Big Cross-Modal Social Media Data Analytics With Deep Intelligence
TL;DR: The thirteen papers in this special section focus on social media data analytics with deep intelligence, which aims to handle data sampling from multimodal deep spaces, so as to well characterize the big data.
Abstract: The thirteen papers in this special section focus on social media data analytics with deep intelligence. Big cross-model social media data analytics with deep intelligence aims to handle data sampling from multimodal deep spaces, so as to well characterize the big data. The addressed topic span from the range of human action recognition to affective computing, disaster detection, classification, retrieval, clustering, vehicle reidentification, and data security.
Citations
More filters
TL;DR: A compression-then-encryption-based dual watermarking to protect the EPR data for the healthcare system, which produces several significant features and offers better performance in terms of robustness and security.
Abstract: The smart healthcare system is an electronic patient records (EPR) sharing system, which significantly helps sharing of EPR data and provides appropriate medical assistance for the patients and a more suitable platform for the potential researchers. However, the security of EPR data is still a major issue in such systems. In this paper, we develop a compression-then-encryption-based dual watermarking to protect the EPR data for the healthcare system, which produces several significant features. Experiments conducted on a large set of medical data indicate the capability of our proposed method for smart healthcare. Finally, when compared with the existing technique, the proposed work offers better performance in terms of robustness and security.
57Â citations
TL;DR: An Android malicious application detection framework termed multiview information integration technology (MVIIDroid), which has superior classification performances when separating malware from benign applications and its ability to attribute malicious applications to their actual families is evaluated.
Abstract: With the rapid growth of Android applications, there is an urgent need for powerful Android malware detection technology nowadays. Existing classification models can be summarized with the following two steps-feature extraction and classification model learning. To further enhance the representation ability of existing classification models, this article presents an Android malicious application detection framework termed multiview information integration technology (MVIIDroid). To be specific, in our approach, we extract applications’ multiple components, transform them into embedding feature vectors and train a multiple Kernel learning model as the classifier. To illustrate the effectiveness of our model, we evaluate MVIIDroid on two Android malware datasets of 6820 malware and 6820 benign applications. Results show that we have superior classification performances when separating malware from benign applications. Moreover, we further evaluate MVIIDroid's ability to attribute malicious applications to their actual families. The experimental results well demonstrate the effectiveness of the proposed model.
12Â citations
TL;DR: The continuous emotional symptoms of three kinds of mental disorders are accurately and quantitatively described by the newly introduced interpretable psychological computing model and the relationship between two complex emotions and the basic emotions is established, breaking through the cognitive limitations of the traditional psychology field.
Abstract: With the current global outbreak of COVID-19, an increasing number of people are suffering from negative mental states and mental disorders. We propose a multimodal psychological computational technology in a universal environment. We establish a mental health database following a naturalistic paradigm as well as a long-term ubiquitous interpretable psychological computing model based on prior knowledge and multimodal information fusion. The proposed model achieves state-of-the-art accuracy in both basic and complex emotion detection on the proposed mental health database and effectively solves scientific and accuracy-related problems in long-term complex mental health status recognition and prediction. Regarding psychology and the medicine of mental disorders, we identify the continuous emotional symptoms of three kinds of mental disorders, which have not previously been accurately observed based on multimodal big data. They are accurately and quantitatively described by the newly introduced interpretable psychological computing model. At the same time, we establish the relationship between two complex emotions and the basic emotions, breaking through the cognitive limitations of the traditional psychology field.
7Â citations
TL;DR: A novel clustering model is proposed, which exploits Pearson Correlation Coefficient to auto-balance the optimal equation according to the category distribution, and is extended to the incremental model, which only learns samples at the current frame to update the original model.
Abstract: FU-PCM is an effective clustering mode by inducing the regularization constraint in C-means to avoid the interference of noises and outliers. However, it is difficult to obtain the satisfactory performance in the context of imbalance category distribution. To address above problem, first, we propose a novel clustering model, which exploits Pearson Correlation Coefficient to auto-balance the optimal equation according to the category distribution. Then, we extend it to the incremental model, which only learns samples at the current frame to update the original model by mapping centers of two adjacent frames to the distinguishable space, and mining hard centers to recognize new categories in online datasets. In this article, offline and online methods are verified in popular datasets, and experimental results demonstrate the effectiveness and efficiency of the proposed models.
2Â citations
07 Jan 2022
TL;DR: This work modifies CABAC using the double bit range estimation in the VVC standard and considering range updates for probability predictions, and improved results show that it provides more significant gains in RA and LD, which is better than the AI configuration.
Abstract: This work modifies CABAC using the double bit range estimation in the VVC standard and considering range updates for probability predictions. The inclusion of the arithmetic coding engine with multi-hypothesis probability estimation, and their consideration of the context modelling of entropy coding at the transform coefficient levels. In addition to describing the implementation details, this work also discusses the hardware implementation, which is based on simple operations such as bitwise operations and single subsampling for subinterval updates. The experimental results show that these improvements offer some gain in rate-distortion efficiency while incurring a controlled and adjustable complexity overhead. The improved results show that it provides more significant gains in RA and LD, which is better than the AI configuration.
1Â citations
References
More filters
TL;DR: A compression-then-encryption-based dual watermarking to protect the EPR data for the healthcare system, which produces several significant features and offers better performance in terms of robustness and security.
Abstract: The smart healthcare system is an electronic patient records (EPR) sharing system, which significantly helps sharing of EPR data and provides appropriate medical assistance for the patients and a more suitable platform for the potential researchers. However, the security of EPR data is still a major issue in such systems. In this paper, we develop a compression-then-encryption-based dual watermarking to protect the EPR data for the healthcare system, which produces several significant features. Experiments conducted on a large set of medical data indicate the capability of our proposed method for smart healthcare. Finally, when compared with the existing technique, the proposed work offers better performance in terms of robustness and security.
57Â citations
TL;DR: An Android malicious application detection framework termed multiview information integration technology (MVIIDroid), which has superior classification performances when separating malware from benign applications and its ability to attribute malicious applications to their actual families is evaluated.
Abstract: With the rapid growth of Android applications, there is an urgent need for powerful Android malware detection technology nowadays. Existing classification models can be summarized with the following two steps-feature extraction and classification model learning. To further enhance the representation ability of existing classification models, this article presents an Android malicious application detection framework termed multiview information integration technology (MVIIDroid). To be specific, in our approach, we extract applications’ multiple components, transform them into embedding feature vectors and train a multiple Kernel learning model as the classifier. To illustrate the effectiveness of our model, we evaluate MVIIDroid on two Android malware datasets of 6820 malware and 6820 benign applications. Results show that we have superior classification performances when separating malware from benign applications. Moreover, we further evaluate MVIIDroid's ability to attribute malicious applications to their actual families. The experimental results well demonstrate the effectiveness of the proposed model.
12Â citations
TL;DR: The continuous emotional symptoms of three kinds of mental disorders are accurately and quantitatively described by the newly introduced interpretable psychological computing model and the relationship between two complex emotions and the basic emotions is established, breaking through the cognitive limitations of the traditional psychology field.
Abstract: With the current global outbreak of COVID-19, an increasing number of people are suffering from negative mental states and mental disorders. We propose a multimodal psychological computational technology in a universal environment. We establish a mental health database following a naturalistic paradigm as well as a long-term ubiquitous interpretable psychological computing model based on prior knowledge and multimodal information fusion. The proposed model achieves state-of-the-art accuracy in both basic and complex emotion detection on the proposed mental health database and effectively solves scientific and accuracy-related problems in long-term complex mental health status recognition and prediction. Regarding psychology and the medicine of mental disorders, we identify the continuous emotional symptoms of three kinds of mental disorders, which have not previously been accurately observed based on multimodal big data. They are accurately and quantitatively described by the newly introduced interpretable psychological computing model. At the same time, we establish the relationship between two complex emotions and the basic emotions, breaking through the cognitive limitations of the traditional psychology field.
7Â citations
TL;DR: A novel clustering model is proposed, which exploits Pearson Correlation Coefficient to auto-balance the optimal equation according to the category distribution, and is extended to the incremental model, which only learns samples at the current frame to update the original model.
Abstract: FU-PCM is an effective clustering mode by inducing the regularization constraint in C-means to avoid the interference of noises and outliers. However, it is difficult to obtain the satisfactory performance in the context of imbalance category distribution. To address above problem, first, we propose a novel clustering model, which exploits Pearson Correlation Coefficient to auto-balance the optimal equation according to the category distribution. Then, we extend it to the incremental model, which only learns samples at the current frame to update the original model by mapping centers of two adjacent frames to the distinguishable space, and mining hard centers to recognize new categories in online datasets. In this article, offline and online methods are verified in popular datasets, and experimental results demonstrate the effectiveness and efficiency of the proposed models.
2Â citations