions. To address this issue, in this paper, an importance-factor (IF)-based multiple correspondence analysis (MCA) framework is proposed to deal with the imbalanced datasets. Specifically, a hierarchical information gain analysis method, which is inspired by the decision tree algorithm, is presented for critical feature selection and IF assignment. Then, the derived IF is incorporated with the MCA algorithm for effective concept detection and retrieval. The comparison results in video concept detection using the disaster dataset and the soccer dataset demonstrate the effectiveness of the proposed framework.