检索结果-内蒙古大学图书馆

Dynamic Sampling-Based Meta-Learning Using Multilingual Acoustic Data for Under-Resourced Speech Recognition

IEEE Access 2024年 12卷 106070-106083页

作者： Hsieh, I-Ting Wu, Chung-Hsien Zhao, Zhe-Hong National Cheng Kung University Graduate Program of Multimedia Systems and Intelligent Computing Tainan70101 Taiwan National Cheng Kung University Department of Computer Science and Information Engineering Tainan70101 Taiwan

Under-resourced automatic speech recognition (ASR) has become an active field of research and has experienced significant progress during the past decade. However, the performance of under-resourced ASR trained by existing methods is still far inferior to high-resourced ASR for practical applications. In this paper, speech data from languages that share the most phonemes with the under-resourced language are selected as supplementary resources for meta-training based on the Model-Agnostic Meta-Learning (MAML) strategy. Besides supplementary language selection, this paper proposes a dynamic sampling method instead of the original random sampling method to select support and query sets for each task in MAML to improve meta-training performance. In this study, Taiwanese is selected as the under-resourced language, and the speech corpus of five languages, including Mandarin, English, Japanese, Cantonese, and Thai, are chosen as supplementary training data for acoustic model training. The proposed dynamic sampling approach uses phonemes, pronunciation, and speech recognition models as the basis to determine the proportion of each supplementary language to select helpful utterances for MAML. For evaluation, with the selected utterances from each supplementary language for meta-training, we obtained a Word Error Rate of 20.24% and a Syllable Error Rate of 8.35% for Taiwanese ASR, which were better than the baseline model (26.18% and 13.99%) using only the Taiwanese corpus and other methods. © 2013 IEEE.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

Key Performance Indicators of Corporate Information systems Strategic Planning for Competitive Advantage 5

Key Performance Indicators of Corporate Information Systems ...

引用

5th International Conference on Advancement in Data Science, E-Learning and Information systems, ICADEIS 2023

作者： Sardjono, Wahyu Bina Nusantara University BINUS Graduate Program - Master of Information Systems Management Information Systems Management Department Jakarta11480 Indonesia

ISBN: (纸本)9798350303414

This study aims to identify the determinants of information system strategic planning in companies that are considered not in line with corporate strategic planning. There is a discrepancy between performance targets and their achievement, it is urgently necessary to carry out an evaluation related to the performance indicators used and the method of measuring performance. So that the long-term targets that are not achieved, the annual targets for the current year, can be identified with certainty the causal factors. Once these factors are identified, a model that succeeds in achieving the strategic goals developed by the company will provide steps for the company that can be used as a reference in the future. A quantitative approach through a questionnaire as a research instrument was carried out and data processing used factor analysis and regression analysis. The results obtained indicate that there are three factors that affect alignment between corporate strategic planning and corporate information system strategic planning, namely new factors related to the leadership model & company support, the target of achieving business performance, and awareness of understanding company policies and directives. The existence of this new factor is important for the success of companies engaged in the management of clean water and sanitation because it achieves alignment between the company's planning strategy and the company's information system planning strategy and can be modeled. © 2023 IEEE.

关键词： Factor analysis

来源：评论

学校读者我要写书评

暂无评论

Image Classification of the Fertility Level of Chili Using Convolutional Neural Network 8

Image Classification of the Fertility Level of Chili Using C...

引用

8th International Conference on Information Management and Technology, ICIMTech 2023

作者： Salim, Richard Fajar, Ahmad Nurul Binus Graduate Program - Master of Information Systems Management Bina Nusantara University Information Systems Management Department Jakarta11480 Indonesia

ISBN: (纸本)9798350326093

Potential exports from Indonesia's economy include agricultural products. Red chilies, especially the curly red chili, are one possible horticultural product that might be produced. For the longest time, the process of sorting chili by various firms in the processing sector, chili exporters, and chili farmers has traditionally been done manually and includes several human resources as the decision-maker. The process of sorting chili manually has many disadvantages, such as the long time required, human visual limitations, and the psychic conditions of the observer. The development of machine learning image processing technology makes it possible to sort agricultural products and plantations automatically. One of the deep machine learning methods techniques is VGG16 models from the Convolutional Neural Network method which currently has the most significant result in image recognition is VGG16 models. The classification accuracy value obtained is 93.3% with kernel size 3x3 and batch size 32 at epochs 50 using the RGB input image. © 2023 IEEE.

关键词： Agricultural products

来源：评论

学校读者我要写书评

暂无评论

Customer Clustering and Classification Based on Location and Online Store Transaction: a Case Study 5

Customer Clustering and Classification Based on Location and...

引用

5th International Conference on Cybernetics and Intelligent systems, ICORIS 2023

作者： Mauritsius, Tuga Dhimathera, Artby Susilo, Arief Ageng Dwi Asima Purba, Yohana Bina Nusantara University Binus Graduate Program - Master of Information Systems Management Information Systems Management Department Jakarta11480 Indonesia

ISBN: (纸本)9798350369489

XYZ Store is an online retailer that offers Rhea Health Tone (RHT), a supplement that helps build immune system strength. Until now, the shop has never contemplated grouping customers and distinguishing successful transactions from failed ones. The research aims to categorize consumers to determine which are purchasing RHT from local retail outlets. It is possible to create a distinction between successful and failed transactions. We have got two datasets to tackle. Tokopedia, an Indonesian marketplace, shared transaction data from its platform spanning three months and 26 days. The second dataset provides additional details, such as the outcomes of the transactions and their dates between September 2020 and June 2021. The purchase data is segmented into five data sets depending on the different locations where data is collected, and the characteristics of the customer base are determined through cluster analysis. This study will help to make market tactics more adaptable via the use of its findings. Based on the results classifications, we figured out that our data had been prepared poorly, which caused an imbalance in data. A better result, more accuracy, and a higher AUC might have been achieved had we prepared adequately. © 2023 IEEE.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

Real Time Tracking System Integrated with IoT for Package Delivery in Indonesia's Logistic Company with TOGAF Implementation 10

Real Time Tracking System Integrated with IoT for Package De...

引用

10th International Conference on ICT for Smart Society, ICISS 2023

作者： Wang, Gunawan Listya, Maria Alianto, Hendra Nugroho, D'Mitry Adam BINUS Graduate Program-Master of Information Systems Management Bina Nusantara University Information Systems Management Department Jakarta11480 Indonesia

ISBN: (纸本)9798350339543

Indonesia is a country with the highest e-commerce adoption in Southeast Asia with a contribution of up to 50%. The growth of e-commerce has a domino effect on other industries such as the shipping industry (logistic). During the Covid19 pandemic, the purchases within e-commerce have increased tremendously. However, due to the large-scale social restriction and rules imposed by the government, the process of sending packages took longer than the standard delivery. With the increasing number of shipments and so do the shipping problems. This occurs due to the lack of shipping updates that provided by logistics and e-commerce for the buyers. The proposal utilizes the use of Internet of Things (IoT) tracking device to eliminate mishandling delivery purposes. Customers enable to track their packages in real-time and speed up the claims immediately with the couriers or return the product to the nearest dropping point © 2023 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Designing Service Oriented Architecture Model in Sehatin Application with a Domain Driven Design Approach 8

Designing Service Oriented Architecture Model in Sehatin App...

引用

8th International Conference on Information Management and Technology, ICIMTech 2023

作者： Legowo, Nilo Erin Lee, Eugenius Hansel Djakaria, Merryta Bina Nusantara University Jakarta Binus Graduate Program - Master of Information Systems Management Information Systems Management Department 11480 Indonesia

ISBN: (纸本)9798350326093

Technology is required to be a solution in dealing with various fields that are negatively affected by the spread of the Covid-19 virus, one of which is in the health sector. There is a change in people's behavior who previously did not exercise to Started the habit of exercising during the pandemic and exercising at home (home workout) became the most popular type of exercise done. There are so many Home workout Application has been developed with their unique features but this makes people have to open many applications according to their needs so it is less effective. This research aims to design service oriented architecture model for home workout applications in this case Call a 'sehatin' application, that can help people to use various features in one integrated platform, The method used in designing the service oriented architecture (SOA) model uses the Domain Driven Design (DDD). The results obtained are the integration of several integrated home workout applications, the system can provide services according to the applications needed, so that the applications can be used by the public to access data service in a situation more efficient. © 2023 IEEE.

关键词： Design

来源：评论

学校读者我要写书评

暂无评论

Customer Segmentation Based On RFM: A Case Study in the context of Pandemic 5

Customer Segmentation Based On RFM: A Case Study in the cont...

引用

5th International Conference on Cybernetics and Intelligent systems, ICORIS 2023

作者： Mauritsius, Tuga Turmawan, Andrew Herdiansyah, Herdiansyah Bina Nusantara University Binus Graduate Program - Master of Information Systems Management Information Systems Management Department Jakarta11480 Indonesia

ISBN: (纸本)9798350369489

Any pandemic such as Covid-19 forces government's policy to limit people's mobility. It impacted to motorcycle workshop's operational and delaying customers in maintaining their motorcycle quality. Overcoming the situation, workshop business has to choose high value of their customers to optimize their service capacity. This can be done by using customer segmentation. Using segmentation approach, it becomes easier to provide services to targeted specific people with fit offering service package. This study presents customer segmentation using Recency, Frequency and Monetary model. K-Means Clustering is used to study motorcycle customer behavior and validated using Davies Bouldin Index. The clustering result can be used to target high potential customers and deliver motorcycle service for them. © 2023 IEEE.

关键词： Motorcycles

来源：评论

学校读者我要写书评

暂无评论

Credit Risk Prediction System For MSME Loan Process 8

Credit Risk Prediction System For MSME Loan Process

引用

8th International Conference on Information Management and Technology, ICIMTech 2023

作者： Purwadi, Ignasius Kenny Bagus Wirahardi, Adhi Hutasoit, Andrew Franico Mauritsius, Tuga Bina Nusantara University Binus Graduate Program - Master of Information Systems Management Information Systems Management Department Jakarta11480 Indonesia

ISBN: (纸本)9798350326093

BPR XYZ is one of the BPR (Rural Bank) located in West Java - Indonesia, that has two primary businesses, which are: (1) collecting funds from customers in the forms of savings and deposits and (2) providing loans to customers. BPR XYZ successfully disbursed approximately more than IDR 300 billion micro-loans to the Micro, Small and Medium Enterprise (MSME) segment and will keep increasing gradually in 2022. Their Non-Performing Loan (NPL) rate in 2021 was above 5% (five percent), more significant than the national standard, and most likely to increase in 2022. This happens because many of their MSME customers cannot pay the debt. This study will conduct predictive analysis using Naïve Bayes and K-Nearest Neighbors algorithm (K-NN) to predict the NPL. From our experiments using historical data and five classes of NPL, we found that naïve bayes do not perform well, with average accuracy just only 19%;meanwhile, K-NN obtains a performance of 59%. Random Forest gained 52%, and SVM with 48%. We tried to restructure the NPL classification label to only two classes with these results. We got better results where K-NN algorithm gained the best performance with an accuracy level of 74%. © 2023 IEEE.

关键词： Sales

来源：评论

学校读者我要写书评

暂无评论

Introducing Cloud-Based Performance Monitoring Technology for Industrial Facilities that Keeps up with the 4.0 Industrial Revolution 10

Introducing Cloud-Based Performance Monitoring Technology fo...

引用

10th International Conference on Smart Computing and Communication, ICSCC 2024

作者： Sardjono, Wahyu Maryani Retnowardhani, Astari BINUS Graduate Program - Master of Information Systems Management Bina Nusantara University Information Systems Management Department Jakarta11480 Indonesia Bina Nusantara University School of Information Systems Information Systems Department Jakarta11480 Indonesia

ISBN: (纸本)9798350363104

This study examines the state of affairs now and what lies ahead in terms of technology from the viewpoint of pilot projects, which are crucial in promoting the adoption of Industry 4.0. The program culminates in the development of a dynamic platform designed to evaluate and contrast many customized solutions that cater to the distinct needs of enterprises. Industry 4.0 developments present a plethora of design options, but businesses must develop internal solutions that best fit their requirements, taking into account factors like safety, affordability, and the possibility of requiring specialist knowledge that may need to be outsourced. The platform, which is now in use, has a cloud-based system for tracking control loop performance (CLPM), which serves as an all-inclusive communication module that efficiently collects and manages data. The document provides specific technical information about the pilot-scale module's implementation, providing an overview of ongoing work. The project's scope has expanded to encompass the system's integration of new capabilities, including live data analysis, job supervision, and a CLPM scheme that can handle factories with dispersed operations across several regions. As such, the discussion delves into the platform's incremental evolution, emphasizing the progress made in the technologies utilized. © 2024 IEEE.

关键词： 4.0 industrial revolution control loop performance industrial facilities performance monitoring presenting cloud-based

来源：评论

学校读者我要写书评

暂无评论

Building Mathematical Models for Company Strategy Formulation in Implementing Knowledge Management systems 10

Building Mathematical Models for Company Strategy Formulatio...

引用

10th International Conference on Smart Computing and Communication, ICSCC 2024

作者： Sardjono, Wahyu Retnowardhani, Astari Kristin, Desi Maya Bina Nusantara University BINUS Graduate Program Master of Information Systems Management Information Systems Management Department Jakarta11480 Indonesia School of Information Systems Bina Nusantara University Information Systems Department Jakarta11480 Indonesia

ISBN: (纸本)9798350363104

This manuscript provides an overview of building a mathematical model to formulate the conditions of a company's knowledge management system that is currently running. Tight competition in the business world makes efforts to seek innovation from several knowledge assets owned by organizations to maximize the use of intangible asset resources available to each individual and organization. Through optimal management, a competitive advantage will be obtained for the company by obtaining innovative ideas based on knowledge management theory. The use of factor analysis methods can be used to look for determining factors for the success of implementing a knowledge management system (KMS) and regression analysis methods can also be used to build mathematical models. The research results provide strategies to improve success factors in KMS implementation and contribute to the development of more effective and efficient implementation strategies for knowledge management and organizational goals toward corporate competitive advantage. © 2024 IEEE.

关键词： competitive advantage exploratory factor analysis innovation knowledge management mathematic model

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：