检索结果-内蒙古大学图书馆

27th China Conference on Information Retrieval, CCIR 2021

作者： Liu, Zhaohui Yan, Hongfei Chen, Chong Su, Qi Peking University Beijing China Beijing Normal University Beijing China National Engineering Laboratory for Big Data Analysis and Application Technology Center for Big Data Research Peking University Beijing China

ISBN: (纸本)9783030881887

With the development of Internet technology, online learning has gradually been widely used, which has further enriched the teaching methods. However, the indicators provided by online learning platforms to characterize students’ learning behaviors are still relatively small, it is hard to effectively evaluate students’ learning effects and provide personalized education for students. While students produce a large number of online data records during the learning process, if these data can be used to evaluate the learning effect of students and find indicators that are significantly related to the learning effect, it can help teachers know the learning status of students and provide them with more targeted guidance. The framework of this article includes: firstly collect online data of students’ learning, and determine major types of factors that affect the learning effect. Next, the original data is processed, and new features are constructed from the original data according to the factors that affect the learning effect. Then use several different methods to select the features that have the most significant impact on the learning effect, and use several regression methods to predict students’ academic performance. Finally, this article selects important features that are more relevant to learning effects, studies the relationship between these features. © 2021, Springer Nature Switzerland AG.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Empirical research on futures trading strategy based on time series algorithm 26th

Empirical research on futures trading strategy based on time...

引用

26th China Conference on Information Retrieval, CCIR 2020

作者： Yao, Shi Hongfei, Yan Siping, Ying Chong, Chen Qi, Su Peking University Beijing China Beijing Normal University Beijing China National Engineering Laboratory for Big Data Analysis and Application Technology Center for Big Data Research Peking University Beijing China

ISBN: (纸本)9783030567248

This article attempts to establish a trading strategy framework based on deep neural networks for the futures market, which consists of two parts: time series forecasting and trading strategies based on trading signals. In the time series forecasting task, we experimented with three types of methods with different entry points, namely recurrent neural networks with gate structure, networks combining time and frequency domain information, and network structures using attention mechanism. In the trading strategy part, the buying and selling signals and the corresponding trading volume are established according to the prediction results, and trading is conducted with the frequency of hours. In the empirical exploration part, we tested the prediction effect and strategic rate of return of various models on the copper contract. The data shows that in general, the best strategy can obtain a relatively stable income growth that has nothing to do with market fluctuations, but lacks countermeasures for rare external events with greater impact. © Springer Nature Switzerland AG 2020.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

An integrated machine learning framework for stock price prediction 26th

An integrated machine learning framework for stock price pre...

引用

26th China Conference on Information Retrieval, CCIR 2020

作者： Bi, Quanzhi Yan, Hongfei Chen, Chong Su, Qi Peking University Beijing China Beijing Normal University Beijing China National Engineering Laboratory for Big Data Analysis and Application Technology Center for Big Data Research Peking University Beijing China

ISBN: (纸本)9783030567248

Predicting the future price of financial assets has always been an important research topic in the field of quantitative finance. This paper attempts to use the latest artificial intelligence technologies to design and implement a framework for financial asset price prediction. The framework we use is divided into three modules: Feature engineering, Regressor, and Hyper Optimizer. The Feature engineering module extract multiple features using technical indicators, FinBERT, FFT, ARIMA, stacked auto-encoder, PCA and XGBoost. The Regressor module consists of a generative adversarial network, where the generator is Seq2Seq and the discriminator is GRU. The HyperOptimizer module will tune the parameter in GAN using the Bayesian optimization algorithm. Finally, we conducted numerical experiments on our framework, which shows that the framework implemented in this paper performs better than the benchmark method. © Springer Nature Switzerland AG 2020.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Semi-supervised sentiment analysis for chinese stock texts in scarce labeled data scenario and price prediction 26th

Semi-supervised sentiment analysis for chinese stock texts i...

引用

26th China Conference on Information Retrieval, CCIR 2020

作者： Zhaoyan, Ji Hongfei, Yan Siping, Ying Chong, Chen Qi, Su Peking University Beijing China Beijing Normal University Beijing China National Engineering Laboratory for Big Data Analysis and Application Technology Center for Big Data Research Peking University Beijing China

ISBN: (纸本)9783030567248

The application of neural network in stock prediction is developing rapidly these years because of its excellency in series data processing. However, as most of research are conducted in English, data sources and labeled data are inadequate in Chinese. Especially for natural language processing tasks in specific domain where specialized labeled data are required to train models to adapt to terminology processing, specialized labeled Chinese in text data are very scarce, such as financial text data. To tackle this challenge, we proposed a semi-supervised learning method to generate well-labeled data and train BERT, a leading natural language processing model, to obtain a trained sentiment machine. Then we got stock-related text data sentiment score based on this machine and further combine the sentiment score and other transaction data as inputs for different neural networks to predict stock price. The experimental results on a large scale of Chinese stock data and texts showed that our proposed method successfully improved prediction accuracy compared to other established methods. Besides, we also examined our method’s applicability combined with different neural networks when predicting different types of stock. © Springer Nature Switzerland AG 2020.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Short-term origin-destination passenger flow forecasting in urban rail transit systems: An ensemble deep learning approach based on data augmentation

引用

Computers & Industrial engineering 2025年 206卷

作者： Xin Wang Dingjun Chen Xuze Ye Xia Zhao Shaoquan Ni School of Transportation and Logistics Southwest Jiaotong University Chengdu 611756 China National and Local Joint Engineering Laboratory of Comprehensive Intelligent Transportation Southwest Jiaotong University Chengdu 611756 China National Engineering Laboratory of Integrated Transportation Big Data Application Technology Southwest Jiaotong University Chengdu 611756 China

Accurate network-level Origin-Destination (OD) passenger flow forecasting is crucial for enhancing the efficiency and service quality of urban rail transit (URT). In URT networks, a significant portion of the passenger flow comes from medium and low flow OD pairs. However, network-level OD passenger flow data exhibit characteristics such as high-dimensional sparsity, data availability, and strong randomness, which severely constrain the performance of forecasting models for medium and low flow OD pairs. In view of this, this study proposes an ensemble deep learning framework (PatchPF) with data augmentation at its core, aimed at short-term OD passenger flow forecasting at the URT network level. We introduce a novel BaggingT mechanism to implement time series ensemble forecasting in PatchPF to further improve the forecasting performance and robustness. The PatchPF architecture is tested on real-world metro datasets from Chongqing and Chengdu, China. The results indicate that it outperforms the other benchmark models. Moreover, the PatchPF architecture does not exhibit the performance bottlenecks in forecasting medium and low flow OD pairs that other state-of-the-art models do, demonstrating the effectiveness of PatchPF and its key components in OD passenger flow forecasting.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Time-aware bike flow prediction framework with dynamic edge fusion and memory integration

引用

Knowledge-Based Systems 2025年 322卷

作者： M.Rajeh, Taha Li, Tianrui Luo, Zhpeng Javed, Muhammad Hafeez Alhaek, Fares School of Computing and Artificial Intelligence Southwest Jiaotong University Chengdu611756 China Engineering Research Center of Sustainable Urban Intelligent Transportation Ministry of Education Chengdu611756 China National Engineering Laboratory of Integrated Transportation Big Data Application Technology Southwest Jiaotong University Chengdu611756 China Manufacturing Industry Chains Collaboration and Information Support Technology Key Laboratory of Sichuan Province Southwest Jiaotong University Chengdu611756 China Stirling College Chengdu University Chengdu610106 China Computing Science Faculty of Natural Sciences University of Stirling StirlingFK9 4LA United Kingdom

Bike-Sharing Systems (BSSs) have become a popular solution to urban sustainability concerns, providing affordable and environmentally friendly transportation. However, modeling and predicting shared-bike flow patterns are essential for developing intelligent transportation systems and facilitating city management. Despite the existence of several inflow and outflow prediction models, certain unresolved issues remain, including the inability to differentiate and capture heterogeneous dynamic relationships within daily flows, and the suboptimal utilization of historical flow data for forecasting. To address these limitations, we propose MVRC (Multi-View Relational Framework based on Clustering), a novel deep-learning framework for precise inflow and outflow prediction. The MVRC framework models three types of heterogeneous dynamic flow relationships during different periods of the day. Initially, we introduce a novel clustering method to group nearby stations exhibiting similar patterns, followed by creating a dynamic memory module to store and update edge attributes based on the time of day. Furthermore, the dynamic edge fusion mechanism effectively integrates heterogeneous, time-evolving flow features, capturing their varying impacts. To further enhance the model's capabilities, an offline component that leverages historical data is incorporated, improving performance during testing. Extensive experiments conducted on real datasets from New York City and Washington, D.C., demonstrate the superior prediction efficiency and framework effectiveness of the MVRC framework compared to existing state-of-the-art techniques. These results highlight the practicality and efficacy of MVRC in terms of prediction accuracy and stability. © 2025

关键词： Urban transportation

来源：评论

学校读者我要写书评

暂无评论

Accelerating local SGD for non-IID data using variance reduction

引用

Frontiers of Computer Science 2023年第2期17卷 73-89页

作者： Xianfeng LIANG Shuheng SHEN Enhong CHEN Jinchang LIU Qi LIU Yifei CHENG Zhen PAN Anhui Province Key Laboratory of Big Data Analysis and Application University of Science and Technology of ChinaHefei 230027China Ant Financial Services Group Hangzhou 310000China Department of Computer Science and Engineering Hong Kong University of Science and TechnologyHong Kong 999077China

Distributed stochastic gradient descent and its variants have been widely adopted in the training of machine learning models,which apply multiple workers in *** them,local-based algorithms,including Local SGD and FedAvg,have gained much attention due to their superior properties,such as low communication cost and ***,when the data distribution on workers is non-identical,local-based algorithms would encounter a significant degradation in the convergence *** this paper,we propose Variance Reduced Local SGD(VRL-SGD)to deal with the heterogeneous *** extra communication cost,VRL-SGD can reduce the gradient variance among workers caused by the heterogeneous data,and thus it prevents local-based algorithms from slow convergence ***,we present VRL-SGD-W with an effectivewarm-up mechanism for the scenarios,where the data among workers are quite *** from eliminating the impact of such heterogeneous data,we theoretically prove that VRL-SGD achieves a linear iteration speedup with lower communication complexity even if workers access non-identical *** conduct experiments on three machine learning *** experimental results demonstrate that VRL-SGD performs impressively better than Local SGD for the heterogeneous data and VRL-SGD-W is much robust under high data variance among workers.

关键词： distributed optimization variance reduction local SGD federated learning non-IID data

来源：评论

学校读者我要写书评

暂无评论

WOBTree:a write-optimized B+-tree for non-volatile memory

引用

Frontiers of Computer Science 2021年第5期15卷 37-51页

作者： Haitao WANG Zhanhuai LI Xiao ZHANG Xiaonan ZHAO Song JIANG School of Computer Science and Engineering Northwestern Polytechnical UniversityXi'An 710072China Key Laboratory of Big Data Storage and Management Northwestern Polytechnical UniversityMinistry of Industry and Information TechnologyXi'An 710072China National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology Xi’An 710072China Department of Computer Science and Engineering University of Texas at ArlingtonArlingtonTX 76010USA

The emergence of non-volatile memory(NVM)has introduced new opportunities for performance optimizations in existing storage *** better utilize its byte-addressability and near-DRAM performance,NVM can be attached on the memory bus and accessed via load/store memory instructions rather than the conventional block *** this scenario,a cache line(usually 64 bytes)becomes the data transfer unit between volatile and non-volatile ***,the failure-atomicity of write on NVM is the memory bit width(usually 8 bytes).This mismatch between the data transfer unit and the atomicity unit may introduce write amplification and compromise data consistency of node-based data structures such as B+-*** this paper,we propose WOBTree,a Write-Optimized B+-Tree for NVM to address the mismatch problem without expensive *** minimizes the update granularity from a tree node to a much smaller subnode and carefully arranges the write operations in it to ensure crash consistency and reduce write *** results show that compared with previous persistent B+-tree solutions,WOBTree reduces the write amplification by up to 86× and improves write performance by up to 61× while maintaining similar search performance.

关键词： non-volatile memory B+-tree atomic granularity mismatch write amplification performance optimization

来源：评论

学校读者我要写书评

暂无评论

MSLKCNN: A Simple and Powerful Multi-scale Large Kernel CNN for Hyperspectral Image Classification

引用

IEEE Transactions on Geoscience and Remote Sensing 2025年 63卷

作者： Liu, Xun Ng, Alex Hay-Man Lei, Fangyuan Ren, Jinchang Guo, Li Du, Zheyuan Guangdong University of Technology School of Information Engineering Guangzhou510006 China Guangdong University of Technology School of Civil and Transportation Engineering Guangzhou510006 China Guangdong Polytechnic Normal University Guangdong Provincial Key Laboratory of Intellectual Property Big Data Guangzhou510665 China Robert Gordon University National Subsea Centre AberdeenAB21 0BH United Kingdom Ministry of Natural Resources Land Satellite Remote Sensing Application Center Beijing100048 China Geoscience Australia Canberra2609 Australia University of New South Wales School of Civil and Environmental Engineering SydneyNSW2052 Australia

Deep learning-based hyperspectral image (HSI) classification models typically utilize multiple feature extraction layers to learn the features of land covers. Nevertheless, they encounter challenges, e.g., 1) Transformers require substantial computational resources, and 2) these layers are carefully assembled and designed. Recently, large kernel convolutional neural networks (LKCNNs) show excellent performance in natural visual tasks. To tackle these limitations and explore the capability of LKCNNs for HSI classification, we present a novel simple and powerful multi-scale large kernel convolutional neural network architecture (MSLKCNN) with the largest kernel size as large as 15 × 15, in contrast to commonly used 3 × 3, for HSI classification. MSLKCNN avoids these specialized designs, comprising a noise suppression module (NSM) and a multi-scale large kernel convolution (MSLKC). Specifically, NSM is first used to suppress the noise and reduce the number of the bands before extracting the features. Then, MSLKC, as the only feature extraction layer of MSLKCNN, joints three parallel convolutions to capture the features of various types (i.e. spectral, spectral-spatial) and ranges (i.e., small local, larger local, and global) from the dimension of scale: (C1) convolution with a kernel size of 1 × 1 is used to extract spectral features;(C2) multi-scale large kernel depthwise separable convolution (MLKDC) is proposed to learn the spectral-spatial features of different ranges including short-range, middle-range, and long-range;and (C3) multi-scale dilated depthwise separable convolution (MDDC) is designed to aggregate the spectral-spatial features between land covers at various distances. Extensive experimental results on three public HSI datasets demonstrate the competitiveness of the proposed MSLKCNN compared with several state-of-the-art methods. © 1980-2012 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

The Existence of a Class of Balanced Multi-output Rotation Symmetric Boolean Functions

引用

Chinese Journal of Electronics 2018年第5期27卷 995-1001页

作者： DU Jiao FU Shaojing QU Longjiang LI Chao PANG Shanqi College of Mathematics and Information Science Henan Normal University Henan Engineering Laboratory for Big Data Statistical Analysis and Optimal Control Henan Normal University College of Computer Science National University of Defense Technology College of Liberal Arts and Science National University of Defense Technology

A new characterization of balanced rotation symmetric（n, m）-functions is presented. Based on the characterization, the nonexistence of balanced rotation symmetric（pr, m）-functions is determined, where p is an odd prime and m ≥ 2. And there exist balanced rotation symmetric（2r, m）-functions for 2 ≤ m ≤ 2r-r. With the help of these results, we also prove that there exist rotation symmetric resilient（2r, m）-functions for 2 ≤ m ≤ 2r-r-1.

关键词： Cryptography Rotation symmetric Orbit matrix Correlation immune Resilient functions

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：