检索结果-内蒙古大学图书馆

International Conference on Machine Learning and Cybernetics (ICMLC)

作者： Angga Wahyu Wibow Eri Sato-Shimokawara Kurnianingsih Department of Computer Science Graduate School of System Design Tokyo Metropolitan University Tokyo Japan Department of Electrical Engineering Politeknik Negeri Semarang Semarang Indonesia

ISBN: (数字)9798331528041

ISBN: (纸本)9798331528058

The field of Human Activity Recognition (HAR) is growing significantly in several areas but little research focuses on cultural behavior. How machine learning can explain human activity as a promotional tool in understanding cultural differences in a region is very challenging. Studies using Recurrent Neural Network (RNN), especially Long Short-Term Memory (LSTM), require large data when used to research human activity recognition. Using small data is a challenge when using LSTM. This study aims to determine which data augmentation is most appropriate when used using LSTM to predict Japanese greeting gestures, namely eshaku, keirei, saikeirei, waving hands, and Indonesian greeting gestures, namely the movement of bringing both hands together in front of the chest with a slight bow of the head. This study proposes to compare several data augmentations, namely jittering, scaling, permutation, cropping, reverse, time shifting, frequency domain, and synthetic data generation. We evaluate model performance using the Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and coefficient of determination (R 2 ) metrics. The experimental results show that Time Shifting has the best MSE, RMSE, and MAE values compared to others and Frequency Domain has the best R 2 values compared to others. This study concludes that Time Shifting and Frequency Domain are the best data augmentation methods when using LSTM in predicting greeting gestures using small data. These findings provide valuable insights into the effectiveness of various data augmentation methods in motion prediction tasks with limited data sets, as well as the importance of selecting appropriate data augmentation methods in human motion analysis. Future work could use data from two or more people using model explanations generated by machine learning and developing a new method for cultural behavior.

关键词： Hands Recurrent neural networks Frequency-domain analysis Machine learning Data augmentation Human activity recognition Cultural differences Motion analysis Long short term memory Synthetic data

来源：评论

学校读者我要写书评

暂无评论

Genre Classification of Bangla Poem Using Machine Learning and Deep Learning Techniques

Genre Classification of Bangla Poem Using Machine Learning a...

引用

AI IoT Congress (AIIoT), World

作者： Syed Tangim Pasha Ashraful Islam Mohammed Masudur Rahman Eshtiak Ahmed Md. Ferdouse Ahmed Foysal Md Zahangir Alam Department of Computing and Information System Daffodil International University Bangladesh Center for Computational and Data Sciences Independent University Bangladesh Department of Computer Science and Engineering Independent University Bangladesh Department of Computer Science and Engineering International Standard University Bangladesh Department of Computer Science and Engineering Daffodil International University Bangladesh

The computational analysis of the Bangla poems is a challenging task due to the diverse linguistic, stylistic, and semantic features of the Bangla language. In this work, we prepared a dataset of 1311 Bangla poems of two separate categories: Love and Miscellaneous poem, which contain 500 and 811 poems respectively. We used word or semantic-based features to classify Bangla poems using the TF-IDF feature techniques. We used Logistic Regression, Naïve Bayes (NB), and Support Vector Machine (SVM) models for classification through machine learning, and we used Bayesian optimization techniques for hyperparameters tuning of these three models. We also used LSTM, CNN, and transformer models for this research. For the performance evaluation of the classification models, we used four evaluation metrics of precision, recall, F1-score, and accuracy. We also used the ROC-AUC curve to distinguish between all the machine learning and deep learning models. The experimental results expressed that, the transformer model achieved the highest accuracy compared to all the typical machine learning and deep learning models with an accuracy of 87%.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Identify the potential pathways and candidate biomarkers of stroke associated with bipolar disorder: Bioinformatics and system biology approach

引用

Journal of Proteins and Proteomics 2025年第2期16卷 141-159页

作者： Hosen, Md. Faruk Basar, Md. Abul Yasmin, Mst. Farjana Hasan, Md. Rakibul Uddin, Muhammad Shahin Department of Information and Communication Technology (ICT) Mawlana Bhashani Science and Technology University (MBSTU) Tangail Bangladesh Department of Computing and Information System (CIS) Daffodil International University (DIU) Dhaka Bangladesh Department of Computer Science and Engineering (CSE) Uttara University Dhaka Bangladesh Department of Software Engineering (SWE) Daffodil International University (DIU) Dhaka Bangladesh

The field of computational bioinformatics and systems biology analysis is rapidly expanding due to advancements in bioinformatics tools. Bipolar disorder (BD) is a severe psychiatric condition affecting both adults and adolescents. In recent years, the risk of stroke has increased among individuals with BD. Background studies indicate that BD and stroke share numerous biochemical and genetic characteristics. Due to traditional endocrinological approaches, no collaborative research exists, and no treatment options have been established for BD and stroke patients. The aim of this study is to identify common molecular pathways and potential therapeutic targets in BD and stroke that could be utilized to predict disease progression. To achieve this, shared genes were identified to determine common pathways. Based on the biochemical, molecular, and genetic interactions among these shared genes, this study identifies the most significant hub genes. To further explore these interactions, analysis such as the protein–protein interaction (PPI) network, topological properties, enrichment analysis, co-expression network, gene regulatory network (GRN), and physical interaction network are conducted. The GO annotation indicates that most of the genes are linked to MAP kinase activity and tumor necrosis factor receptor superfamily binding. The KEGG pathway analysis revealed that the predominant genes are associated with apoptosis and the IL-18 signaling pathway. This comparison helps understand the biochemical and genetic characteristics shared between BD and stroke. Predictive drug analysis has identified consistent potential biomarkers associated with both conditions, providing a scientific foundation for investigating their diagnosis, treatment, and prognosis. Finally, chemical experiments could be conducted to further validate the efficacy of these drugs.

关键词：

来源：评论

学校读者我要写书评

暂无评论

DFIL: an Efficient Web Service Prediction Method based on Deep Feature Interactive Learning 25

DFIL: an Efficient Web Service Prediction Method based on De...

引用

25th IEEE International Conference on computer Supported Cooperative Work in Design, CSCWD 2022

作者： Zhu, Ke Wang, Zhixin Jiang, Tao Xiao, Yingyuan Zheng, Wenguang Hsu, Ching-Hsien Tianjin University of Technology Engineering Research Center of Learning-Based Intelligent System Ministry of Education Tianjin China He'nan University of Economics and Law School of Computer and Information Engineering Zhengzhou China Asia University Department of Computer Science and Information Engineering Taichung Taiwan

ISBN: (纸本)9781665405270

As a representative of nonfunctional features, the Quality-of-Service (QoS) plays an important role in recommending the best service to users. To obtain the QoS value of web services, researchers have proposed many web service QoS prediction methods. However, most existing methods, e.g., the collaborative filtering method, leverage the historical user service invocation information to predict the QoS value, and only consider the QoS information of similar users or services, and ignore the attributes and characteristics of users and services. In this paper, we propose a Deep Feature Interaction Learning method, called DFIL for short, for web service QoS prediction. Specifically, (1) DFIL constructs a feature extraction method to effectively obtain the features of users and services. (2) DFIL trains a novel neural network for QoS prediction, which can accurately find the potential relationship between users and services based on the feature information, and provide good web service recommendation. © 2022 IEEE.

关键词： Quality of service

来源：评论

学校读者我要写书评

暂无评论

Enhancing Short-Term Load Forecasting Accuracy in the Power systems Using a Hybrid CNN-GRU Model 5

Enhancing Short-Term Load Forecasting Accuracy in the Power ...

引用

5th International Conference on Electrical, Communication and computer engineering, ICECCE 2024

作者： Ahsan, Muhammad Aziz, Muhammad Osama Bokhari, Syed Ahmad Saleem Haris, Muhammad Iqbal, Mansoor Ullah, Kaleem University of Alabama at Birmingham Department of Computer Science BirminghamAL35294 United States University of Engineering and Technology Buildings Energy Research Center Peshawar Pakistan Harvard Extension School Harvard University CambridgeMA02138 United States Institute for Insight - J Mack Robinson College of Business Georgia State University Georgia United States University of Wah Department of Computer Science Wah Cantt Pakistan CASE UET Peshawar Department of Electrical Energy System Peshawar Pakistan

ISBN: (纸本)9798331529437

In modern power systems, Short-Term Load Forecasting (STLF) is critical in ensuring the dynamic allocation of operational reserves and accurate prediction of electrical load demand. STLF is pertinent to grid operators to ensure system stability, allocate generation resources efficiently, and provide adequate reserves to deal with unexpected fluctuations. Considering the pivotal role of STLF, this paper introduces a hybrid deep learning model for short-term load forecasting, combining Convolutional Neural Networks (CNN) and Gated Recurrent Units (GRU). The model features two parallel 32-filter convolutional layers (3×3 kernel, ReLU) for spatial feature extraction, followed by two series-connected GRU layers (42 units) to capture temporal dependencies. A final dense layer with ReLU activation generates the load forecast. This architecture effectively leverages CNNs for local pattern recognition and GRUs for modeling temporal sequences, offering a robust solution for accurate short-term load forecasting. Therefore, the efficacy of the CNN-GRU model is tested and validated using an extensive dataset sourced from the National Transmission & Dispatch Company (NTDC), Pakistan. The performance analysis reveals that hybrid CNN-GRU models achieve a substantial reduction in Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Root Mean Square Error (RMSE), with values of 400.33, 2.89, 580.95, for single-step forecasts, respectively, and 1031.11, 5.31, and 733.19, for 24-hour forecasts. Furthermore, the simulation results demonstrate notable enhancements in load forecasting accuracy and robustness across various forecast horizons. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Technical Job Distribution at BSD SHARP Service Center Using Combination of Naïve Bayes and K-Nearest Neighbour

Technical Job Distribution at BSD SHARP Service Center Using...

引用

2022 IEEE Symposium on Wireless Technology and Applications, ISWTA 2022

作者： Pebrianti, Dwi Ariawan, Angga Bayuaji, Luhur Mahdiana, Deni Rusdah, Rusdah Universitas Budi Luhur Magister of Computer Science Faculty of Information Technology Jakarta Indonesia International Islamic University Malaysia Faculty of Engineering Dept. of Mechanical Engineering Gombak Malaysia Universitas MNC Faculty of Computer Science Department of Computer and Media Jakarta Indonesia Universitas Budi Luhur Faculty of Information Technology Department of Information System Jakarta Indonesia

ISBN: (数字)9781665484824

ISBN: (纸本)9781665484824

Works distribution is a routine carried out every day by the head of the branch in the SHARP Service Center. The accuracy of the labor division is very important to get customer satisfaction. Inappropriate work distribution can increase complaints from customers. Currently, works distribution in SHARP Service Center is carried out manually, where the works received on the selected system is then shared through the document provided. Time taken for this process is about 1.42 minutes on average for each damage reports. Speed of Service also depends on the Head of department's expertise and experience. In this study, an automatic system based on Machine Learning will be designed for the technicians work distribution by using a combination of k Nearest Neighbor (k-NN) and Naïve Bayes. Naïve Bayes algorithm is used to improve the feature extraction accuracy by considering the feature below the average (α). Meanwhile, k-NN algorithm is used to classify the experimental data. From the study, it is found that the best of k value for k-NN algorithm is 15. It is known that a high number of accuracy values, the labor distribution can be more accurate. The validation of the proposed method is conducted by using a confusion matrix with a composition of 80% training data and 20% test data. The single Classifier test with the Naïve Bayes algorithm produces the highest accuracy value of 72.7%, while using k-NN algorithm is 81.5%. With a combination of Naive Bayes and k-NN algorithms, the accuracy value is increasing to 86%. This result shows that the proposed method improves the accuracy by 13.3% on single Naive Bayes algorithms and 4% on a single k-NN algorithm. The results obtained show that in the manual process, the average time per job is 1.42 minutes, while by using the proposed method, the average processing time is around 0.03 seconds per job. An increase of 2480 times faster is found and confirmed during the implementation of the proposed method. © 2022 IEEE.

关键词： Customer satisfaction

来源：评论

学校读者我要写书评

暂无评论

LBA: Matching Theory Based Latency-Sensitive Binary Offloading in IoT-Fog Networks

LBA: Matching Theory Based Latency-Sensitive Binary Offloadi...

引用

International Communication systems and Networks and Workshops, COMSNETS

作者： Priyanka Soni Omkar Chand Deshlahre Anurag Satpathy Sourav Kanti Addya Department of Computer Science and Engineering Cloud and Smart System Services Lab National Institute of Technology Karnataka India Missouri University of Science and Technology Rolla MO USA

The Internet of Things (IoT) is growing more popular with applications like healthcare services, traffic monitoring, video streaming, smart homes, etc. These applications produce an enormous amount of data, so a realistic option in this instance is to offload computational tasks to their proximity fog nodes (FNs) instead of the remote cloud. However, a negligent offloading strategy may cause anomalous computational traffic load at the FNs, causing congestion that may adversely affect the latency. However, the latency of task flows from IoT devices comprises communications latency at BS and computational latency at FNs. Therefore, designing offloading algorithms to distribute the computational load at FN evenly and efficiently utilize the FN resources is crucial. To solve this problem, we proposed LBA in a fog network with a binary offloading strategy using the matching theory-based approach. We utilize the Analytic Hierarchy Process (AHP) to generate the preference list. Furthermore, the binary offloading technique follows the deferred acceptance algorithm (DAA) to produce a stable assignment, and the complete offloading problem is modeled as a one-to-many matching game. Comprehensive simulations ensure that LBA can accomplish a better-balanced assignment for homogeneous and heterogeneous input concerning all the baseline algorithms.

关键词：

来源：评论

学校读者我要写书评

暂无评论

PipeEdge: A Trusted Pipelining Collaborative Edge Training based on Blockchain 23

PipeEdge: A Trusted Pipelining Collaborative Edge Training b...

引用

2023 World Wide Web Conference, WWW 2023

作者： Yuan, Liang He, Qiang Chen, Feifei Dou, Ruihan Jin, Hai Yang, Yun School of Computer Science and Technology Huazhong University of Science and Technology China Department of Computing Technologies Swinburne University of Technology Australia School of Information Technology Deakin University Australia Faculty of Mathematics University of Waterloo Canada National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab Huazhong University of Science and Technology Wuhan430074 China

ISBN: (纸本)9781450394161

Powered by the massive data generated by the blossom of mobile and Web-of-Things (WoT) devices, Deep Neural Networks (DNNs) have developed both in accuracy and size in recent years. Conventional cloud-based DNN training incurs rapidly-increasing data and model transmission overheads as well as privacy issues. Mobile edge computing (MEC) provides a promising solution by facilitating DNN model training on edge servers at the network edge. However, edge servers often suffer from constrained resources and need to collaborate on DNN training. Unfortunately, managed by different telecoms, edge servers cannot properly collaborate with each other without incentives and trust. In this paper, we introduce PipeEdge, a scheme that promotes collaborative edge training between edge servers by introducing incentives and trust based on blockchain. Under the PipeEdge scheme, edge servers can hire trustworthy workers for pipelined DNN training tasks based on model parallelism. We implement PipeEdge and evaluate it comprehensively with four different DNN models. The results show that it outperforms state-of-the-art schemes by up to 173.98% with negligible overheads. © 2023 ACM.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Light Shift Detection Using a Multiple Photodetection Method Suitable for a Chip-Scale Atomic Clock

Light Shift Detection Using a Multiple Photodetection Method...

引用

IEEE International Frequency Control Symposium

作者： Masahiro Fukuoka Shigeyoshi Goka Department of Electrical Engineering and Computer Science Graduate School of System Design Tokyo Metropolitan University Tokyo Japan

SummaryIn this study, we demonstrated light shift detection in a running atomic clock using the multi-photodetection method. This method enables the observation of atomic resonances with different intensities using the intensity distribution of the laser beam. As it requires only two photodiodes to detect light shifts, it is easily applicable to miniature vapor-cell-based atomic clocks such as chip-scale atomic clocks and miniature atomic clocks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

ML-Driven HR system: Candidate Assessment for Enhanced Recruitment Outcomes

ML-Driven HR System: Candidate Assessment for Enhanced Recru...

引用

Automation, Computing and Renewable systems (ICACRS), International Conference on

作者： G C Dixon S Srithanujan H D T N Ranaweera R.M.P M Ariyarathne K L Weerasinghe N Gamage Department of Information Technology SriLanka Institure of Information Technology Sri Lanka Department of Computer System Engineering SriLanka Institure of Information Technology Sri Lanka Department of Computer Science and Software Engineering SriLanka Institure of Information Technology Sri Lanka

ISBN: (数字)9798331532420

ISBN: (纸本)9798331532437

This study developed and implemented a machine learning-driven HR system designed to enhance recruitment outcomes, particularly in online interviews. Challenges such as candidate impersonation, interview question memorization, and deepfake manipulations were addressed. The system incorporated several critical processes utilizing advanced algorithms: pre-interview shortlisting employed Support Vector Machines (SVM) achieving 92% accuracy; cognitive skills assessment utilized adaptive testing methods powered by fine-tuned Large Language Models (LLMs) with 88% accuracy; emotional intelligence evaluation was conducted using deep learning-based emotion detection algorithms attaining 85% accuracy; face verification leveraged Convolutional Neural Networks (CNNs) for identity and deepfake detection with 90% accuracy; online video interviews were analyzed through integrated audio-visual processing techniques; language proficiency testing employed Natural Language Processing (NLP) models reaching 87% accuracy; analysis of thinking patterns was performed using sentiment analysis algorithms with 83% accuracy; and candidate grading was facilitated by ensemble learning methods ensuring consistent and reliable evaluations. The results demonstrated the system’s effectiveness in verifying candidate authenticity and assessing competence, providing a robust solution for modern recruitment challenges.

关键词： Support vector machines Deepfakes Analytical models Sentiment analysis Renewable energy sources Accuracy Reliability Interviews Recruitment Testing

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：