检索结果-内蒙古大学图书馆

IEEE International Conference on big data

作者： Antonela Tommasel Ira Assent Department of Computer Science DIGIT Aarhus University Centre for Digitalisation Big Data and Data Analytics Denmark ISISTAN CONICET-UNCPBA Aarhus University Argentina Department of Computer Science DIGIT Aarhus University Centre for Digitalisation Big Data and Data Analytics Aarhus University Denmark

Recommender systems play a crucial role in how users consume information and establish new social relations. However, different factors (such as the data collection process, the designed recommendation model, or even the interpretation of findings) could make recommenders (unintendedly) prone to biases, favouring certain user groups or items, thus resulting in unfair outcomes. Recommenders also face fairness criticism for inducing filter bubbles, echo chambers, and, more generally, facilitating opinion manipulation. In this work, we study the impact of user recommender systems on fairness. To this end, we carry out a user recommendation task on a politically polarized Twitter data collection. Then, we evaluate how the different politically aligned user groups experience recommendation quality. Finally, we explore causal models to identify data and model-related features that could affect the fairness of recommender outcomes. Our study shows that political alignment is associated with the unfairness of recommenders affecting not only the relevance of recommendations, but also their diversity and the resulting interaction patterns.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Predicting the Impact of University Majors on the Length of Post-Graduation Unemployment Periods in Bahrain

Predicting the Impact of University Majors on the Length of ...

引用

2022 International Conference on data analytics for Business and Industry, ICDABI 2022

作者： Alkhayyat, Aysha Musaaed, Elham Alheddi, Isa Zeki, Ahmed University of Bahrain MSc. in Big Data Science and Analytics College of Science Bahrain University of Bahrain College of Information Technology Department of Information Systems Bahrain

ISBN: (纸本)9781665490580

Unemployment rate is among the most important measures that reflects the country's development. International investors and creditors will closely monitor the unemployment rate prior to making any decision to invest in any country. In addition, this measure is being monitored by the governments to set plans to keep it within the norm, should it exceed the acceptable limits. By understanding the importance of this measure and in pursuance of guiding high school graduates to enroll in majors of high demand and hence reduce the waiting time, this research studied the ability of three classifiers, namely Support Vector Machine (SVM), KNearest Neighbors (KNN), and Naive Bayes (NB) to accurately predict the length of post-graduation unemployment periods based on the university major of University of Bahrain (UOB) graduates. SVM classifier was found to provide the best predictions, though not significantly outperforming the KNN and NB classifiers. Moreover, other factors such as the Grade Point Average (GPA) and gender can be tested for their impact on unemployment period individually or grouped with the major. Additionally, this paper can be used to set the basis for deploying real life applications that will predict the unemployment period based on multiple factors. © 2022 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Changes in frailty and incident cardiovascular disease in three prospective cohorts (vol 45, pg 1048, 2024)

引用

EUROPEAN HEART JOURNAL 2024年第19期45卷 1768-1768页

作者： He, Di Wang, Zhaoping Li, Jun Yu, Kaixin He, Yusa He, Xinyue Liu, Yuanjiao Li, Yuhao Fu, Ruiyi Zhou, Dan Zhu, Yimin Department of Epidemiology and Biostatistics and Department of Respiratory Disease Sir Run Run Shaw Hospital Zhejiang University School of Medicine 866 Yuhangtang Rd Hangzhou 310058 Zhejiang China Department of Big Data in Health Science School of Public Health and the Second Affiliated Hospital Zhejiang University School of Medicine 866 Yuhangtang Rd Hangzhou 310058 Zhejiang China Center of Clinical Big Data and Analytics of the Second Affiliated Hospital Zhejiang University School of Medicine 866 Yuhangtang Rd Hangzhou 310058 Zhejiang China Cancer Center Zhejiang University 866 Yuhangtang Rd Hangzhou 310058 Zhejiang China

Background and Aims Previous studies found that frailty was an important risk factor for cardiovascular disease (CVD). However, previous studies only focused on baseline frailty status, not taking into consideration the changes in frailty status during follow-up. The aim of this study was to investigate the associations of changes in frailty status with incident *** This study used data of three prospective cohorts: China Health and Retirement Longitudinal Study (CHARLS), English Longitudinal Study of Ageing (ELSA), and Health and Retirement Study (HRS). Frailty status was evaluated by the Rockwood frailty index and classified as robust, pre-frail, or frail. Changes in frailty status were assessed by frailty status at baseline and the second survey which was two years after the baseline. Cardiovascular disease was ascertained by self-reported physician-diagnosed heart disease (including angina, heart attack, congestive heart failure, and other heart problems) or stroke. Cox proportional hazard models were used to calculate the hazard ratio (HR) and 95% confidence interval (95% CI) after adjusting for potential *** A total of 7116 participants from CHARLS (female: 48.6%, mean age: 57.4 years), 5303 from ELSA (female: 57.7%, mean age: 63.7 years), and 7266 from HRS (female: 64.9%, mean age: 65.1 years) were included according to inclusion and exclusion criteria. The median follow-up periods were 5.0 years in the CHARLS, 10.7 years in the ELSA, and 9.5 years in the HRS. Compared with stable robust participants, robust participants who progressed to pre-frail or frail status had increased risks of incident CVD (CHARLS, HR = 1.84, 95% CI: 1.54–2.21; ELSA, HR = 1.53, 95% CI: 1.25–1.86; HRS, HR = 1.59, 95% CI: 1.31–1.92). In contrast, frail participants who recovered to robust or pre-frail status presented decreased risks of incident CVD (CHARLS, HR = 0.62, 95% CI: 0.47–0.81; ELSA, HR = 0.49, 95% CI: 0.34–0.69; HRS, HR = 0.70, 95% CI: 0.55–0.89) whe

关键词： Frailty Cardiovascular disease Dynamic nature Epidemiology

来源：评论

学校读者我要写书评

暂无评论

Comparing K-Nearest Neighbors, Random Forest and Naïve Bayes Models to Classify Fetal Health Using Resampling Methods

Comparing K-Nearest Neighbors, Random Forest and Naïve Baye...

引用

2022 International Conference on data analytics for Business and Industry, ICDABI 2022

作者： Aldossary, Yasmeen Alhaddad, Salman Ebrahim, Majeed Zeki, Ahmed M. University of Bahrain MSc. in Big Data Science and Analytics College of Science Bahrain University of Bahrain College of Information Technology Department of Information System Bahrain

ISBN: (纸本)9781665490580

To preserve the lives of pregnant women and their fetuses, Cardiotocography (CTG) data is used to check the fetus's condition and for early intervention if necessary. As a result of the existence of imbalanced classes of fetus's condition, the performance of some models is affected when applied to real data. Accordingly, this research aims to observe the effect of data resampling methods on models' performance by adopting the CRISP methodology. The performance of k-Nearest Neighbors (KNN), Random Forest (RF), and Naive Bayes (NB) models are compared using resampled data with several methods with the help of the CorrelationAttributeEval feature selection method for high dimension reduction. The results stated that NB and RF models' performance has not improved with the resampling methods, while KNN has significantly improved across all metrics. RF is considered a good classifier as it was the best, with an accuracy of 93.8%. NB was the only improved model using feature selection with imbalanced data. Also, NB has the worst performance in terms of accuracy compared to KNN and RF. © 2022 IEEE.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Predicting Performance of Primary School Students and Determining The Dominant Factors Using data Mining with Feature Selection Methods

Predicting Performance of Primary School Students and Determ...

引用

2022 International Conference on data analytics for Business and Industry, ICDABI 2022

作者： Aldossary, Yasmeen Ebrahim, Majeed Alhaddad, Salman Zeki, Ahmed M. University of Bahrain MSc. in Big Data Science and Analytics College of Science Bahrain University of Bahrain College of Information Technology Department of Information System Bahrain

ISBN: (纸本)9781665490580

Educational organizations consistently attempt to investigate students' behaviors in learning and give early predictions for intervening and improving their learning performance, especially in primary education, which is the core stone of education. Educational data mining offers different practical prediction algorithms to predict student performance and various feature selection methods to determine the dominant factors which affect student performance. This study applied three data mining algorithms, K-Nearest Neighbor, Naive Bayes, and Decision Tree, to predict the performance of students based on their socio-demographic characteristics. The feature selection methods Information Gain Attribute Evaluation, Wrapper Subset Evaluation, and Classifier Subset Evaluation were used to identify the dominant factors. The research was done by adopting the CRISP methodology using a data set obtained from one of Bahrain's public primary boys' schools. The results showed that the model of the k-Nearest Neighbors algorithm with Wrapper Subset Evaluation as a feature selection method had achieved the highest accuracy rate of 67.6%. For the dominant factors, Student Age and Student Level have the most impact on student performance, followed by Parent education level. © 2022 IEEE.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

Parallel Implementation of KNN Algorithm for Dry Beans dataset

Parallel Implementation of KNN Algorithm for Dry Beans Datas...

引用

2022 International Conference on data analytics for Business and Industry, ICDABI 2022

作者： Hammad, Salman Alhaddad, Salman Yusuf, Husain Alqaddoumi, Abdulla University of Bahrain MSc. in Big Data Science and Analytics College of Science Bahrain College of Information Technology University of Bahrain Department of Information System Bahrain

ISBN: (纸本)9781665490580

In recent years, the interest in using machine learning to solve complex problems in different sectors using parallel algorithms has increased. The KNN algorithm is the most popular method to classify the dataset. The main problem in implementing this algorithm is how to quickly find K nearest neighbors. Accordingly, this research aims to study the effect of the parallel execution of the KNN algorithm by using different metrics such as speedup and efficiency. The performance of serial and parallel KNN algorithms was compared using different values of K and a different number of processers. The results show that The Message Passing Interface (MPI4py) improves performance by increasing speedup and decreasing execution time. Also, the impact of serial execution on the size of the dataset is discussed. It was found that as the size of the dataset increases, the serial time increases. Meanwhile, the use of seven processers in parallel execution with the KNN algorithm needs less time. In addition, as the number of processers increases, the speedup increases, and the efficiency decreases © 2022 IEEE.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Evolutionary Transfer Neural Architecture Search Across Spaces via Representation Learning

引用

IEEE Transactions on Evolutionary Computation 2025年

作者： Hou, Boyu Feng, Liang Chen, Xuefeng Tang, Jing Tan, Kay Chen Liao, Xiaofeng Chongqing University College of Computer Science China Chongqing Key Laboratory of Big Data Intelligence and Privacy Computing Chongqing400044 China The Hong Kong Polytechnic University Department of Data Science and Artificial Intelligence Hong Kong999077 Hong Kong Data Science and Analytics Thrust Guangzhou511453 China

Neural Architecture Search (NAS) has emerged as a crucial method for automating the design of deep learning models. Despite its potential, NAS frequently requires substantial computational and hardware resources. To mitigate these challenges, transferable NAS (TNAS) has been introduced, leveraging prior NAS results to enhance performance on new tasks. However, existing methods largely focus on knowledge transfer within identical neural search spaces, overlooking the potential for cross-domain transferability. Motivated by this gap, we explore evolutionary TNAS across heterogeneous search spaces by learning common neural representations. In particular, we introduce a novel approach that encodes both operational and topological information of neural architectures into a unified sequence using a simple tokenizer. This sequence is then processed by a variational auto-encoder, with a Transformer-based encoder to capture rich neural representations and a decoder that reconstructs the original sequence. By utilizing these latent representations, we further establish an inter-domain mapping that acts as a bridge, enabling effective explicit solution transfer among diverse search spaces to enhance the evolutionary NAS process. To harness this capability, we develop an evolutionary sequential transfer optimization approach that transfers knowledge during population initialization, providing both flexibility and adaptability. To the best of our knowledge, this work serves as the first attempt in the literature exploring evolutionary TNAS across diverse spaces. Moreover, we demonstrate the utility of our method through comprehensive empirical studies using different architecture spaces, including NAS-Bench-101, NAS-Bench-201, and the DARTS search space. Our results show that the proposed method significantly enhances the adaptability and performance of NAS across varied domains. © 1997-2012 IEEE.

关键词： Transfer learning

来源：评论

学校读者我要写书评

暂无评论

Reducing a Set of Regular Expressions and Analyzing Differences of Domain-specific Statistic Reporting

arXiv

引用

arXiv 2022年

作者： Kalmbach, Tobias Hoffmann, Marcel Lell, Nicolas Scherp, Ansgar Data Science and Big Data Analytics Group Ulm University Germany

Due to the large amount of daily scientific publications, it is impossible to manually review each one. Therefore, an automatic extraction of key information is desirable. In this paper, we examine STEREO, a tool for extracting statistics from scientific papers using regular expressions. By adapting an existing regular expression inclusion algorithm for our use case, we decrease the number of regular expressions used in STEREO by about 33.8%. We reveal common patterns from the condensed rule set that can be used for the creation of new rules. We also apply STEREO, which was previously trained in the life-sciences and medical domain, to a new scientific domain, namely Human-Computer-Interaction (HCI), and re-evaluate it. According to our research, statistics in the HCI domain are similar to those in the medical domain, although a higher percentage of APA-conform statistics were found in the HCI domain. Additionally, we compare extraction on PDF and LATEX source files, finding LATEX to be more reliable for extraction. Copyright © 2022, The Authors. All rights reserved.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

A Survey of Orthographic Information in Machine Translation

引用

SN Computer science 2021年第4期2卷 330页

作者： Chakravarthi, Bharathi Raja Rani, Priya Arcan, Mihael McCrae, John P. Unit for Linguistic Data Insight SFI Research Centre for Data Analytics Data Science Institute National University of Ireland Galway Galway Ireland Unit for Natural Language Processing Insight SFI Research Centre for Data Analytics Data Science Institute National University of Ireland Galway Galway Ireland

Machine translation is one of the applications of natural language processing which has been explored in different languages. Recently researchers started paying attention towards machine translation for resource-poor languages and closely related languages. A widespread and underlying problem for these machine translation systems is the linguistic difference and variation in orthographic conventions which causes many issues to traditional approaches. Two languages written in two different orthographies are not easily comparable but orthographic information can also be used to improve the machine translation system. This article offers a survey of research regarding orthography’s influence on machine translation of under-resourced languages. It introduces under-resourced languages in terms of machine translation and how orthographic information can be utilised to improve machine translation. We describe previous work in this area, discussing what underlying assumptions were made, and showing how orthographic knowledge improves the performance of machine translation of under-resourced languages. We discuss different types of machine translation and demonstrate a recent trend that seeks to link orthographic information with well-established machine translation methods. Considerable attention is given to current efforts using cognate information at different levels of machine translation and the lessons that can be drawn from this. Additionally, multilingual neural machine translation of closely related languages is given a particular focus in this survey. This article ends with a discussion of the way forward in machine translation with orthographic information, focusing on multilingual settings and bilingual lexicon induction. © 2021, The Author(s).

关键词： Machine translation Neural machine translation Orthography Rule-based machine translation Statistical machine translation Under-resourced languages

来源：评论

学校读者我要写书评

暂无评论

A Dynamic Cross-Scale Transformer with Dual-Compound Representation for 3D Medical Image Segmentation 48

A Dynamic Cross-Scale Transformer with Dual-Compound Represe...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Zhang, Ruixia Wang, Zhiqiong Wang, Zhongyang Xin, Junchang Northeastern University College of Medicine and Biological Information Engineering China Northeastern University School of Computer Science & Engineering China Northeastern University Key Laboratory of Big Data Management and Analytics Liaoning Province Shenyang China

ISBN: (纸本)9781728163277

Transformer models exploit multi-head self-attention to capture long-range information. Further, window-based self-attention solves the problem of quadratic computational complexity and provides a new solution for dense prediction of 3D images. However, Transformers miss structural information due to the naive tokenization scheme. Furthermore, single-scale attention fails to achieve a balance between feature representation and semantic information. Aiming at the above problems, we propose a window-based dynamic crossscale cross-attention transformer (DCS-Former) for precise representation of the diversity features. DCS-former first constructs dual-compound feature representations through Antehoc Structure-aware Module and Post-hoc Class-aware Module. Then, the bidirectional attention structure is designed to interactively fuse structural features with class representations. The experimental results show that our method outperforms various competing segmentation models on three different public datasets. © 2023 IEEE.

关键词： Deep Learning Medical Image Segmentation Self-attention Vision Transformer

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：