Computational Intelligence最新文献

Electroencephalogram (EEG), as a tool capable of objectively recording brain electrical signals during emotional expression, has been extensively utilized. Current technology heavily relies on datasets, with its performance being limited by the size of the dataset and the accuracy of its annotations. At the same time, unsupervised learning and contrastive learning methods largely depend on the feature distribution within datasets, thus requiring training tailored to specific datasets for optimal results. However, the collection of EEG signals is influenced by factors such as equipment, settings, individuals, and experimental procedures, resulting in significant variability. Consequently, the effectiveness of models is heavily dependent on dataset collection efforts conducted under stringent objective conditions. To address these challenges, we introduce a novel approach: employing a self-supervised pre-training model, to process data across different datasets. This model is capable of operating effectively across multiple datasets. The model conducts self-supervised pre-training without the need for direct access to specific emotion category labels, enabling it to pre-train and extract universally useful features without predefined downstream tasks. To tackle the issue of semantic expression confusion, we employed a masked prediction model that guides the model to generate richer semantic information through learning bidirectional feature combinations in sequence. Addressing challenges such as significant differences in data distribution, we introduced adaptive clustering techniques that manage by generating pseudo-labels across multiple categories. The model is capable of enhancing the expression of hidden features in intermediate layers during the self-supervised training process, enabling it to learn common hidden features across different datasets. This study, by constructing a hybrid dataset and conducting extensive experiments, demonstrated two key findings: (1) our model performs best on multiple evaluation metrics; (2) the model can effectively integrate critical features from different datasets, significantly enhancing the accuracy of emotion recognition.

In recent years, with the development of low-earth orbit broadband satellites, the combination of multi-path transmission and software-defined networking (SDN) for satellite networks has seen rapid advancement. The integration of SDN and multi-path transmission contributes to improving the efficiency of transmission and reducing network congestion. However, the current SDN controllers do not support the multi-path QUIC protocol (MPQUIC), and the routing algorithm used in current satellite networks based on minimum hop count struggles to meet the real-time requirements for some applications. Therefore, this paper designs and implements an SDN controller that supports the MPQUIC protocol and proposes a multi-objective optimization-based routing algorithm. This algorithm selects paths with lower propagation delays and higher available bandwidth for subflow transmission to improve transmission throughput. Considering the high-speed mobility of satellite nodes and frequent link switching, this paper also designs a flow table update algorithm based on the predictability of satellite network topology. It enables proactive rerouting upon link switching, ensuring stable transmission. The performance of the proposed solution is evaluated through satellite network simulation environments. The experimental results highlight that SDN-MPQUIC significantly improves performance metrics: it reduces average completion time by 37.3% to 59.3% compared to QSMPS and by 52.8% to 72.4% compared to Disjoint for files with different sizes. Additionally, SDN-MPQUIC achieves an average throughput improvement of 81.4% compared to QSMPS and 147.8% compared to Disjoint, while demonstrating a 26.3% lower retransmission rate than QSMPS.

The quality of machine learning (ML) models deployed in dynamic environments tends to decline over time due to disparities between the data used for training and the upcoming data available for prediction, which is commonly known as drift. Therefore, it is important for ML models to be capable of detecting any changes or drift in the data distribution and updating the ML model accordingly. This study presents various drift detection techniques to identify drift in the survival outcomes of patients who underwent cardiac surgery. Additionally, this study proposes several drift adaptation strategies, such as adaptive learning, incremental learning, and ensemble learning. Through a detailed analysis of the results, the study confirms the superior performance of ensemble model, achieving a minimum mean absolute error (MAE) of 10.684 and 2.827 for predicting hospital stay and ICU stay, respectively. Furthermore, the models that incorporate a drift adaptive framework exhibit superior performance compared to the models that do not include such a framework.

Entity recognition of product titles is essential for retrieving and recommending product information. Due to the irregularity of product title text, such as informal sentence structure, a large number of professional attribute words, a large number of unrelated independent entities of various combinations, the existing general named entity recognition model is limited in the e-commerce field of product title entity recognition. Most of the current studies focus on only one of the two challenges instead of considering the two challenges together. Our approach proposes NEZHA-CNN-GlobalPointer architecture with the addition of label semantic network, and uses multigranularity contextual and label semantic information to fully capture the internal structure and category information of words and texts to improve the entity recognition accuracy. Through a series of experiments, we proved the efficiency of our approach over a dataset of Chinese product titles from JD.com, improving the F1-value by 5.98%, when compared to the BERT-LSTM-CRF model on the product title corpus.

The above article, published online on 17 February 2022 in Wiley Online Library (wileyonlinelibrary.com), has been retracted by agreement between the Editor-in-Chief, Diana Inkpen, and Wiley Periodicals LLC. The article was published as part of a guest-edited special issue. Following publication, it came to our attention that two of those named as Guest Editors of this issue were being impersonated and/or misrepresented by a fraudulent entity. An investigation by the publisher found that all of the articles, including this one, experienced compromised editorial handling and peer review which was not in line with the journal's ethical standards. Therefore, a decision has been made to retract this article. We did not find any evidence of misconduct by the authors. The authors have been informed of the decision to retract but do not agree with this decision.

Personalized health monitoring and prediction are indispensable in advancing healthcare delivery, particularly amidst the escalating prevalence of chronic illnesses and the aging population. Deep learning (DL) stands out as a promising avenue for crafting personalized health monitoring systems adept at forecasting health outcomes with precision and efficiency. As personal health data becomes increasingly accessible, DL-based methodologies offer a compelling strategy for enhancing healthcare provision through accurate and timely prognostications of health conditions. This article offers a comprehensive examination of recent advancements in employing DL for personalized health monitoring and prediction. It summarizes a diverse range of DL architectures and their practical implementations across various realms, such as wearable technologies, electronic health records (EHRs), and data accumulated from social media platforms. Moreover, it elucidates the obstacles encountered and outlines future directions in leveraging DL for personalized health monitoring, thereby furnishing invaluable insights into the immense potential of DL in this domain.

The above article, published online on 14 November 2021 in Wiley Online Library (wileyonlinelibrary.com), has been retracted by agreement between the Editor-in-Chief, Diana Inkpen, and Wiley Periodicals LLC. The article was published as part of a guest-edited special issue. Following publication, it came to our attention that two of those named as Guest Editors of this issue were being impersonated and/or misrepresented by a fraudulent entity. An investigation by the publisher found that all of the articles, including this one, experienced compromised editorial handling and peer review which was not in line with the journal's ethical standards. Therefore, a decision has been made to retract this article. We did not find any evidence of misconduct by the authors. The authors have been informed of the decision to retract.

Dynamic texts in scene videos provide valuable insights and semantic cues crucial for video applications. However, the movement of this text presents unique challenges, such as blur, shifts, and blockages. While efficient in tracking text, state-of-the-art systems often need help when text becomes obscured or complicated scenes. This study introduces a novel method for detecting and tracking video text, specifically designed to predict the location of obscured or occluded text in subsequent frames using a tracking-by-detection paradigm. Our approach begins with a primary detector to identify text within individual frames, thus enhancing tracking accuracy. Using the Kalman filter, Munkres algorithm, and deep visual features, we establish connections between text instances across frames. Our technique works on the concept that when text goes missing in a frame due to obstructions, we use its previous speed and location to predict its next position. Experiments conducted on the ICDAR2013 Video and ICDAR2015 Video datasets confirm our method's efficacy, matching or surpassing established methods in performance.

The above article, published online on 21 October 2021 in Wiley Online Library (wileyonlinelibrary.com), has been retracted by agreement between the Editor-in-Chief, Diana Inkpen, and Wiley Periodicals LLC. The article was published as part of a guest-edited special issue. Following publication, it came to our attention that two of those named as Guest Editors of this issue were being impersonated and/or misrepresented by a fraudulent entity. An investigation by the publisher found that all of the articles, including this one, experienced compromised editorial handling and peer review which was not in line with the journal's ethical standards. Therefore, a decision has been made to retract this article. We did not find any evidence of misconduct by the authors. The authors have been informed of the decision to retract.

Radiologists and clinicians must automatically examine breast and tumor locations and sizes accurately. In recent years, several neural network-based feature fusion versions have been created to improve medical image segmentation. Multi-modal image fusion photos may efficiently identify tumors. This work uses image fusion to identify computed tomography and magnetic resonance imaging alterations. A Gauss-log ratio operator is recommended for difference image production. The Gauss-log ratio and log ratio difference image complement the objective of improving the difference map through image fusion. The feature change matrix extracts edge, texture, and intensity from each picture pixel. The final change detection map classifies feature vectors as “changed” or “unchanged” which has been mapped for high-resolution or low-resolution pixels. This paper proposes a multi-feature blocks (MFB) based neural network for multi-feature fusion. This neural network modeling approach globalizes pixel spatial relationships. MFB-based feature fusion also aims to capture channel interactions between feature maps. The proposed technique outperforms state-of-the-art approaches which have been discussed in detail in experimental results section.