检索结果-内蒙古大学图书馆

MTTF: a multimodal transformer for temperature forecasting

International Journal of Computers and Applications 2024年第2期46卷 122-135页

作者： Cao, Yang Zhai, Junhai Zhang, Wei Zhou, Xuesong Zhang, Feng Key Laboratory of Machine Learning and Computational Intelligence College of Mathematics and Information Science Hebei University Hebei Province Baoding China

Accurate weather forecasting is crucial for various applications, including agriculture and environmental monitoring. However, existing deep learning based methods typically use only temperature observations as input, which do not consider spatial location (e.g. neighboring regions usually show similar temperature trends) and increase the difficulty in predicting anomalous fluctuations in temperature. To address the issue, a multimodal transformer model for temperature forecasting is proposed. This model not only uses observational data (time series data) as input but also incorporates spatially rich ERA5 reanalysis temperature data (spatio-temporal data). The proposed method has two distinctive features: (1) a sequence merging module to highlight dominating features and reduce the cost of attention calculations. (2) a cross-all-modal attention mechanism to capture dependencies between the current modality and all modalities. The method was evaluated on three temperature datasets from a plain and two challenging high-altitude mountainous regions. Results showed that the proposed method outperforms existing methods in terms of MAE and MSE, offering a promising new solution for temperature prediction. Our code and data are availab.e at https://***/Adam618/MTTF. © 2023 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Weather forecasting

来源：评论

学校读者我要写书评

暂无评论

Two-stage dimensionality reduction approach based on 2DLDA and fuzzy rough sets technique

引用

NEUROCOMPUTING 2011年第17期74卷 3722-3727页

作者： Zhao, Hao-Xin Xing, Hong-Jie Wang, Xi-Zhao Hebei Univ Coll Math & Comp Sci Key Lab Machine Learning & Computat Intelligence Baoding 071002 Hebei Province Peoples R China

Traditional two-dimensional linear discriminant analysis (2DLDA) can deal with discriminant information between classes and directly extract features from image matrices. However, 2DLDA essentially works solely in the row-direction of images. Therefore, the features extracted by 2DLDA may contain redundant information. In this letter, a dimensionality reduction method based on 2DLDA and fuzzy rough sets technique is proposed to deal with the foresaid problem. Experimental results on the four benchmark face databases demonstrate that the proposed method is superior to its related methods. (C) 2011 Elsevier B.V. All rights reserved.

关键词： 2DLDA Fuzzy rough sets Feature extraction Dimensionality reduction Face recognition

来源：评论

学校读者我要写书评

暂无评论

Case-based multi-task pathfinding algorithm

Case-based multi-task pathfinding algorithm

引用

2012 International Conference on machine learning and Cybernetics, ICMLC 2012

作者： Li, Yan Su, Lan-Ming He, Qiang Key Lab. in Machine Learning and Computational Intelligence College of Mathematics and Computer Science Hebei University Baoding 071002 Hebei Province China

ISBN: (纸本)9781467314855

Pathfinding is an important task in computer games, where the algorithm efficiency is the key issue. In this paper, we introduce case-based reasoning method in the process of A* algorithm in multi-task pathfinding. Firstly, we save some typical paths as cases. When a new task is coming, it no longer uses A* to find a path from scratch, but firstly computes the similarity of the new task and the stored cases to decide whether to go along the previous found paths or not. A solution to the new task will be obtained after adapting to the found similar case(s). Obviously, this memory-based pathfinding can reduce the search time at the cost of using more memory to store found paths as cases. Through experimental results, it is demonstrated that, as the number of stored paths is increasing, fewer nodes are needed to be searched during the pathfinding process. © 2012 IEEE.

关键词： Case based reasoning

来源：评论

学校读者我要写书评

暂无评论

PSO-based method for learning similarity measure of nominal features

PSO-based method for learning similarity measure of nominal ...

引用

International Conference on machine learning and Cybernetics

作者： Li, Yan Zhang, Xiu-Li Key Lab. in Machine Learning and Computational Intelligence Faculty of Mathematics and Computer Science Hebei University Baoding 071002 Hebei Province China

ISBN: (纸本)9781424465262

This paper presents a PSO-based method for learning similarity measure of nominal features for case based reasoning classifiers (i.e. CBR classifiers). The symbolic features considered here takes completely unordered values. It has been indicated in [3] that in specific classification task, the similarities between these nominal feature values can not be simply considered as either 0 or 1. A GA-based approach has been developed for learning similarity measure of such feature values. However, when the number of features and feature values become larger, the GA-based algorithm's convergence speed obviously slows down, and the accuracy of classification may be also affected. To address this problem, we propose a PSO-based algorithm for learning similarity measure of nominal features and further describe feature importance through the learned similarity measure. The experimental results show that, using the proposed PSO-based algorithm, the convergence speed is much faster than that of GA-based algorithm and the accuracy is also improved. In addition, we also explain that the feature importance defined through the learned similarities is essentially consistent with that in rough sets, and an illustrative example is finally provided. © 2010 IEEE.

关键词： Particle swarm optimization (PSO)

来源：评论

学校读者我要写书评

暂无评论

Reductions in inconsistent decision systems based on dominance relations

Reductions in inconsistent decision systems based on dominan...

引用

International Conference on machine learning and Cybernetics

作者： Li, Yan Sun, Na-Xin Zhao, Jin Key Lab. In Machine Learning and Computational Intelligence College of Mathematics and Computer Science Hebei University Baoding 071002 Hebei Province China

ISBN: (纸本)9781424465262

By incorporating domination principle in inconsistent decision systems based on dominance relations, we define the concept of distribution function for a decision system to directly reflect the inconsistent degree of this system. A new type of distribution reduction and maximum distribution reduction are correspondingly introduced. The relationships between these two reductions are discussed, and their judgment theorems are given. In addition, this article also discusses the relations of several existing reductions, including compatible reduction, absolute reduction, and the proposed distribution reduction and maximum distribution reduction. © 2010 IEEE.

关键词： Distribution functions

来源：评论

学校读者我要写书评

暂无评论

Chinese functional chunk identification based on part of speech

Chinese functional chunk identification based on part of spe...

引用

International Conference on Communication Systems and Network Technologies, CSNT 2012

作者： Wang, Cheng-Yan Li, Xin-Fu Tian, Xue-Dong College of Mathematics and Computer Science Hebei University Baoding China Key Lab. in Machine Learning and Computational Intelligence of Hebei Province China

ISBN: (纸本)9780769546926

Chinese functional chunk describes the basic skeleton of the Chinese sentences. It is the important bridge for joining syntax and semantic description, and the Chinese functional chunk identification plays a key role in the field of natural language processing. Identifying the functional chunk is mostly based on methods of rules which come from the syntactic analysis. In this paper, the functional chunk is mainly identified according to the effect of the part of speech in the syntactic analysis. The experimental results show the proposed method works well on the functional chunk identification. © 2012 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Image classification by combining local and mid-level features 18

Image classification by combining local and mid-level featur...

引用

2nd International Conference on Innovation in Artificial intelligence, ICIAI 2018

作者： Lu, Yao Zhang, Hui Key Lab. of Machine Learning and Computational Intelligence College of Mathematics and Information Science Hebei University Hebei China

ISBN: (纸本)9781450363457

It is meaningful to study high performance image classification algorithms for massive image management and effective organization. Image feature representations directly affect the performance of classification algorithm, however, in the current popular sparse coding framework for image classification, the relationship between local features is neglected, and there is still considerable room left for further improvement. In this paper, mid-level features are extracted efficiently in the superpixel region, which embeds the structural information of local features. Then we combine the local features and mid-level features to achieve better discriminative ability under the sparse coding framework. Compared to using only local features, the new methods improve the classification performance about 3% on several popular benchmarks. © 2018 Association for Computing machinery.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Selection of deep web database based on retrieval performance

Selection of deep web database based on retrieval performanc...

引用

2nd International Workshop on Education Technology and Computer Science, ETCS 2010

作者： Li, Weijing Yuan, Fang Zhang, Ming Key Lab. in Machine Learning and Computational Intelligence College of Mathematics and Computer Science Hebei University Baoding Hebei China

ISBN: (纸本)9780769539874

A mass of high-quality information included in Deep Web can be accessed, which is still growing rapidly with the rapid development of the World Wide Web. Therefore it becomes more and more important to find the Web databases which are most relevant to the user queries. In this paper we propose a selection method of Web database based on retrieval performance. This method can fix the topic based on website characteristics and then classify the websites. Finally it decides which Web database can be chosen based on the retrieval performance. This method can not only accurately select the Web databases which satisfy the user queries but also improve the speed of the database query and the quality of retrieval. © 2010 IEEE.

关键词： Websites

来源：评论

学校读者我要写书评

暂无评论

Potential support vector machine based on the reduced samples

Potential support vector machine based on the reduced sample...

引用

International Conference on Information Science and Engineering

作者： Lu, Shu-Xia Cao, Gui-En Meng, Jie Wang, Hua-Chao Key Lab. of Machine Learning and Computational Intelligence College of Mathematics and Computer Science Hebei University Baoding Hebei China

ISBN: (纸本)9781424480968

When the training dataset is very large, the learning process of potential support vector machine takes up so large memory that the training speed is very slow. To accelerate the training speed of the potential support vector machine (PSVM) for large-scale datasets, a new method is proposed, which introduces PSVM based on the reduced samples. The new method removes most non-support vectors, and keeps the samples on and near the boundary, which may be the support vectors, as the new training samples. This method is more suitable to large-scale datasets. The experimental results show that the proposed method performs well to decrease the consumption of computer memory,and accelerate the training speed of PSVM. © 2010 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

A rank reduced matrix method in extreme learning machine

A rank reduced matrix method in extreme learning machine

引用

9th International Symposium on Neural Networks, ISNN 2012

作者： Lu, Shuxia Zhang, Guiqiang Wang, Xizhao Key Lab. of Machine Learning and Computational Intelligence College of Mathematics and Computer Science Hebei University Baoding Hebei China

ISBN: (纸本)9783642313455

Extreme learning machine (ELM) is a learning algorithm for single-hidden layer feedforward neural networks (SLFNs) which randomly chooses hidden nodes and analytically determines the output weights of SLFNs. but when dealing with large datasets, we need more hidden nodes to enhance training and testing accuracy, in this case, this algorithm can't achieve high speed any more, sometimes its training can't be executed because the bias matrix is out of memory. We focus on this issue and use the Rank Reduced Matrix (MMR) method to calculate the hidden layer output matrix, the result showed this method can not only reach much higher speed but also better improve the generalization performance whenever the number of hidden nodes is large or not. © 2012 Springer-Verlag.

关键词： Singular value decomposition

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：