International Journal of Cognitive Computing in Engineering最新文献

{"title":"Prenatal depression level prediction using ensemble based deep learning model","authors":"Abinaya Gopalakrishnan ,&nbsp;Xujuan Zhou ,&nbsp;Revathi Venkataraman ,&nbsp;Raj Gururajan ,&nbsp;Ka Ching Chan ,&nbsp;Guohun Zhu ,&nbsp;Niall Higgins","doi":"10.1016/j.ijcce.2024.12.002","DOIUrl":"10.1016/j.ijcce.2024.12.002","url":null,"abstract":"<div><h3>Background and objective:</h3><div>Many people find that the emotional and mental strain of labor and delivery is greater than they anticipated. However, there are few reports on stress levels during pregnancy, and there is limited research into stress observation during delivery. Prenatal depression during the delivery has to be monitored continuously without disturbing the mothers during the childbirth.</div></div><div><h3>Methods:</h3><div>We explore the potential of employing EDA for Prenatal Depression prediction. The proposed model applies a novel method for motion artifacts followed by data labeling using PHQ-9 score values and LOOCV applied to train robustly. This culminated in the development of a novel EBDL model to accurately predict stress levels.</div></div><div><h3>Results:</h3><div>We subsequently applied the ensemble based deep learning model on a testing dataset and our method proved to be 93.87 percent accurate, proving its superiority over the standard supervised classification models. The accuracy of this approach applied to three benchmark datasets produced better results compared to all commonly applied machine learning models, including an Ensemble based Deep Learning model.</div></div><div><h3>Conclusion:</h3><div>The preliminary results are promising, and indicate a superior utility of EDA for monitoring stress levels in real-life scenarios. This approach should be applied to a clinical setting, it potentially could continuously monitor stress levels in pregnant women and provide real-time feedback of clinically important data for clinicians.</div></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"6 ","pages":"Pages 267-279"},"PeriodicalIF":0.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143139940","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development of an efficient method for object detection and localization in 3D space using RGBD cameras for autonomous systems","authors":"Nataliya Boyko","doi":"10.1016/j.ijcce.2025.04.005","DOIUrl":"10.1016/j.ijcce.2025.04.005","url":null,"abstract":"<div><div>The work presents an efficient algorithm for object detection, orientation estimation, and isometric positioning in 3D space using RGBD camera data. The goal of the study is to improve the accuracy and processing speed of autonomous navigation and manipulation systems under conditions of limited computational resources. The proposed approach combines heuristic isometry estimation with segmentation methods (DBSCAN), plane estimation (RANSAC), and orientation analysis, enabling effective processing of scenes with planar backgrounds. The main advantage of the algorithm lies in its ability to operate in real time: the processing time for a single frame is only 20 ms, achieving object positioning accuracy up to 5.48 cm. The results of experimental research confirm a high level of accuracy and stability even under challenging conditions. The algorithm outperforms existing models in terms of processing speed while demonstrating comparable or superior positioning accuracy. The practical significance of the proposed method lies in its potential application in mobile robotics, automated warehouse systems, and machine vision systems where high autonomy and precision are required. The algorithm can also be adapted to a broader range of tasks due to its flexible hyperparameter tuning. A key limitation remains the requirement for object placement on a planar surface and the use of a depth camera, which necessitates a specific environmental setup. The proposed method makes a significant contribution to the advancement of computer vision and autonomous robotics technologies, opening prospects for its implementation in next-generation systems.</div></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"6 ","pages":"Pages 537-551"},"PeriodicalIF":0.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143942054","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhanced priority based task scheduling with integrated fault tolerance in distributed systems","authors":"Sheikh Umar Mushtaq ,&nbsp;Sophiya Sheikh ,&nbsp;Sheikh Mohammad Idrees","doi":"10.1016/j.ijcce.2024.12.006","DOIUrl":"10.1016/j.ijcce.2024.12.006","url":null,"abstract":"<div><div>Fault tolerance in task scheduling is a fundamental challenge in cloud computing, particularly in dynamic environments where failures can disrupt task execution and degrade performance. Existing scheduling procedures lack effective mechanisms to adjust task priorities leading to suboptimal performance vigorously. Moreover, Fault Tolerance (FT) is regularly addressed either through replication or checkpointing, but rarely in a synergistic manner. While Replication and checkpointing provide several strengths. However, these also suffer from significant weaknesses when used in isolation. Replication increases cost and resource usage while Checkpointing suffers from frequent storage overheads. This research addresses these limitations by proposing the dynamic priority-based task and integrated fault-tolerance method that combines the strengths of both replication and check pointing methods to achieve high recovery rates and efficient resource utilization. Replication supplies instant backup options, while checkpointing reduces recovery time and computational loss, and addresses the need for more adaptive and robust scheduling solutions in complex computing environments. This paper portrays a novel Priority-Based Task Scheduling and Fault Tolerant model (PBTS-FT) that computes priorities via response ratios, intending to optimize resource allocation and system responsiveness. We also accompany PBTS-FT with an integrated FT scheme that combines replication and checkpointing techniques, improving system flexibility to failures by considering both VM and task failures. The results were carried out in two setups that promise enhanced QoS (Quality of Service) including recovery, makespan, speedup, efficiency, resource utilization, and progress, thereby presenting a valuable contribution to the field. The same is proved in simulation experiments which showed recovery improvements of 60.75 %, 63.75 %, and 8.79 % compared to Replication, Checkpointing, and FTMSFIS methods respectively. Compared to Modified Min Min, HEFT, and PETS, the proposed PBTS-FT improves resource utilization by about 6.27 %, 12.89 %, and 14.20 % respectively. This provides empirical evidence that the proactive choice of FT methods substantially boosts the success rate of task completion and other parameters.</div></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"6 ","pages":"Pages 152-169"},"PeriodicalIF":0.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143139985","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning for object recognition: A comprehensive review of models and algorithms","authors":"Paschalis Tsirtsakis,&nbsp;Georgios Zacharis,&nbsp;George S. Maraslidis,&nbsp;George F. Fragulis","doi":"10.1016/j.ijcce.2025.01.004","DOIUrl":"10.1016/j.ijcce.2025.01.004","url":null,"abstract":"<div><div>The rapid advancements in artificial intelligence (AI) and machine learning (ML) have significantly enhanced progress in computer vision, opening doors to innovative technological possibilities and enabling a range of real-world applications. Despite these developments, object recognition remains a complex domain with persistent challenges and limitations. This work seeks to address these challenges by investigating the effectiveness of deep learning (DL) methods in object detection tasks. Leveraging DL allows for the direct learning of feature representations from image data, resulting in advanced performance. This review provides a comprehensive examination of fundamental models and algorithms, with a particular focus on neural network (NN) frameworks utilized for feature extraction. Additionally, it evaluates the benchmark datasets commonly employed to assess their performance in object recognition tasks.</div></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"6 ","pages":"Pages 298-312"},"PeriodicalIF":0.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143139941","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cognitive computing method based on decoding psychological emotional states","authors":"Baihui Huangfu ,&nbsp;Wenjuan Cheng","doi":"10.1016/j.ijcce.2024.10.002","DOIUrl":"10.1016/j.ijcce.2024.10.002","url":null,"abstract":"<div><div>The current artificial intelligence is constrained to passively executing human command control. It cannot perceive, learn, and guide itself. Furthermore, the majority of these systems are unable to comprehend the human psychological cognitive state or emotional intention actively. To enhance the universality and generalization ability of existing emotion recognition methods, a psychological emotion state decoding method combining singular spectrum analysis and entropy measure is designed based on an electroencephalography signal. This method is developed to empower the cognitive computing system with the ability to perceive emotions. The entropy measure is accelerated by the vector dissimilarity criterion. Finally, dynamic sample entropy model learning is used to realize the cognitive calculation of psychological and emotional state recognition. The results demonstrated that the computational efficiency of the proposed decoding algorithm was significantly improved. The average recognition accuracy of positive and negative emotional states reached 82.78±16.22. The best recognition accuracy of the proposed electroencephalography emotion recognition method was 84.64 %. The proposed electroencephalography emotion recognition method demonstrated an accuracy of 64.15 % in recognizing cross-individual positive, neutral, and negative emotional states. The proposed emotion recognition method shows better universality and generalization performance, and can effectively identify people's psychological and emotional states.</div></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"6 ","pages":"Pages 32-43"},"PeriodicalIF":0.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142554008","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Empirical research on the evolution trend of heat and sentiment for emergencies","authors":"Shihong Wu ,&nbsp;Wei Yu ,&nbsp;Yanxia Zhao ,&nbsp;Hongyan Li ,&nbsp;Jiatong Wang ,&nbsp;Weiyan Yang ,&nbsp;Yue Lin","doi":"10.1016/j.ijcce.2025.01.006","DOIUrl":"10.1016/j.ijcce.2025.01.006","url":null,"abstract":"<div><div>Emergencies inflict heavy casualties, economic losses, ecological damage, and significant social harm to society. By segmenting information topics and analysing emotional shifts, we can identify corresponding real-world events and their impacts, thereby providing guidance for timely responses to emergencies. In the past, public opinion monitoring of emergencies was based mainly on single-topic detection or emotion analysis, which cannot comprehensively evaluate the evolution of public opinion. In this work, word segmentation is applied to video comments related to various emergency situations. By utilizing the co-word network and Louvain algorithm for theme division, along with sentiment analysis constructed through time series analysis of sentiment value changes for various emergencies employing the naive Bayes method, the evolution of public opinion is comprehensively assessed. As a result, the pivotal nodes in the evolution of public opinion are identified and the evolution process is divided into stages. Using this method, relevant management departments can effectively address the majority of public opinions for various types of emergencies, addressing them from the perspectives of prevention, adjustment, and recovery. This approach not only enhances rescue efficiency and strengthens safety management but also actively guides the evolution of public opinion, ultimately providing society with solid and reliable security safeguards.</div></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"6 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143139942","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A tweet sentiment classification approach using an ensemble classifier","authors":"Vidyashree KP ,&nbsp;Rajendra AB ,&nbsp;Gururaj HL ,&nbsp;Vinayakumar Ravi ,&nbsp;Moez Krichen","doi":"10.1016/j.ijcce.2024.04.001","DOIUrl":"https://doi.org/10.1016/j.ijcce.2024.04.001","url":null,"abstract":"<div><p>Social media users are more receptive to products or events and share their thoughts through raw textual data, which is classified as semi-structured data. This data, which is presented using a variety of terminologies, is noisy by nature but yet contains important information and superfluous details, giving analysts a way to identify patterns and knowledge. This hidden information must be extracted from language data in order to make informed decisions and create strategic plans for entering new markets. Among the most prominent fields of study are natural language processing (NLP) and data mining techniques, especially when it comes to sentiment analysis—the process of identifying the feelings and insights concealed in the data. Twitter is one of the significant microblogging platform with millions of users. These users use Twitter to share sentiments using hash tags on different topics and to make status updates known as tweets. Twitter is therefore regarded as a significant real-time source and as one of the most active opinion indicators. The volume of information is produced by Twitter is enormous and manually scanning the entire data set is difficult process. The paper proposed an ensemble classifier to categorize emotion of the tweets on the basis of polarities such as positive and negative.</p><p>In our study, we ensemble classifiers which is a combination of Random Forest (RF), Support Vector Machine (SVM) and Decision Tree (DT). The data is collected from Twitter API and the Twitter data is analysed autonomously to define public view on particular topic. The features obtained after the process of dimensionality reduction using LDA undergoes the stage of feature selection using Wrapper based technique. The iterative Wrapper based technique predict score for the features, the features with low score are ignored and high score is proceeded for classification. The ensemble classifier used Adaptive Boosting (AdaBoost) technique where the output from the Machine Learning (ML) classifiers are combined to produce a single output. Adaboost combines the poor classifiers and extracts the prediction value to make a better classifier. The experimental results show that the proposed ensemble classifier provides better accuracy of 93.42 % that is comparatively better than existing Convolutional Bidirectional - Long Short-Term Memory (ConvBiLSTM) classifier and Hybrid Lexicon- Naïve Bayes Classifier (HL-NBC) which produce classification accuracy of 91.53 % and 89.61 % respectively.</p></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"5 ","pages":"Pages 170-177"},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2666307424000123/pdfft?md5=64e2fa7ef45bd04e15a561dd4e416a52&pid=1-s2.0-S2666307424000123-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140843930","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A multi-fused convolutional neural network model for fruit image classification","authors":"Bam Bahadur Sinha ,&nbsp;R. Dhanalakshmi","doi":"10.1016/j.ijcce.2024.09.003","DOIUrl":"10.1016/j.ijcce.2024.09.003","url":null,"abstract":"<div><p>Fruit categorization presents a significant challenge due to the diverse range of fruit types and their similarities in color, shape, size, and structure. This challenge is addressed in this research by proposing a multi-fused CNN approach that integrates three state-of-the-art deep learning models, namely EfficientNet-B0, MobileNetV2, and ResNet50V2, into a unified pipeline. The primary objective is to enhance prediction accuracy and mitigate overfitting issues commonly encountered with conventional techniques. Utilizing the Fruit-360 dataset (both small and full versions), the proposed multi-fused CNN model demonstrates superior performance, achieving an impressive accuracy rate of 99.32% and 97.15% in classifying fruit types across 24 and 131 different categories, respectively. This substantial improvement underscores the potential of the proposed approach in reliably detecting fruit types. In future work, efforts will be directed towards optimizing the model for enhanced processing speed, facilitating real-time performance in practical applications, particularly in agricultural automation. Additionally, rigorous experiments on diverse real-world fruit images will be conducted to further validate the robustness of the proposed model.</p></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"5 ","pages":"Pages 416-424"},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2666307424000354/pdfft?md5=88d9e15d3de689340a58162058be802e&pid=1-s2.0-S2666307424000354-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142228537","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning-based human activity recognition using CNN, ConvLSTM, and LRCN","authors":"Md. Ashraf Uddin ,&nbsp;Md. Alamin Talukder ,&nbsp;Muhammad Sajib Uzzaman ,&nbsp;Chandan Debnath ,&nbsp;Moumita Chanda ,&nbsp;Souvik Paul ,&nbsp;Md. Manowarul Islam ,&nbsp;Ansam Khraisat ,&nbsp;Ammar Alazab ,&nbsp;Sunil Aryal","doi":"10.1016/j.ijcce.2024.06.004","DOIUrl":"https://doi.org/10.1016/j.ijcce.2024.06.004","url":null,"abstract":"<div><p>Human activity recognition (HAR) plays a crucial role in assisting the elderly and individuals with vascular dementia by providing support and monitoring for their daily activities. This paper presents a deep learning (DL)-based approach to HAR, leveraging convolutional neural network (CNN), convolutional long short-term memory (ConvLSTM), and long-term recurrent convolutional network (LRCN) architectures. These models are designed to extract spatial features and capture temporal dependencies in video data, enhancing the accuracy of activity classification. We conducted experiments on the UCF50 and HMDB51 video datasets, encompassing diverse human activities. Our evaluation demonstrates that the ConvLSTM model achieves an accuracy of 82% on UCF50 and 68% on HMDB51, while the LRCN model gives accuracies of 93.44% and 71.55%, respectively. Finally, the CNN model outperforms with an accuracy rate of 99.58% for the UCF50 and 92.70% for the HMDB51 datasets. These significant improvements showcase the effectiveness of integrating convolutional and recurrent neural networks for HAR tasks. Our research contributes to advancing HAR systems with potential applications in healthcare, assisted living, and surveillance. By accurately recognizing human activities, our models can assist in remote patient monitoring, fall detection, and public safety initiatives. These findings underscore the importance of DL in enhancing the quality of life and safety for individuals in various contexts.</p></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"5 ","pages":"Pages 259-268"},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2666307424000214/pdfft?md5=76e7649d648013ec357377724eac33ad&pid=1-s2.0-S2666307424000214-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141543703","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-objective Harris Hawks optimization algorithm for selecting best location and size of distributed generation in radial distribution system","authors":"Sandeep Gogula ,&nbsp;V․ S. Vakula","doi":"10.1016/j.ijcce.2024.08.003","DOIUrl":"10.1016/j.ijcce.2024.08.003","url":null,"abstract":"<div><p>Distributed Generating Units (DGUs) are widely used in backup power, renewable energy integration, voltage regulation, and energy storage applications. The DGUs are small-scale power-generating units located near the load centers of a power system, as opposed to large, centralized power plants located far away. However, the DGUs must reduce power losses and environmental impacts and improve the voltage profile. This was achieved by effective optimization of the position and rating of DGU. However, the conventional optimization methods failed to optimize the power losses, voltage-current balancing issues, and harmonic distortions. So, this article presents the selection of the best location and size of DGU in radial distribution systems using a nature-inspired multi-objective Harris Hawks optimization (HHO) algorithm while considering improvements in the voltage profile and reductions in active power losses. The HHO algorithm is inspired by the supportive actions of intelligent birds, such as Harris Hawks, while searching for prey. The distinguished hunting process of Harris Hawks using other family members makes it unique, which is useful for searching for better quality solutions while optimizing multi-objective fitness functions. Fitness function is determined by considering each bus's voltage profile and branches' active power losses. The proposed method is tested on two circle distribution systems (69 bus and 118 bus). The results are compared to those obtained using some of the more well-known optimization techniques given by scholars in recent years. The optimization of seven DGU in 118 bus radial distribution systems resulted in 405.32 kW of active power losses, a 68.78% decrease in losses, and 0.9723 Vs of minimum voltage.</p></div>","PeriodicalId":100694,"journal":{"name":"International Journal of Cognitive Computing in Engineering","volume":"5 ","pages":"Pages 436-452"},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2666307424000287/pdfft?md5=b0e1c17a253d3e12fd5852d1ef56ca90&pid=1-s2.0-S2666307424000287-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142241250","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}