检索结果-内蒙古大学图书馆

international conference on Acoustics, Speech, and Signal processing (ICASSP)

作者： Keqi Deng Philip C. Woodland Department of Engineering University of Cambridge Cambridge UK

End-to-end (E2E) automatic speech recognition (ASR) implicitly learns the token sequence distribution of paired audio-transcript training data. However, it still suffers from domain shifts from training to testing, and domain adaptation is still challenging. To alleviate this problem, this paper designs a replaceable internal language model (RILM) method, which makes it feasible to directly replace the internal language model (LM) of E2E ASR models with a target-domain LM in the decoding stage when a domain shift is encountered. Furthermore, this paper proposes a residual softmax (R-softmax) that is designed for CTC-based E2E ASR models to adapt to the target domain without re-training during inference. For E2E ASR models trained on the LibriSpeech corpus, experiments showed that the proposed methods gave a 2.6% absolute WER reduction on the Switchboard data and a 1.0% WER reduction on the AESRC2020 corpus while maintaining intra-domain ASR results.

关键词： Training Adaptation models Training data Switches Signal processing data models Decoding

来源：评论

学校读者我要写书评

暂无评论

A Statistical-Based Approach to Load Model Parameter Identification

引用

IEEE ACCESS 2021年 9卷 66915-66928页

作者： Gulakhmadov, Aminjon Tavlintsev, Alexander Pankratov, Aleksey Suvorov, Anton Kovaleva, Anastasia Lipnitskiy, Ilya Safaraliev, Murodbek Semenenko, Sergey Gubin, Pavel Dmitriev, Stepan Rasulzoda, Khusrav Chinese Acad Sci Xinjiang Inst Ecol & Geog Res Ctr Ecol & Environm Cent Asia Urumqi 830011 Peoples R China Chinese Acad Sci Xinjiang Inst Ecol & Geog State Key Lab Desert & Oasis Ecol Urumqi 830011 Peoples R China Minist Energy & Water Resources Tajikistan Dushanbe 734064 Tajikistan Ural Fed Univ Dept Automated Elect Syst Ekaterinburg 620002 Russia United Power Syst Tomsk 634041 Russia HP Inc Portland OR 97211 USA Andritz Hydro GmbH D-88212 Ravensburg Germany

In the last few years, a great number of methods for identifying the load model parameters have been proposed. This article discusses the use of statistical approach to estimate the substation equivalent load model parameters for supplying to oil-producing industrial region. The disadvantages of existing statistical approach are the low accuracy obtained for the parameter estimates, especially when using samples size is small. To eliminate this deficiency, the current measurement data archive from SCADA system of electrical parameters for 15 months was collected. For the purpose of verifying the obtained results of statistical processing of SCADA data, a full-scale experiment was carried out in relation to the studied substation. The article describes the statistical method used to process the current SCADA measurement data, the results of archived statistical processing and experimental SCADA data. The electrical load models' parameters received from the experimental studies results are of practical importance.

关键词： Load modeling Mathematical model Voltage measurement data models Clustering algorithms Biological system modeling Aggregates Load modeling power system power system study static load model ZIP model

来源：评论

学校读者我要写书评

暂无评论

Multi-domain Recommendation with Embedding Disentangling and Domain Alignment 23

Multi-domain Recommendation with Embedding Disentangling and...

引用

32nd ACM international conference on Information and Knowledge Management (CIKM)

作者： Ning, Wentao Yan, Xiao Liu, Weiwen Cheng, Reynold Zhang, Rui Tang, Bo Univ Hong Kong Hong Kong Peoples R China Southern Univ Sci & Technol Hong Kong Peoples R China Huawei Noahs Ark Lab Montreal PQ Canada Univ Hong Kong Dept Comp Sci Hong Kong Peoples R China Univ Hong Kong Musketeers Fdn Inst Data Sci Hong Kong Peoples R China Guangdong Hong Kong Macau Joint Lab Hong Kong Peoples R China Southern Univ Sci & Technol Res Inst Trustworthy Autonomous Syst Shenzhen Peoples R China Southern Univ Sci & Technol Dept Comp Sci & Engn Shenzhen Peoples R China

ISBN: (纸本)9798400701245

Multi-domain recommendation (MDR) aims to provide recommendations for different domains (e.g., types of products) with overlapping users/items and is common for platforms such as Amazon, Facebook, and LinkedIn that host multiple services. Existing MDR models face two challenges: First, it is difficult to disentangle knowledge that generalizes across domains (e.g., a user likes cheap items) and knowledge specific to a single domain (e.g., a user likes blue clothing but not blue cars). Second, they have limited ability to transfer knowledge across domains with small overlaps. We propose a new MDR method named EDDA with two key components, i.e., embedding disentangling recommender and domain alignment, to tackle the two challenges respectively. In particular, the embedding disentangling recommender separates both the model and embedding for the inter-domain part and the intra-domain part, while most existing MDR methods only focus on model-level disentangling. The domain alignment leverages random walks from graph processing to identify similar user/item pairs from different domains and encourages similar user/item pairs to have similar embeddings, enhancing knowledge transfer. We compare EDDA with 12 state-of-the-art baselines on 3 real datasets. The results show that EDDA consistently outperforms the baselines on all datasets and domains. All datasets and codes are available at https://***/Stevenn9981/EDDA.

关键词： multi-domain recommendation disentanglement alignment

来源：评论

学校读者我要写书评

暂无评论

Medical data Clustering Based on Multi-objective Clustering Algorithm 4th

Medical Data Clustering Based on Multi-objective Clustering ...

引用

4th international conference on Machine Learning for Cyber Security, ML4CS 2022

作者： Chen, Shilian Tan, Yingsi Guo, Junkai He, Yuqin Geng, Shuang College of Management Shenzhen University Shenzhen518060 China

ISBN: (纸本)9783031201011

With the development of massive medical data, clustering algorithm becomes an effective way for medical data processing and data mining. On the one hand, it helps medical learners find effective information patterns from massive data;on the other hand, it promotes the development of medical technology and increase productivity. For traditional clustering algorithm, a single clustering index is difficult to meet people's needs of diversity and comprehensiveness. In contrast, multi-objective clustering (MOC) considers multiple objectives at the same time, and comprehensively deals with various clustering problems and standards, such as compactness, diversity of feature selection and high data dimension. Artificial bee colony algorithm (ABC) has a faster speed and embodies the idea of swarm intelligence. It imitates the optimization process of bees, and finally obtains the global optimal value. On this basis, this paper proposed a multi-objective artificial bee colony clustering algorithm (MOC-NABC) that is combined with current better-performed clustering algorithm. It takes normalized mutual information (NMI), Calinski-Harabasz (CH), Fowlkes-Mallows index (FMI) and silhouette coefficient (SC) of clustering as the final evaluation indexes. The experiment on UCI mouse protein gene dataset shows that the overall performance effect is greatly improved, e.g. compact clustering and the effective utilization of data features. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

WANE: Workload Adaptive Neuro-genetic Engine for Container Usage Prediction

WANE: Workload Adaptive Neuro-genetic Engine for Container U...

引用

Advances in Electrical Engineering and Computer Applications( AEECA), 2020 IEEE international conference on

作者： Soyeon Park Hyokyung Bahn Dept. of Computer Engineering Ewha University Seoul Republic of Korea

ISBN: (纸本)9798350355260

Microservice architectures in Kubernetes-based container environments often face challenges in optimally allocating resources due to fluctuating workload usage. Traditional methods, which rely on heuristic computing resource requests, typically result in overprovisioning and inefficient utilization. Predicting the actual runtime resource demand of containerized applications accurately is challenging. To address this issue, the authors introduce a novel hybrid cloud resource prediction engine that combines the strengths of genetic algorithms and Bayesian neural networks. Trained with real-world trace data, the proposed model outperforms existing techniques like ARIMA and exponential smoothing, particularly in reducing the risk of underprediction. Notably, these enhancements significantly improve CPU prediction accuracy, demonstrating potential for optimizing resource allocation and enhancing cost efficiency in cloud environments.

关键词： Cloud computing Costs Smoothing methods Neural networks Containers data models Resource management Engines Faces Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Show Me the Whole World: Towards Entire Item Space Exploration for Interactive Personalized Recommendations 22

Show Me the Whole World: Towards Entire Item Space Explorati...

引用

15th ACM international conference on Web Search and data Mining (WSDM)

作者： Song, Yu Sun, Shuai Lian, Jianxun Huang, Hong Li, Yu Jin, Hai Xie, Xing Natl Engn Res Ctr Big Data Technol & Syst Wuhan Peoples R China Serv Comp Technol & Syst Lab Wuhan Peoples R China Luazhong Univ Sci & Technol Wuhan Peoples R China Microsoft Res Asia Beijing Peoples R China Meituan Grp Beijing Peoples R China

ISBN: (纸本)9781450391320

User interest exploration is an important and challenging topic in recommender systems, which alleviates the closed-loop effects between recommendation models and user-item interactions. Contextual bandit (CB) algorithms strive to make a good trade-off between exploration and exploitation so that users' potential interests have chances to expose. However, classical CB algorithms can only be applied to a small, sampled item set (usually hundreds), which forces the typical applications in recommender systems limited to candidate post-ranking, homepage top item ranking, ad creative selection, or online model selection (A/B test). In this paper, we introduce two simple but effective hierarchical CB algorithms to make a classical CB model (such as LinUCB and Thompson Sampling) capable to explore users' interest in the entire item space without limiting to a small item set. We first construct a hierarchy item tree via a bottom-up clustering algorithm to organize items in a coarse-to-fine manner. Then we propose a hierarchical CB (HCB) algorithm to explore users' interest on the hierarchy tree. HCB takes the exploration problem as a series of decision-making processes, where the goal is to find a path from the root to a leaf node, and the feedback will be back-propagated to all the nodes in the path. We further propose a progressive hierarchical CB (pHCB) algorithm, which progressively extends visible nodes which reach a confidence level for exploration, to avoid misleading actions on upper-level nodes in the sequential decision-making process. Extensive experiments on two public recommendation datasets demonstrate the effectiveness and flexibility of our methods.

关键词： Recommender System Contextual Bandit Interest Exploration

来源：评论

学校读者我要写书评

暂无评论

Classification and prediction of ECG data based on decision trees and their optimization algorithms

Classification and prediction of ECG data based on decision ...

引用

Computers, Information processing and Advanced Education (CIPAE), 2020 international conference on

作者： Runqiang Xing Xi’an Technological University Xi’an China

The heart beat is usually synchronized with the electrical activity of the heart, and the ECG records the representation of the heart's electrical signals on the body surface. In this study, we developed a classification algorithm model for cardiac arrhythmias by analyzing ECG data. We first constructed a classification model to determine the danger level of ECG data using a decision tree approach to achieve real-time judgment and alarm of arrhythmias. The model was then post-pruned and optimized to further classify and predict specific categories for each risk level in conjunction with the actual situation, resulting in an algorithmic model with relatively high prediction accuracy. This will provide an important monitoring tool for cardiac patients and promote timely treatment.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Research on Automobile User Association Recommendation Algorithm Based on Feature Attribute and Matrix Decomposition

Research on Automobile User Association Recommendation Algor...

引用

Computers, Information processing and Advanced Education (CIPAE), 2020 international conference on

作者： Yang Jing Hu Huiying Zhang Fan Zhao Yishu China Automotive Technology and Research Center Co. Ltd. Tianjin China China Auto Information Technology (Tianjin) Co. Ltd. Tianjin China

ISBN: (数字)9798331527662

ISBN: (纸本)9798331527679

In order to solve the problem of association between automobile users and automobile configuration feature attributes under sparse matrix of big data, this paper proposes a matrix decomposition association recommendation algorithm based on automobile users and automobile configuration. Firstly, the crawler technology is used to obtain the data of related automobile websites. Secondly, the vehicle user-vehicle feature attribute system is built, and the crawler data is input into the system to form a sparse matrix. Then, gradient descent algorithm is used to decompose the sparse matrix into two low-dimensional matrices, P and Q. Finally, the prediction matrix is used to recommend the car configuration with large factor to the car users. This method makes up for the research on the association recommendation algorithm between automobile users and automobile product configuration under the background of sparse crawler big data matrix, and provides a reliable theoretical basis for the development of the industry.

关键词： Industries Crawlers Information processing Big data Reliability theory Prediction algorithms Reliability engineering Automobiles Sparse matrices Matrix decomposition

来源：评论

学校读者我要写书评

暂无评论

Research on Product Quality Control Algorithm System of Intelligent Manufacturing Based on Big data

Research on Product Quality Control Algorithm System of Inte...

引用

Computers, Information processing and Advanced Education (CIPAE), 2020 international conference on

作者： Zhang Jingwen Zheng Min Zhang Yifan School of Logistics Linyi University Linyi China

ISBN: (数字)9798331527662

ISBN: (纸本)9798331527679

In this paper, an improved PRO-PSO algorithm is proposed for intelligent manufacturing industry to achieve efficient performance in product quality control and anomaly location. The effect of the number of neurons on the detection accuracy and the risk of misjudgment, and the significant effect of data characteristics on the performance of the model were verified by experiments. After optimization, the model achieves an anomaly detection accuracy of 88.02%, a false detection rate of 1.13%, and a positioning accuracy of 99.4%. This achievement provides cutting-edge quality control means for intelligent manufacturing industry, significantly improves production efficiency and product quality by accurately identifying and locating abnormal data, and lays a solid foundation for the construction of intelligent manufacturing system in the era of Industry 4.0.

关键词： Manufacturing industries Accuracy Neurons Quality control Production Solids Product design data models Quality assessment Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Automatic Feasibility Study via data Quality Analysis for ML: A Case-Study on Label Noise 39

Automatic Feasibility Study via Data Quality Analysis for ML...

引用

39th IEEE international conference on data Engineering, ICDE 2023

作者： Renggli, Cedric Rimanic, Luka Kolar, Luka Wu, Wentao Zhang, Ce Eth Zurich Switzerland Microsoft Research

ISBN: (纸本)9798350322279

In our experience of working with domain experts who are using today's AutoML systems, a common problem we encountered is what we call "unrealistic expectations"- when users are facing a very challenging task with a noisy data acquisition process, while being expected to achieve startlingly high accuracy with machine learning (ML). Many of these are predestined to fail from the beginning. In traditional software engineering, this problem is addressed via a feasibility study, an indispensable step before developing any software system. In this paper, we present Snoopy, with the goal of supporting data scientists and machine learning engineers performing a systematic and theoretically founded feasibility study before building ML applications. We approach this problem by estimating the irreducible error of the underlying task, also known as the Bayes error rate (BER), which stems from data quality issues in datasets used to train or evaluate ML models. We design a practical Bayes error estimator that is compared against baseline feasibility study candidates on 6 datasets (with additional real and synthetic noise of different levels) in computer vision and natural language processing. Furthermore, by including our systematic feasibility study with additional signals into the iterative label cleaning process, we demonstrate in end-to-end experiments how users are able to save substantial labeling time and monetary efforts. © 2023 IEEE.

关键词： Machine learning

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：