检索结果-内蒙古大学图书馆

Decoding Queries: An In-Depth Survey of Quality Techniques for Question Analysis in Arabic Question Answering Systems

引用

IEEE ACCESS 2024年 12卷 135241-135264页

作者： Essam, Mariam Deif, Mohanad A. Attar, Hani Alrosan, Ayat Kanan, Mohammad A. Elgohary, Rania Misr Univ Sci & Technol MUST Coll Informat Technol Dept Artificial Intelligence 6th October City 12566 Egypt Zarqa Univ Fac Engn Zarqa 13110 Jordan Univ Business & Technol Coll Engn Jeddah 21448 Saudi Arabia Skyline Univ Coll Sch Comp Sharjah U Arab Emirates Ain Shams Univ Fac Comp & Informat Sci Informat Syst Dept Cairo 11566 Egypt

In the field of natural language processing (NLP), natural language understanding (NLU) plays a critical role in transforming human languages into machine-interpretable formats. This paper provides an overview of methodologies and resources that have been developed so far concerning Arabic QAS, focusing on NLU regarding question analysis and classification. These components perform an important role in obtaining accurate, quality, context-sensitive answers. Findings indicate that deep learning models work wonders for complex languages, but machine learning algorithms usually do the job in most classification tasks. Further, there is mention of the potential of rule-based and hybrid approaches, whose research in the future should be integrated with evolving evaluation methods necessary to keep pace with the advances in NLP. Challenges especially pertinent to Arabic QAS are complex syntax, dialectal diversity, limited tool support, and a lack of benchmark datasets. Other directions for the future are the development of complete datasets, standardized test bed frameworks, and extra question classification, adopting a hybrid approach considering interrogative words along with question multiplicity. This survey would therefore be helpful in highlighting shortcomings in the literature, suggesting new directions for research, and emphasizing that innovation concerning NLU within QASs is an ongoing necessity.

关键词： Surveys Natural language processing Question answering (information retrieval) Chatbots Semantics Search engines machine learning algorithms Natural language understanding (NLU) Arabic question answering systems Arabic understanding question analysis challenges in Arabic question analysis

来源：评论

学校读者我要写书评

暂无评论

Where Does Minimum Error Entropy Outperform Minimum Mean Square Error? A New and Closer Look

引用

IEEE ACCESS 2018年 6卷 5856-5864页

作者： Heravi, Ahmad Reza Hodtani, Ghosheh Abed Ferdowsi Univ Mashhad Elect Engn Dept Mashhad *** Razavi Khorasan Iran

The past decade has seen a rapid application of information theoretic learning (ITL) criteria in robust signal processing and machine learning problems. Generally, in ITL's literature, it is seen that, under non-Gaussian assumptions, especially when the data are corrupted by heavy-tailed or multi-modal non-Gaussian distributions, information theoretic criteria [such as minimum error entropy (MEE)] outperform second order statistical ones. The objective of this research is to investigate this better performance of MEE criterion against that of minimum mean square error. Having found similar results for MEE- and MSE-based methods, in the non-Gaussian environment under particular conditions, we need a precise demarcation between this occasional similarity and occasional outperformance. Based on the theoretic findings, we reveal a better touchstone for the outperformance of MEE versus MSE.

关键词： Entropy mean square error methods machine learning algorithms information theoretic learning Kullback-Leibler divergence

来源：评论

学校读者我要写书评

暂无评论

Multilabel Ranking With Inconsistent Rankers

引用

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND machine INTELLIGENCE 2022年第9期44卷 5211-5224页

作者： Geng, Xin Zheng, Renyi Lv, Jiaqi Zhang, Yu Southeast Univ Sch Comp Sci & Engn Minist Educ Nanjing 211189 Peoples R China Southeast Univ Key Isib Comp Network & Informat Integrat Minist Educ Nanjing 211189 Peoples R China

While most existing multilabel ranking methods assume the availability of a single objective label ranking for each instance in the training set, this paper deals with a more common case where only subjective inconsistent rankings from multiple rankers are associated with each instance. Two ranking methods are proposed from the perspective of instances and rankers, respectively. The first method, Instance-oriented Preference Distribution learning (IPDL), is to learn a latent preference distribution for each instance. IPDL generates a common preference distribution that is most compatible to all the personal rankings, and then learns a mapping from the instances to the preference distributions. The second method, Ranker-oriented Preference Distribution learning (RPDL), is proposed by leveraging interpersonal inconsistency among rankers, to learn a unified model from personal preference distribution models of all rankers. These two methods are applied to natural scene images dataset and 3D facial expression dataset BU_3DFE. Experimental results show that IPDL and RPDL can effectively incorporate the information given by the inconsistent rankers, and perform remarkably better than the compared state-of-the-art multilabel ranking algorithms.

关键词： Training Predictive models Adaptation models Task analysis machine learning machine learning algorithms Encoding Multilabel ranking instance-oriented preference distribution learning ranker-oriented preference distribution learning machine learning

来源：评论

学校读者我要写书评

暂无评论

Gradual-N-Justification (GNJ) to Reduce False-Positive Hardware Trojan Detection in Gate-Level Netlist

引用

IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS 2022年第4期30卷 515-525页

作者： Salmani, Hassan Howard Univ Dept Elect Engn & Comp Sci Washington DC 20059 USA

The integrated circuit design flow is highly susceptible to hardware Trojan (HT) insertion. While there have been significant efforts to detect HTs, techniques usually demand a golden model, suffer limited scalability, or experience high false positives. This article introduced a new technique called gradual-N-justification (GNJ) to reduce false-positives HT detection in the gate-level netlist. The GNJ technique combines the signal justification and unsupervised K-means machine learning (ML) algorithm. The GNJ technique is a general technique that can be applied to a set of reported suspicious signals (SSs) in order to identify the most SSs and reduce false-positive rates (FPRs) in detecting HTs. The GNJ technique is applied to 60 different combinations of full-scan and partial-scan circuits and hard-to-detect combinational HTs. To realize valid and hard-to-detect HTs, a configurable HT insertion platform is developed. Furthermore, to the best of our knowledge, it is for the first time that an extensive evaluation of partial-scan circuits for HT detection at the gate level is being performed. The comprehensive results on both full- and partial-scan circuits have shown that the GNJ technique is highly scalable as it presents a linear relationship between the number of SSs and the execution time of GNJ. Furthermore, the GNJ technique does not miss an HT circuit if exists. The GNJ technique offers an FPR as low as 3.89% for full-scan circuits and 3.31% for partial-scan circuits on average.

关键词： Integrated circuits Logic gates machine learning algorithms Trojan horses Hardware Clustering algorithms Integrated circuit modeling False-positive rate (FPR) gradual-N-justification (GNJ) hardware Trojans (HT) machine learning (ML) postprocessing technique

来源：评论

学校读者我要写书评

暂无评论

A CNN-Based Structured Light Communication Scheme for Internet of Underwater Things Applications

引用

IEEE INTERNET OF THINGS JOURNAL 2020年第10期7卷 10038-10047页

作者： Trichili, Abderrahmen Ben Issaid, Chaouki Ooi, Boon S. Alouini, Mohamed-Slim King Abdullah Univ Sci & Technol Comp Elect & Math Sci & Engn Div Thuwal 239556900 Saudi Arabia

Underwater optical wireless communication is an emerging field that can provide reliable connectivity for future generation Internet of Underwater Things devices. In this article, we propose a communication system based on single and superposition of Laguerre-Gaussian modes to transfer information and rely on a convolutional neural network for the mode identification in an underwater environment. A 100% recovery fidelity is reported at clear and turbid water. Beyond 90% of identification, accuracy is achieved under different laboratory-emulated underwater turbulence conditions. The practical implementation of the proposed spatial-mode-based communication scheme is further discussed.

关键词： Structural beams Convolutional neural networks machine learning algorithms Encoding Laser beams Indexes Convolutional neural network (CNN) index modulation Internet of Underwater Things (IoUT) spatially structured light underwater wireless optical communication (UWOC)

来源：评论

学校读者我要写书评

暂无评论

An Automatic Framework Recognizing the Relationships of Cultural Heritage

引用

IEEE ACCESS 2024年 12卷 171689-171705页

作者： Zhang, Zizhan Zhou, Zijun Cao, Yingchun Hebei Univ Architecture Sch Architecture & Art Zhangjiakou 075000 Hebei Peoples R China Jilin Univ Sch Artificial Intelligence Changchun 130012 Jilin Peoples R China Jiangsu Univ Sci & Technol Sch Civil Engn & Architecture Zhenjiang 212003 Jiangsu Peoples R China

The field of cultural heritage has developed over a long period, accumulating a wealth of research findings. Researchers are now focusing on systematic relationships and taxonomic studies of heritage, exploring the underlying cultural information embedded within. Inspired by the fields of machine learning and biology, we propose a research approach that combines data processing with unsupervised algorithms ("Feature Sparsity Module + N"), which can be utilized to unveil the systematic relationships of cultural heritage study subjects. We construct the Cultural Heritage Relationship Evaluation Framework (CHREF), framework using the structure of "FSM + PCA + HCA", which offers a workflow characterized by interpretability, visualization capabilities, and automation. The framework utilizes the FSM module to transform the research subjects into matrices, employing PCA and HCA to obtain intuitive charts and reliable data results with minimal manual intervention. Additionally, we provide experiments and a user study on traditional Chinese brick kilns to validate the effectiveness and universality of the proposed framework. The "FSM + N" methodology and CHREF can provide tools for various stages of work in cultural heritage, making significant contributions to the digital development and database construction in the field of cultural heritage.

关键词： Cultural differences Principal component analysis Vectors Data analysis machine learning algorithms Geoscience Unsupervised learning Systematics Matrix decomposition Encoding Cultural heritage machine learning interdisciplinary research embedding one-hot encoding principal component analysis hierarchical clustering analysis traditional Chinese kilns

来源：评论

学校读者我要写书评

暂无评论

Joint Pedestrian Motion State and Device Pose Classification

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2020年第8期69卷 5862-5874页

作者： Kasebzadeh, Parinaz Radnosrati, Kamiar Hendeby, Gustaf Gustafsson, Fredrik Linkoping Univ Dept Elect Engn S-58183 Linkoping Sweden

Novel features for joint classification of gait and device modes are proposed and multiple machine learning methods are adopted to jointly classify the modes. The classification accuracy as well as the F-1 score of two standard classification algorithms, K-nearest neighbor (KNN) and Gaussian process (GP), are evaluated and compared against a proposed neural network (NN)-based classifier. The proposed features are the correlation scores of a detected gait cycle relative to a set of unique gait signatures as well as the gait cycle time, all extracted from handheld inertial measurement units (IMUs). Each gait signature is defined such that it contains one full cycle of the human gait. In order to take the temporal correlation between classes into account, the initial classifiers' estimates are fed into a hidden Markov model (HMM) unit to obtain the final class estimates. The performance of the proposed method is evaluated on a large data set, including two classes of gait modes (walking and running) and four classes of device modes (fixed and face-up in the hand, swinging in the hand, in the pocket, and in the backpack). The experimental results validate the reliability of the considered features and effectiveness of the HMM unit. The initial classification accuracy of the NN-based approach is 91%, which is further improved to 99% after the smoothing stage on the validation set and 98% on the test set.

关键词： Feature extraction Hidden Markov models Smart phones Sensors Legged locomotion machine learning machine learning algorithms Activity classification feedforward neural network (NN) gaussian process (GP) hidden markov model (HMM) inertial measurement unit (IMU) pedestrian gait analysis

来源：评论

学校读者我要写书评

暂无评论

An Evolutionary Forest for Regression

引用

IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION 2022年第4期26卷 735-749页

作者： Zhang, Hengzhe Zhou, Aimin Zhang, Hu East China Normal Univ Sch Comp Sci & Technol Shanghai Inst AI Educ Shanghai 200241 Peoples R China Beijing Electromech Engn Inst Sci & Technol Complex Syst Control & Intelligent Beijing 100074 Peoples R China

Random forest (RF) is a type of ensemble-based machine learning method that has been applied to a variety of machine learning tasks in recent years. This article proposes an evolutionary approach to generate an oblique RF for regression problems. More specifically, our method induces an oblique RF by transforming the original feature space to a new feature space through the evolutionary feature construction method. To speed up the searching process, the proposed method evaluates each set of features based on a decision tree (DT) rather than an RF. In order to obtain an RF, we archive top-performing features and corresponding trees during the search. In this way, both the features and the forest can be constructed simultaneously in a single run. The proposed evolutionary forest is applied to 117 benchmark problems with different characteristics and compared with some state-of-the-art regression methods, including several variants of the RF and gradient boosted DTs (GBDTs). The experimental results suggest that the proposed method outperforms the existing RF and GBDT methods.

关键词： Random forests Decision trees Forestry machine learning algorithms Regression tree analysis Optimization Boosting Evolutionary feature construction evolutionary forest (EF) genetic programming (GP) random forest (RF)

来源：评论

学校读者我要写书评

暂无评论

Intrusion Detection Based on Sequential Information Preserving Log Embedding Methods and Anomaly Detection algorithms

引用

IEEE ACCESS 2021年 9卷 58088-58101页

作者： Kim, Czangyeob Jang, Myeongjun Seo, Seungwan Park, Kyeongchan Kang, Pilsung Korea Univ Sch Ind Management Engn Seoul 02841 South Korea Univ Oxford Dept Comp Sci Oxford OX1 3QD England

Previous methods for system intrusion detection have mainly consisted of those based on pattern matching that employs prior knowledge extracted from experts' domain knowledge. However, pattern matching-based methods have a major drawback that it can be bypassed through various modified techniques. These advanced persistent threats cause limitation to the pattern matching-based detecting mechanism, because they are not only more sophisticated than usual threats but also specialized in the targeted attacking object. The defense mechanism should have to comprehend unusual phenomenons or behaviors to successfully handles the advanced threats. To achieve this, various security techniques based on machine learning have been developed recently. Among these, anomaly detection algorithms, which are trained in unsupervised fashion, are capable of reducing efforts of security experts and securing labeled dataset through post analysis. It is further possible to distinguish abnormal behaviors more precisely by training classification models if sufficient amounts of labeled dataset is obtained through post analysis of anomaly detection results. In this study, we proposed an end-to-end abnormal behavior detection method based on sequential information preserving log embedding algorithms and machine learning-based anomaly detection algorithms. Contrary to other machine learning based system anomaly detection models, which borrow domain experts' knowledge to extract significant features from the log data, raw log data are transformed into a fixed size of continuous vector regardless of their length, and these vectors are used to train the anomaly detection models. Experimental results based on a real system call trace dataset, our proposed log embedding method with unsupervised anomaly detection model yielded a favorable performance, at most 0.8708 in terms of AUROC, and it can be further improved up to 0.9745 with supervised classification algorithms if sufficient labeled attack lo

关键词： Intrusion detection Anomaly detection machine learning algorithms Data models Training Support vector machines Feature extraction System anomaly detection cyber security system log embedding advanced persistent threat ADFA-LD

来源：评论

学校读者我要写书评

暂无评论

APPROXIMATE INFERENCE OF FUNCTIONAL-DEPENDENCIES FROM RELATIONS

引用

THEORETICAL COMPUTER SCIENCE 1995年第1期149卷 129-149页

作者： KIVINEN, J MANNILA, H UNIV HELSINKI DEPT COMP SCI SF-00014 HELSINKI FINLAND

The functional dependency inference problem is the following. Given a relation I, find a set of functional dependencies that is equivalent with the set of all functional dependencies holding in r. All known algorithms for this task have running times that can be in the worst case exponential in the size of the smallest cover of the dependency set. We consider approximate dependency inference. We define various measures for the error of a dependency in a relation. These error measures have the value 0 if the dependency holds and a value close to 1 if the dependency clearly does not hold. Depending on the measure used, all dependencies with error at least epsilon in r can be detected with high probability by considering only 0(1/epsilon) or 0(/r/(1/2)/epsilon) random tuples of v. We also show how a machine learning algorithm due to Angluin et al. can be applied to give in output-polynomial time an approximately correct cover for the set of dependencies holding in v.

关键词： Armstrong relations Design DEPENDENCE Set (Psychology) machine learning algorithms

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：