检索结果-内蒙古大学图书馆

作者： Nishiyama, Daiki Fukuchi, Kazuto Akimoto, Youhei Sakuma, Jun Master's Program in Computer Science Degree Programs in Systems and Information Engineering Graduate School of Science and Technology University of Tsukuba Tsukuba-shi305-8577 Japan RIKEN Center for Advanced Intelligence Project Tokyo103-0027 Japan Department of Computer Science Faculty of Engineering Information and Systems University of Tsukuba Tsukuba-shi305-8577 Japan

In real world applications of multiclass classification models, misclassification in an important class (e.g., stop sign) can be significantly more harmful than in other classes (e.g., no parking). Thus, it is crucial to improve the recall of an important class while maintaining overall accuracy. For this problem, we found that improving the separation of important classes relative to other classes in the feature space is effective. Existing methods that give a class-sensitive penalty for cross-entropy loss do not improve the separation. Moreover, the methods designed to improve separations between all classes are unsuitable for our purpose because they do not consider the important classes. To achieve the separation, we propose a loss function that explicitly gives loss for the feature space, called class-sensitive additive angular margin (CAMRI) loss. CAMRI loss is expected to reduce the variance of an important class due to the addition of a penalty to the angle between the important class features and the corresponding weight vectors in the feature space. In addition, concentrating the penalty on only the important class hardly sacrifices separating the other classes. Experiments on CIFAR-10, GTSRB, and AwA2 showed that CAMRI loss could improve the recall of a specific class without sacrificing accuracy. In particular, compared with GTSRB's second-worst class recall when trained with cross-entropy loss, CAMRI loss improved recall by 9%. © 2023 The Institute of Electronics, Information and Communication Engineers.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Tree Counting with Deep Forest Algorithm for Kulon Progo District in Yogyakarta, Indonesia Using Pleiades Satellite Imagery

Tree Counting with Deep Forest Algorithm for Kulon Progo Dis...

引用

ICT Conference (AICTC), Arab

作者： David Wijaya Ibrahim Seto Aditama Stanley Armando Austen Calvin Surya Widjaja Bakti Amirul Jabar Edy Irwansyah Computer Science Department School of Computer Science Bina Nusantara University Jakarta Indonesia Computer Science Department School of Computer Science Bina Nusantara University Tangerang Selatan Indonesia Computer Science Department School of Computer Science Bina Nusantara University Bandung Indonesia BINUS Graduate Program Master of Computer Science Bina Nusantara University Jakarta Indonesia

ISBN: (数字)9798350342086

ISBN: (纸本)9798350342093

Trees are one of the most important living things on the planet. Trees help by being one of the largest producers' oxygens on planet, by absorbing groundwater and by maintaining soil fertility. Trees can also be one of the keys to reducing existing air pollution. Therefore, the number of existing trees must be maintained. The first step in maintaining the number of existing trees is to count or map them. One of the easiest ways is to do this use an algorithm. This study uses a tree-counting algorithm called Deep Forest. Deep Forest is a Python package that can detect trees through RGB satellite imagery. This study used satellite imagery from Pleiades on an area in Kulon Progo district, Yogyakarta, Indonesia for the image-one of the four regencies within Yogyakarta Province in Indonesia. In addition to the Deep Forest algorithm, this research also used an application provided by Esri called ArcGIS Pro. With this application, data preprocessing such as labeling the data and creating the deep learning dataset for research is much easier, while ArcGIS Pro provides tools for labeling and exporting the data to jupyter notebook. The result obtained using the Deep Forest algorithm is the F1 score with an accuracy of 0.76 for the first experiment, 0.774 for the second experiment, and 0.779 for the final experiment.

关键词： Deep learning Training Codes Planets Vegetation Soil Regression analysis Satellite images Labeling Random forests

来源：评论

学校读者我要写书评

暂无评论

Single-Phase Transformerless UPS with Interleaved Boost Converters for Battery Discharging

Single-Phase Transformerless UPS with Interleaved Boost Conv...

引用

2023 IEEE Energy Conversion Congress and Exposition, ECCE 2023

作者： Pinto, Alan R. A. Bahia, Filipe A. C. Souza, Reuben P. R. Pinheiro, José R. Costa, Fabiano F. Federal University of Bahia Post-Graduate Program in Electrical Engineering - Ppgee Salvador Brazil Federal University of Bahia Dept. of Electrical and Computer Engineering - Deec Salvador Brazil Federal University of Campina Grande Dept. of Electrical Engineering - Dee Campina Grande Brazil

ISBN: (纸本)9798350316445

This paper proposes a reconfiguration for a single-phase, double conversion uninterruptible power supply (UPS). With the addition of two static switches to the original circuit, the UPS mitigates the ripple in the current, through battery discharge performed by two legs. Other than these switches, the circuit topology is preserved. The preliminary results show a satisfactory level of ripples for the inductor current and suitable power quality for the output voltage. © 2023 IEEE.

关键词： battery mode discharger interleaved boost normal mode

来源：评论

学校读者我要写书评

暂无评论

Simple Fuzzy Decision Support Model for Evaluating the Cryptocurrency’s Performance

Simple Fuzzy Decision Support Model for Evaluating the Crypt...

引用

Science and Information Technology in Smart Administration (ICSINTESA), International Conference of

作者： Ditdit Nugeraha Utama Computer Science Department BINUS Graduate Program – Master of Computer Science Bina Nusantara University Jakarta Indonesia

ISBN: (纸本)9781665472890

Evaluation of cryptocurrency’s performance is performed questionably. There is no role for computer-model, make such an evaluation process does not have guidance. In this study, a simple decision support model (DSM) based on fuzzy logic was academically constructed to observe the cryptocurrency’s performance. By operating the primary method of fuzzy logic and taking into account three types of parameters (i.e. time-series close price data, daily max-min price, and transaction number), a novel DSM for evaluating the cryptocurrency’s performance was fruitfully executed. Based on three types of real cryptocurrency six-month data (i.e. Bitcoin Ethereum, and Dogecoin), the model could irreversibly expose that Bitcoin has the best performance with 36.39 performance points.

关键词： Fuzzy logic Computational modeling Bitcoin Minimax techniques Information technology

来源：评论

学校读者我要写书评

暂无评论

Risk Assessment Analysis on Mobile Banking Using Cobit 5 Framework

Risk Assessment Analysis on Mobile Banking Using Cobit 5 Fra...

引用

Creative Communication and Innovative Technology (ICCIT), IEEE

作者： Joy Nashar Utamajaya Arief Ramadhan Edi Abdurachman Agung Trisetyarso Muhammad Zarlis Computer Science Department BINUS Graduate Program – Computer Science Bina Nusantara University West Jakarta Indonesia

In general, public or private organizations or companies have used information-based technology as a support to improve business performance to be more effective and efficient in order to achieve a company's business goals. Banking companies include companies that use information technology such as Mobile Banking in their services. Mobile banking is a banking facility or service using mobile communication tools such as mobile phones, with the provision of facilities for banking transactions through mobile applications. Even though they have used a good SOP (Standard Operating Procedure), there are still many obstacles that occur, especially system problems and human errors. This can result in a high risk if it occurs continuously and will be fatal to the company's business processes. Therefore, COBIT 5 can be used as a benchmark for the IT assessment process on local bank mobile banking services. The purpose of the study is to determine the level of capability and strategy for improving risk management. The method used in this study is to focus on IT risk management with the COBIT 5 framework in the Evaluate Direct Monitoring (EDM) 3 domain, Align Plain Organize (APO) 12. From the results of the analysis, the calculation of the capability level is at Level 1 Performed Process with a score of 82.04%, thus the status has evidence as well as a systematic approach and this achievement is obtained significantly through the assessment of process attributes.

关键词： Systematics Standards organizations Banking Companies Mobile communication Mobile handsets Risk management

来源：评论

学校读者我要写书评

暂无评论

Towards a Novel Edge AI System for Particle Size Detection in Mineral Processing Plants 25

Towards a Novel Edge AI System for Particle Size Detection i...

引用

25th International Conference on Enterprise Information Systems, ICEIS 2023

作者： Cardoso, Flávio Wellb Silva, Mateus Coelho de Meira, Natália F.C. Oliveira, Ricardo Augusto Rabelo Campos Bianchi, Andrea G. Graduate Program in Instrumentation Control and Automation of Mining Processes Instituto Tecnológico Vale Federal University of Ouro Preto Ouro Preto Brazil VALE S.A. Nova Lima Brazil Department of Computer Science Federal University of Ouro Preto Ouro Preto Brazil

ISBN: (纸本)9789897586484

Monitoring and controlling the particle size is essential to reducing the variability and optimizing energy efficiency in mineral process plants. The industry standard utilizes laboratory processes for particle size characterization;the problems that arise here are obtaining representative sample from the bulk and finding a rapid method of particle size assessment. We propose a machine vision concept based on Edge AI architecture and deep convolutional neural algorithms to enable a real-time analysis of particle size, as an alternative to offline laboratory process. The present paper is part of this proposed concept and aims exclusively to validate a deep convolutional neural network algorithm trained from synthetic datasets. The proposed model reached a mean Average Precision (mAP) of 0.96 and processing times of less than 1s. The results demonstrate the feasibility of deep convolutional neural networks for real-time particle size segmentation and establishes the first step towards a novel Edge AI system for particle size measurement in mineral processing plants. Copyright © 2023 by SCITEPRESS - Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

关键词： Particle size

来源：评论

学校读者我要写书评

暂无评论

Low-Resolution Face Recognition for CCTV and Edge-Powered Smart Attendance Systems

Low-Resolution Face Recognition for CCTV and Edge-Powered Sm...

引用

IEEE Annual International computer Software and Applications Conference (COMPSAC)

作者： Nico Surantha Edward Yose Sani Muhamad Isa Department of Electrical Electronics and Communication Engineering Faculty of Science and Engineering Tokyo City University Tokyo Japan Computer Science Department BINUS Graduate Program-Master of Computer Science Bina Nusantara University Jakarta Indonesia

ISBN: (数字)9798350376968

ISBN: (纸本)9798350376975

Attendance systems have become more modern, and one of the biometric systems without physical contact is face recognition. However, many face-based attendance systems still carry out attendance individually and cannot detect multiple faces simultaneously. In addition, capturing facial data in real-time is still a challenge because the relatively large distance between the camera and the individual reduces the ability to recognize faces. The general solution is to use super-resolution to generate better-quality faces while maintaining the main facial recognition features. One technique still being researched is super-resolution generative adversarial networks (SRGAN). SRGAN can enlarge the resolution of captured images and maintain image quality sufficient for face recognition. The attendance system can be easily integrated into edge devices such as the Jetson Nano. This paper proposes automatic and effective attendance systems with the super-resolution technique to detect and recognize faces in low-resolution input. The experimental results show that using face data capture with a resolution of 40 × 40 pixels and a four-fold magnification results in a resolution of 160 × 160 pixels. Combining Face SRGAN with FaceNet architecture as the basis of face recognition can achieve an accuracy rate of 78.19% and an F1-Score of 81.13% with an average processing time of 1.61 seconds per frame on a PC and 14.55 seconds per frame on a Jetson Nano at an average of face recognition per frame of as many as up to 8 faces simultaneously.

关键词： Support vector machines Image quality Accuracy Face recognition Image edge detection Superresolution Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Analysis of Several Quantum Encoding Methods Implemented on A Quantum Circuit Architecture to Improve Classification Accuracy

Analysis of Several Quantum Encoding Methods Implemented on ...

引用

International Conference on Electrical, Telecommunication and computer Engineering (ELTICOM)

作者： Rahmad Fauzi Muhammad Zarlis Herman Mawengkang Poltak Sihombing Graduate Program of Computer Science Faculty of Computer Science and Information Technology Universitas Sumatera Utara Medan Indonesia

ISBN: (纸本)9798350320725

Quantum encoding is a process to transform classical information into quantum states. It plays a crucial role in using quantum algorithms to solve classical problems, especially in quantum machine learning tasks. There are many QE methods. It is very difficult to determine which QE method to choose to improve classification accuracy. Therefore, this paper will analyze several QE methods. Training and testing on Iris flower datasets were performed in a architecture quantum circuit and some performances parameters were evaluated. The expected result is that we can compare the classification accuracy of some of these Quantum encodings.

关键词： Training Quantum computing Quantum algorithm computer architecture Transforms Quantum state Encoding

来源：评论

学校读者我要写书评

暂无评论

Systematic Literature Review on Statistics and Machine Learning Predictive Models for Rice Phenotypes

引用

Procedia computer Science 2023年 227卷 1054-1061页

作者： Nicholas Dominic Tjeng Wawan Cenggoro Bens Pardamean Bioinformatics and Data Science Research Center Bina Nusantara University Jakarta 11480 Indonesia Computer Science Department School of Computer Science Bina Nusantara University Jakarta 11480 Indonesia Computer Science Department BINUS Graduate Program – Master of Computer Science Bina Nusantara University Jakarta 11480 Indonesia

Predicting the best-quality of rice phenotypes is the priority among agricultural researchers to fulfill worldwide food security. Trend development of predictive models from statistics to machine learning is the subject of this review. Gathered from the Google Scholar database, 14 appropriate papers (2016-2020) related to the rice phenotypes prediction were selected through title and abstract content filtering. The outputs show that Support Vector Machine, Multi-layer Perceptron, and regression are the most used models, while yield is the priority prediction point besides tiller, panicle, and 1000-grain weight of rice. However, finding the accurate predictor is invariably challenging due to distinct rice varieties in the world and high confounding factors. Thus, developing an advanced deep learning model that accommodates these needs is worth considering further.

关键词： rice phenotypes prediction statistics machine learning

来源：评论

学校读者我要写书评

暂无评论

UML Design for Decision Support Model in Determining the Sustainability of Online Course Materials

UML Design for Decision Support Model in Determining the Sus...

引用

International Conference on Information Technology, Information Systems and Electrical Engineering (ICITISEE)

作者： Erna Fransisca Angela Sihotang Afdhal Kurniawan Ditdit Nugeraha Utama Computer Science Department BINUS Graduate Program Master of Computer Science Bina Nusantara University Jakarta Indonesia

The scenario of online learning is a very urgent need in the world of future knowledge. Since the Corona Virus Disease-19 pandemic, the world economy has started to plummet and caused many adults to lose their jobs. The advantage is the flexibility and rapid development of the internet. In 2020, the number of unemployed increased significantly. This reason makes people strive to improve their ability to meet job requirements by taking online courses. Online courses are a way that people can choose to improve their skills anywhere and anytime. The sustainability of online course material that is offered to the course user and issued by the company will be discussed in this study. The novelty of this research is to obtain a decision support model based on fuzzy logic for determining online courses. The method used is decision-making based on UML and fuzzy logic for the final decision. The fuzzy inference model process begins by determining the decision parameters then using fuzzification with absolute input then refracted with fuzzy criteria, and ends with defuzzification with absolute output. There are two groups of parameters in this study, company profits which consist of 5 parameters and user benefits, which consist of 9 parameters. Once the model is verified and valid, the final decision is useful for users looking for online course and also useful for the decision unit of online course companies in determining the sustainability of online course materials.

关键词： Fuzzy logic Electrical engineering Pandemics Unified modeling language Decision making Companies Sustainable development

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：