检索结果-内蒙古大学图书馆

AI-Based Feature Extraction and Cosine Similarity for Automation of Student Learning Assessment

学校读者我要写书评

暂无评论

AI-Based Feature Extraction and Cosine Similarity for Automa...

information technology (ACIT)

作者： Bambang Dwi Wijanarko Bachtiar Rosilah Binti Hassan Dina Fitria Murad Ridho Bramulya Ihsan Yaya Heryadi Engkos Achmad Kuncoro Eka Julianti Computer Science Department BINUS Online Learning Bina Nusantara University Jakarta Indonesia Master of English Education Graduate School Universitas Terbuka Jakarta Indonesia Center for Cyber Security Faculty of Information Science and Technology Universiti Kebangsaan Malaysia Bangi Selangor Malaysia Information Systems Department BINUS Online Learning Bina Nusantara University Jakarta Indonesia Manajemen Department BINUS Online Learning Bina Nusantara University Jakarta Indonesia Computer Science Department BINUS Graduate Program-Doctor of Computer Science Bina Nusantara University Jakarta Indonesia Management Department BINUS Business School Undergraduate Program Bina Nusantara University Jakarta Indonesia Information System Department Universitas Terbuka Jakarta Indonesia

This research is ongoing research into the student learning process which aims to develop artificial intelligence-based technology to calculate essay exam scores automatically, based on the textual proximity of students' answers to the answer key. This research uses an approach based on natural language processing (NLP) and text analysis. Feature extraction was carried out using the BERT model and similarity comparison using the Cosine distance method. Model evaluation was carried out using sensitivity and specificity tests. The model evaluation results show a sensitivity of 98.21%, which means this assessment model is able to detect positive cases well.

关键词：

AVTENet: Audio-Visual Transformer-based Ensemble Network Exploiting Multiple Experts for Video Deepfake Detection

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Hashmi, Ammarah Shahzad, Sahibzada Adil Lin, Chia-Wen Tsao, Yu Wang, Hsin-Min The Social Networks and Human-Centered Computing Program Taiwan International Graduate Program Academia Sinica Taipei11529 Taiwan The Institute of Information Systems and Applications National Tsing Hua University Hsinchu30013 Taiwan The Department of Computer Science National Chengchi University Taipei11605 Taiwan The Department of Electrical Engineering The Institute of Communications Engineering National Tsing Hua University Hsinchu300044 Taiwan The Research Center for Information Technology Innovation Academia Sinica Taipei11529 Taiwan The Institute of Information Science Academia Sinica Taipei11529 Taiwan

Forged content shared widely on social media platforms is a major social problem that requires increased regulation and poses new challenges to the research community. The recent proliferation of hyper-realistic deepfake videos has drawn attention to the threat of audio and visual forgeries. Most previous work on detecting AI-generated fake videos only utilizes visual modality or audio modality. While there are some methods in the literature that exploit audio and visual modalities to detect forged videos, they have not been comprehensively evaluated on multi-modal datasets of deepfake videos involving acoustic and visual manipulations. Moreover, these existing methods are mostly based on CNN and suffer from low detection accuracy. Inspired by the recent success of Transformer in various fields, to address the challenges posed by deepfake technology, in this paper, we propose an Audio-Visual Transformer-based Ensemble Network (AVTENet) framework that considers both acoustic manipulation and visual manipulation to achieve effective video forgery detection. Specifically, the proposed model integrates several purely transformer-based variants that capture video, audio, and audio-visual salient cues to reach a consensus in prediction. For evaluation, we use the recently released benchmark multi-modal audio-video FakeAVCeleb dataset. For a detailed analysis, we evaluate AVTENet, its variants, and several existing methods on multiple test sets of the FakeAVCeleb dataset. Experimental results show that our best model outperforms all existing methods and achieves state-of-the-art performance on Testset-I and Testset-II of the FakeAVCeleb dataset. © 2023, CC BY.

关键词： Statistical tests

Air Quality Prediction Based on Air Pollution Emissions in the City Environment Using XGBoost with SMOTE

学校读者我要写书评

暂无评论

Air Quality Prediction Based on Air Pollution Emissions in t...

International Conference of Computer science and information technology

作者： Adli A Nababan Sutarman Muhammad Zarlis Erna B Nababan Department of Computer Science Faculty of Computer Science and Information Technology Universitas Sumatera Utara Medan Indonesia Department of Mathematics Faculty of Mathematics and Natural Sciences Universitas Sumatera Utara Medan Indonesia Department of Information Systems Management BINUS Graduate Program-Master of Information Systems Management Bina Nusantara University Jakarta Indonesia

ISBN: (纸本)9798350399080

Air pollution makes it worse in populated regions. Indonesia's major cities are also impacted by air pollution. Due to rising traffic, material consumption by vehicles, industrial growth, land burning, and garbage collection, air quality has altered significantly. Accurate measurement and classification of air quality are required. Results of accurate classification aid in forming governmental regulations. To reach the criteria for livable air quality, we intend to manage controls. Some of the characteristics of air pollution include particulate (PM10) or PM25, sulfides (in the form of SO2), carbon monoxide (CO), ozone (O3), and nitrites (NO2). Using the XGBoost algorithm and synthetic minority oversampling (SMOTE) methods based on the Air Pollution Standard Index (ISPU) category, this study seeks to uncover factors that affect air quality. XGBoost, an ensemble machine learning approach built on a decision tree and applying a gradient reinforcement framework, is the classification algorithm in use. Air quality categorization has been tested and shown to work with the XGBoost algorithm. The dataset used in this study is air quality in 2021 which was collected by the Jakarta Environmental Ministry The proposed classification model will be evaluated using the Repeated k-fold cross-validation method for objective results. The results of the study prove that SMOTE and XGBoost have better performance than using only the XGBoost method in predicting air quality, where the overall accuracy value of SMOTE with XGBoost is 99.60%, The overall precision value is 99.60%, the overall recall value is 99.60% and the overall f1-score is 99.60% and the overall ROC AUC value is 99.96%. The overall SMOTE method with the proposed XGBoost has given better performance in air pollution prediction.

关键词： Machine learning algorithms Urban areas Air pollution Prediction algorithms Particle measurements Regulation Classification algorithms

Multi-state Chiral Switching Through Adiabaticity Control in Encircling Exceptional Points

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Li, Aodong Wang, Jian Alù, Andrea Chen, Lin Wuhan National Laboratory for Optoelectronics School of Optical and Electronic Information Huazhong University of Science and Technology Wuhan430074 China Photonics Initiative Advanced Science Research Center City University of New York New YorkNY10031 United States Physics Program Graduate Center City University of New York New YorkNY10016 United States Shenzhen Huazhong University of Science and Technology Research Institute Shenzhen518063 China School of Science and Engineering The Chinese University of Hong Kong Shenzhen518172 China

Dynamic encircling of exceptional points has attracted significant interest in recent years, as it can facilitate chiral transmission selectivity due to a nontrivial eigenstate evolution. Recently, multi-state systems have been explored, associated with more complex topologies supporting a larger number of exceptional points, but chiral switching among multiple eigenstates has remained elusive in experiments. Here, we overcome this challenge by dividing the eigenstate space into multiple subspaces by controlling the adiabaticity. The eigenstates in different subspaces can evolve without crosstalk, and chiral switching occurs as the eigenstates within each subspace are subject to a non-adiabatic transition while they encircle exceptional points. We experimentally demonstrate this phenomenon by reporting chiral switching for two groups of optical modes at telecom wavelengths in a four-state optical system, and theoretically demonstrate that our approach can be extended to higher-order systems. Our findings pave new avenues for studying chiral dynamics based on exceptional-point physics in multi-state systems, and offer opportunities to develop multiplexed photonic devices. © 2024, CC BY.

关键词： Photonic devices

Enabling AI-Generated Content (AIGC) Services in Wireless Edge Networks

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Du, Hongyang Li, Zonghang Niyato, Dusit Kang, Jiawen Xiong, Zehui Shen, Xuemin Kim, Dong In The School of Computer Science and Engineering The Energy Research Institute @ NTU Interdisciplinary Graduate Program Nanyang Technological University Singapore The School of Information and Communication Engineering University of Electronic Sciences and Technology of China Chengdu China The School of Automation Guangdong University of Technology China The Pillar of Information Systems Technology and Design Singapore University of Technology and Design Singapore The Department of Electrical and Computer Engineering University of Waterloo Canada The Department of Electrical and Computer Engineering Sungkyunkwan University Korea Republic of

Artificial Intelligence-Generated Content (AIGC) refers to the use of AI to automate the information creation process while fulfilling the personalized requirements of users. However, due to the instability of AIGC models, e.g., the stochastic nature of diffusion models, the quality and accuracy of the generated content can vary significantly. In wireless edge networks, the transmission of incorrectly generated content may unnecessarily consume network resources. Thus, a dynamic AIGC service provider (ASP) selection scheme is required to enable users to connect to the most suited ASP, improving the users’ satisfaction and quality of generated content. In this article, we first review the AIGC techniques and their applications in wireless networks. We then present the AIGC-as-a-service (AaaS) concept and discuss the challenges in deploying AaaS at the edge networks. Yet, it is essential to have performance metrics to evaluate the accuracy of AIGC services. Thus, we introduce several image-based perceived quality evaluation metrics. Then, we propose a general and effective model to illustrate the relationship between computational resources and user-perceived quality evaluation metrics. To achieve efficient AaaS and maximize the quality of generated content in wireless edge networks, we propose a deep reinforcement learning-enabled algorithm for the optimal ASP selection. Simulation results show that the proposed algorithm can provide a higher quality of generated content to users and achieve fewer crashed tasks by comparing with four benchmarks, i.e., overloading-avoidance, random, round-robin policies, and the upper-bound schemes. © 2023, CC BY.

关键词： Reinforcement learning

Dynamic programming Algorithm using Furniture Industry 0/1 Knapsack Problem

学校读者我要写书评

暂无评论

Dynamic Programming Algorithm using Furniture Industry 0/1 K...

International Conference on Cybernetics and Intelligent System (ICORIS)

作者： Mietha Anditta Natashya Amartya Laurens Spits Warnars Harco Leslie Hendric Spits Warnars Arief Ramadhan Teddy Siswanto Teddy Mantoro Nurulhuda Noordin Computer Science Departement School of Computer Science Bina Nusantara University Jakarta Indonesia Information Systems Departement School of Information Systems Bina Nusantara University Jakarta Indonesia Computer Science Departement BINUS Graduate Program - Computer Science Bina Nusantara University Jakarta Indonesia PJJ Informatika Telkom University Bandung Indonesia Information Systems Department Trisakti University Jakarta Indonesia Faculty of Engineering and Technology Sampoerna University Jakarta Indonesia School of Computing Sciences College of Computing Informatics and Mathematics Universiti Teknologi MARA Selangor Malaysia

Dynamic programming is a fundamental algorithm that can be found in our daily lives easily. One of the dynamic programming algorithm implementations consists of solving the 0/1 knapsack problem. A 0/1 knapsack problem can be seen from industrial production cost. It is prevalent that a production cost has to be as efficient as possible, but the expectation is to get the proceeds of the products higher. Thus, the dynamic programming algorithm can be implemented to solve the diverse knapsack problem, one of which is the 0/1 knapsack problem, which would be the main focus of this paper. The implementation was implemented using C language. This paper was created as an early implementation algorithm using a Dynamic program algorithm applied to an Automatic Identification System (AIS) dataset.

关键词：

Fusion of Mixture of Experts and Generative Artificial Intelligence in Mobile Edge Metaverse

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liu, Guangyuan Du, Hongyang Niyato, Dusit Kang, Jiawen Xiong, Zehui Jamalipour, Abbas Mao, Shiwen Kim, Dong In The College of Computing and Data Science The Energy Research Institute @ NTU Interdisciplinary Graduate Program Nanyang Technological University Singapore The College of Computing and Data Science Nanyang Technological University Singapore The School of Automation Guangdong University of Technology China The Pillar of Information Systems Technology and Design Singapore University of Technology and Design Singapore The Department of Electrical and Computer Engineering Auburn University Auburn United States The School of Electrical and Computer Engineering University of Sydney Australia The Department of Electrical and Computer Engineering Sungkyunkwan University Korea Republic of

In the digital transformation era, Metaverse offers a fusion of virtual reality (VR), augmented reality (AR), and web technologies to create immersive digital experiences. However, the evolution of the Metaverse is slowed down by the challenges of content creation, scalability, and dynamic user interaction. Our study investigates an integration of Mixture of Experts (MoE) models with Generative Artificial Intelligence (GAI) for mobile edge computing to revolutionize content creation and interaction in the Metaverse. Specifically, we harness an MoE model's ability to efficiently manage complex data and complex tasks by dynamically selecting the most relevant experts running various sub-models to enhance the capabilities of GAI. We then present a novel framework that improves video content generation quality and consistency, and demonstrate its application through case studies. Our findings underscore the efficacy of MoE and GAI integration to redefine virtual experiences by offering a scalable, efficient pathway to harvest the Metaverse's full potential. © 2024, CC BY.

关键词： Augmented reality

Database management system design for COVID-19 symptom analysis 6

学校读者我要写书评

暂无评论

Database management system design for COVID-19 symptom analy...

6th International Conference on information Management and technology, ICIMTech 2021

作者： Rahutomo, Reza Darmalim, Ferdinan Cen, Andrew Brilianto, Jayaku Pardamean, Bens School of Information Systems Bina Nusantara University Information Systems Department Jakarta11480 Indonesia PT. Mega Medica Pharmaceuticals IT Department Jakarta Indonesia School of Computer Science Bina Nusantara University Computer Science Department Jakarta11480 Indonesia Faculty of Technology and Design Bunda Mulia University Jakarta Indonesia BINUS Graduate Program - Master of Computer Science Bina Nusantara University Computer Science Department Jakarta11480 Indonesia

ISBN: (纸本)9781665449373

Applying a digital strategy is proven to be one way to reduce the infection rate of COVID-19. As a fundamental part of it, database is capable to create a data repository to predict uncertain impacts of the pandemic. The research proposed a database design for project C-Beta that emphasized on creating an ideal system to record and monitor the daily health condition of people with COVID-19. Mainly, the research adopted Connolly and Begg principles of Database Software Development Life Cycle. As a result, the database design activities produced blueprints in the forms of data dictionary, ERD, and user views. The database design delivered in the research included with several features namely user verification, restricted research membership, and flexibility in survey creation. Based on analysis and design phase, C-Beta provides a method to collect data for COVID-19 clinical study. Researchers who utilize C-Beta is able to manage multiple surveys and extract all collected records at the same time. © 2021 IEEE.

关键词： COVID-19

5-Fold Cross Validation on Supporting K-Nearest Neighbour Accuration of Making Consimilar Symptoms Disease Classification

学校读者我要写书评

暂无评论

5-Fold Cross Validation on Supporting K-Nearest Neighbour Ac...

Computer science and Engineering (IC2SE), International Conference on

作者： Andy Satria Opim Salim Sitompul Herman Mawengkang Graduate Program of Computer Science and Information Technology Universitas Sumatera Utara Medan Indonesia Universitas Sumatera Utara Medan Indonesia

ISBN: (纸本)9781665400473

Classification is a technique in machine learning that is used to built the group of data. The data that consist of class and target are grouped based on the data attachment to the sample data, so that the data group can be used as a basis for making decisions at a later stage. This study aims to find a classification of two diseases, dengue fever and typhus which have similar symptoms using a combination of two techniques, 5-Fold Cross Validation and K-Nearest Neighbor. The combination of these techniques is done to improve the accuracy of the classification, feature selection is done in preprocessing phase to collect verified feature to get information gain to measure which feature is kept and continue to next stage of classification process. A confusion matrix will be applied to test the performance in terms of the accuracy of the constructed classification. K-Nearest Neighbor (KNN) is a classification method that applies a supervised algorithm, while 5-Fold Cross Validation is a statistical method that is applied to test the performance of a method or technique. This investigation resulted in an increase in the value of accuracy and recall, with a score of 1.79 points and a recall value of 96.43%.

关键词： Computer science Phase measurement Statistical analysis Machine learning Gain measurement Feature extraction Classification algorithms