检索结果-内蒙古大学图书馆

computer science and data engineering (csde), ieee asia-pacific conference on

作者： Priynka Sharma Kaylash Chaudhary M.G.M. Khan Michael Wagner School of Computing Information and Mathematical Sciences The University of the South Pacific Suva Suva

ISBN: (数字)9781728163031

ISBN: (纸本)9781728163048

The existence of noise in a Ransomware dataset can negatively affect the classification model constructed. More explicitly, the noisy examples in the dataset can antagonistically influence the learnt hypothesis. Eviction of noisy occurrences will improve the hypothesis; thus, improving the classification precision of the model. This paper acquaints a novel strategy through upgraded inferiority of training ransomware data with a noisy dependent variable for multiclass classification problems. Noise diminishes classification accuracy by disturbing the informational training index and setting off the classifier to assemble erroneous models. Our methodology uses a Machine Learning Fundamental Filters (MLFF) to arrange suspicious noisy examples and prototype selection (PS) in order to recognise the set of real noisy occurrences in ransomware dataset. This paper shows that the tuning of MLFF with prototype selection improves the nature of noisy training data collections; thus, increases the classification precision of the model trained with the training dataset without noise.

关键词： Noise measurement Classification algorithms Training data Ransomware Machine learning algorithms Machine learning data mining

来源：评论

学校读者我要写书评

暂无评论

Self-motivated Information Sharing in Communities for Promoting Regional Revitalization

Self-motivated Information Sharing in Communities for Promot...

引用

computer science and data engineering (csde), ieee asia-pacific conference on

作者： Ryo Shioya Mizuki Tanaka Yurika Shiozu Katsuhiko Yonezaki Ivan Tanev Katsunori Shimohara Graduate School of Science and Engineering Doshisha University Kyoto Japan Faculty of Economics Kyoto Sangyo University Kyoto Japan Department of Regional Development Taisho University Tokyo Japan

ISBN: (数字)9781728163031

ISBN: (纸本)9781728163048

A community is a system composed of “Hito,” “Mono,” and “Koto,” and their relationships, and it cannot be established without the voluntary involvement of people. In order to revitalize the local community, we have worked on a mechanism through which residents can be involved in creating better and richer relationships. For this purpose, we have built a system for collecting, quantifying, and visualizing relationships created by residents in their daily lives, and the system has been in operation in the field so far. In addition, we have introduced a mechanism for people to transmit and share awareness about regional factors and problems. As a result, we can confirm the behavior modification of residents through the said mechanism. In this study, we propose a new mechanism to promote self-motivated information sharing, which results in more effective behavior modification among people in a community. The mechanism incorporates peoples' self-centered behavior, such as walking and exercise, into an information-sharing mechanism based on the map application, with the aim to investigate whether people's self-centered behavior can have altruistic effects.

关键词： MONOS devices data visualization Legged locomotion Media Meters Economics Aging

来源：评论

学校读者我要写书评

暂无评论

Partnerships with Governments to Implement In-Country science Programmes in the South pacific Region

Partnerships with Governments to Implement In-Country Scienc...

引用

computer science and data engineering (csde), ieee asia-pacific conference on

作者： Bibhya Sharma Avinesh Prasad Swasti Narayan Bijeta Kumar Vineet Singh Salsabil Nusair Gavin Khan School of Computing Information and Mathematical Sciences The University of the South Pacific Suva Fiji Faculty of Science Technology and Environment The University of the South Pacific Suva Fiji

ISBN: (数字)9781728163031

ISBN: (纸本)9781728163048

This paper essays the design of a new cohort-based model to deliver science programmes for high school graduates and teachers teaching science in the high schools in the South pacific. The aim is to increase the number of skilled science teachers and ultimately the number of science graduates qualified to work in not only the public-private sectors but also find solutions to unique problems faced in the South pacific. Sustaining such programmes depends heavily on the culminated tripartite efforts between students, respective regional countries and higher education institutes. The University of the South pacific offers selected undergraduate degree programmes through its science Teachers Accelerated Programme (STAP) and the In-country science Programme (ISP) in its member countries. The government bodies decide upon the discipline combinations of programme based on their countries' priority areas. The first STAP launched in Samoa in 2014 had 60 science teachers enrolled in Bachelor of science with 5 different combination of majors. The partnership achieved well and in turn paved the way for STAP offers in Samoa (second cohort), Tonga and Vanuatu. Ideally, science students would travel to the main USP Campus in Fiji for further studies; however, with programmes like STAP and ISP, the university is able to deliver science to its various member countries at much lower costs. This paper will highlight the design and implementation of the new cohort based science model, together with the successes and opportunities for the future.

关键词： Cohort-based science Programmes Government Partnerships South pacific

来源：评论

学校读者我要写书评

暂无评论

Detection Of Skin Cancer Using Deep Neural Networks

Detection Of Skin Cancer Using Deep Neural Networks

引用

computer science and data engineering (csde), ieee asia-pacific conference on

作者： Md. Muzahidul Islam Rahi Farhan Tanvir Khan Mohammad Tanvir Mahtab A. K. M. Amanat Ullah Md. Golam Rabiul Alam Md. Ashraful Alam Department of Computer Science and Engineering BRAC University Dhaka Bangladesh

ISBN: (数字)9781728163031

ISBN: (纸本)9781728163048

Skin cancer is a huge issue which gets neglected very often. Sometimes the human eye is unable to precisely detect diseases from imaging data, in cases of doctor's manual inspection. In this age, we see the rise of use of deep learning methods in our daily life problem solving. Therefore, we develop an automated computerised system for detecting skin diseases using deep neural network algorithms. In the proposed model, we have used several neural network algorithms and analyse their performances to detect five major skin diseases and Figure out the best performing algorithm in terms of accuracy. CNN and by using Keras Sequential API, we have structured a new model to gainan accuracy of around 80%. Later, for comparison and also to increase accuracy we have used architectures that use pre-trained data. These transfer learning model includes VGG11, RESNET50 and DENSENET121. Among the algorithms used in the proposed models, resnet architecture achieve highest accuracy of 90%.

关键词： Skin Neural networks Training Melanoma

来源：评论

学校读者我要写书评

暂无评论

A cloud based architecture for hosting ECG arrhythmia data classification service

引用

AIP conference Proceedings 2023年第1期2764卷

作者： S. Nagarajan V. Nivaskumar M. Vanitha Lakshmi C. Senthilkumar S. Sathish Kumar M. Moorthi Department of Electronics and Communication Engineering SRM Institute of Science and Technology Ramapuram Campus Chennai Tamil Nadu India Department of Electronics and Communication Engineering Jawahar Engineering College Chennai India Department of Electronics and Communication Engineering Saveetha School of Engineering Chennai India Department of Electronics and Communication Engineering Asia Pacific University of Technology and Innovation Kualalumbur Malaysia Department of Medical Electronics Saveetha Engineering College Chennai India

Healthcare industry is in demand to adopt the new technologies that are in the market to improve their service quality. The telecommunication system is integrated with computing, Inter-networking, mobility, data storage, data analytics IoT (Internet of Things) centred technology is the order of the day. In this research work, our objective is to develop software that integrates the cloud computing and mobile technology for health care monitoring system. In present research, the stable fractal values are extracted from the ECG signals using Higuchi’s method, which was not attempted by any researcher in the field of developing a computer aided diagnosis system for arrhythmia. From the results, the maximum classification accuracy for fractal features has been registered by support vector machine. 92.08% and 90.36%. The comparative evaluation was performed on 1) cloud -two node cluster 2) cloud-Ten node cluster. We presented software that integrates mobile computing and cloud computing for analyzing the ECG arrhythmia for five different arrhythmia classes. a) Class N b) Class S c) Class V d) Class F e) Class Q. This work proposed to evaluate cloud based architecture 1) cloud -two node cluster 2) cloud-Ten node cluster. Where the multi nodes cluster computing reduces the execution to almost to 50 per cent and maintains the response time. Dynamic resource allocation frame work has ability to distribute the load and maintain the time response.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Classification in a Distributed System - A Study of Random Forest in the Hadoop MapReduce Framework

Classification in a Distributed System - A Study of Random F...

引用

computer science and data engineering (csde), ieee asia-pacific conference on

作者： Kunal Kumar Neeraj Anand Sharma A B M Shawkat Ali Department of Computer Science And Mathematics The University of Fiji Saweni Lautoka City Fiji School of Engineering and Technology CQUniversity Melbourne Campus Australia

ISBN: (数字)9781728163031

ISBN: (纸本)9781728163048

Due to the massive increase in the amount of data collected and stored globally, new issues of data processing and algorithm compatibility has come to light. Processing large datasets is a challenge for commodity systems due to its limited processing capabilities while specialized and more powerful processing systems are costly to set up and use. This research will focus on a distributed system known as the Hadoop Framework that uses commodity machines to create a combined and more powerful system that is able to process big datasets much more efficiently. The Hadoop Framework processes data in a distributed and parallel environment which is unlike the processing mechanisms used in a traditional commodity computer system. However, this change in the nature of processing demands a change in the underlying structure of the machine learning algorithms that process the data. The research will delve at implementing the Random Forest algorithm in the Hadoop Framework. The Random Forest algorithm will be executed on a four-node Hadoop Cluster inputting data of varying sizes at a time. Performance of the algorithm will be measured by analyzing execution Time, Accuracy, Kappa, Reliability and Standard Deviation measures. Results will show that the execution time reduces constantly as the size of dataset increases thus making the system a key tool for processing large datasets.

关键词： hadoop map-reduce mahout classification random forests Taxonomy Distributed Systems dataset execution time Frameworks algorithms Machine learning algorithms Random

来源：评论

学校读者我要写书评

暂无评论

Optimized RTPS platform for IIoT based smart farming systems

Optimized RTPS platform for IIoT based smart farming systems

引用

computer science and data engineering (csde), ieee asia-pacific conference on

作者： Basem Almadani Saud Mohammad Mostafa Computer Engineering Department King Fahd University of Petroleum and Minerals Dhahran Saudi Arabia

ISBN: (数字)9781728163031

ISBN: (纸本)9781728163048

With the ever-growing population, food and water demands have risen exponentially. Almost 80-85% of food is produced by agriculture but the rise in demand needs to be met by an increase in food production. The agriculture sector tends to tackle this issue by incorporating the Industrial Internet of Things (IIoT) and cloud technologies to improve productivity and reduce wastage of resources needed for it. The prime target of incorporation of these technologies would be to facilitate farmers to grow crops and raise livestock with ease and lower cost with increasing profits. Smart farming systems can comprise of one or more sub-systems that are automated using IIoT technology and processing is done either in local server or cloud. Most researchers work on individual sub-system and improve the performance providing local optimization. These sub-systems can have different vendors and integration is quite difficult at times. Therefore, we propose to use data Distribution Service (DDS) middleware to integrate these heterogeneous sub-systems and provide high performance with low latency system. In this paper, we introduce a novel approach to tackle the issue of heterogeneity and provide global optimization. A suitable mathematical model for our proposed system is given and experiments are conducted to show the efficiency of the system in terms of throughput and delay.

关键词： Agriculture Servers Robot sensing systems Logic gates Temperature measurement Production Cloud computing

来源：评论

学校读者我要写书评

暂无评论

Analysing computer science Course Using Learning Analytics Techniques

Analysing Computer Science Course Using Learning Analytics T...

引用

computer science and data engineering (csde), ieee asia-pacific conference on

作者： Sanjay Jha Meena Jha Liam O’Brien Central Queensland University Sydney NSW Australia Home Affairs Canberra ACT Australia

ISBN: (数字)9781728163031

ISBN: (纸本)9781728163048

The academic community has identified some critical issues facing the sector and are trying to Figure out how these can be addressed. One of these issues is that science, Technology, engineering, and Mathematics (STEM) fields have notoriously low retention rates. According to 2016 report on Australia's STEM workforce only 32% are university qualified. LA has become an emerging research field where knowledge is extracted and patterns are discovered from e-learning systems. LA provides a set of techniques which can help improve the retention rate by identifying students' needs, and analysing students' learning behaviour patterns. The objective of our research is to introduce a data-driven investigation to determine students' learning behaviour in a computer science course. Our research uses a variety of techniques such as Clustering (K-Means), and Classification (Decision Tree Induction) in order to achieve the goal to discover data-driven knowledge from the Moodle Learning Management System (LMS) and the Student Information System (SIS). The analysis was carried out using log data obtained from Moodle and student's information obtained from the SIS in a computer science course. The study collected demographic profiles of students studying a course in Information Systems and Analysis and compared them to find answers to questions such as how student's engagement to online activities and accessing learning resources impact students' success rate and hence improving retention rate by identifying students need on time. For this work, Rapid Miner data mining tool was used to mine and analyse data from the SIS and the Moodle system.

关键词： Xenon

来源：评论

学校读者我要写书评

暂无评论

A Prototype of a Secured File Storing and Sharing System for Cloud Storage Infrastructure

A Prototype of a Secured File Storing and Sharing System for...

引用

computer science and data engineering (csde), ieee asia-pacific conference on

作者： Md. Shahadul Alam Patwary Rajesh Palit Electrical and Computer Engineering North South University Dhaka Bangladesh

ISBN: (数字)9781728163031

ISBN: (纸本)9781728163048

With the increased popularity of third-party cloud storage services, privacy has become a major concern. As cloud storage services can be availed at a minimum price now-a-days, it comes with a greater cost, vulnerability of being accessed by anyone else. To ensure reliability, most of the cloud service providers keep multiple copies of files. This becomes a serious privacy issue when the replicated copies remain in the server even after the owner has deleted the original file. To solve this problem, several models such as FADE, MADS, SFADE+ etc. were proposed that ensures assured file deletion but has limitations of their own. We have designed a system using the existing cryptographic approaches which ensures assured deletion of files resolving the limitations that remained un-patched by the previous models. The primary focus of our proposed model is to ensure user privacy and security through a user-friendly system which is computationally less intensive and does not rely upon any trusted entity. For factual exhibition, we have implemented a prototype named CryptedCloud which assures file deletion without relying upon any trusted body with no compromise to the performance and user-friendliness.

关键词： file deletion Ensure Privacy PROTOTYPE copying

来源：评论

学校读者我要写书评

暂无评论

Using Graph Convolution Network for Predicting Performance of Automatically Generated Convolution Neural Networks

Using Graph Convolution Network for Predicting Performance o...

引用

computer science and data engineering (csde), ieee asia-pacific conference on

作者： Enzhi Zhang Tomohiro Harada Ruck Thawonmas Graduate School of Information Science and Engineering Ritsumeikan University Kusatsu Shiga Japan College of Information Science and Engineering Ritsumeikan University Kusatsu Shiga Japan

ISBN: (数字)9781728163031

ISBN: (纸本)9781728163048

In this paper, we propose a model using a graph convolution network for predicting the accuracy of the automatically generated convolution neural network (CNN). In recent years, there have been many excellent results in the automatic designing of deep learning network structures and their hyperparameters. These works mainly use evolutionary algorithms and reinforcement learning methods. But to verify the accuracy of newly generated neural network samples, it requires a lot of computational time and large scale of hardware. Therefore, how to reduce training time and improve the utilization of samples has become an important problem of automatic design for deep learning networks. On the other hand, the effectiveness of the graph convolution network (GCN) has been shown in processing graph data on the tasks of graph node classification, link prediction, and clustering. We are considering using GCN to predict the accuracy of the automatically designed neural networks. We compare GCN, a multi-layer neural network(NN), and CNN on the accuracy prediction task of automatically generated CNN for image classification tasks on the CIFAR-10 dataset. Our experiment demonstrates the advantages of GCN on small datasets compared to the multi-layer neural network and CNN, while GCN achieves the equivalent accuracy to the other methods on large datasets.

关键词： Training Convolution Predictive models Computational modeling Feature extraction Artificial neural networks

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：