检索结果-内蒙古大学图书馆

11th International Conference on computer engineering and Knowledge, ICCKE 2021

作者： Moradi, Ashkan Shekofteh, Yasser Zarei, Saeed Shahid Beheshti University Faculty of Computer Science and Engineering Tehran Iran

ISBN: (纸本)9781665402088

Automatic Spoken language identification (LID) refers to the automatic process of identifying languages spoken in the audio files. Pure acoustic approaches have shown great potential in LID. Since acoustic approaches have become more and more popular, phonetic information has been largely overlooked. In this paper, we present a genetic-based fusion approach based on the score probabilities of two phonetic LID systems. There are two SVM classifiers trained on perplexities as their feature vectors which are obtained from phone language models of different phone recognizers. Two phone recognizers are here utilized;one decodes the speech file to a sequence of IPA alphabet, as a universal phone recognizer, and the other is a Farsi phone recognizer which is trained on FARSDAT databases. With the help of the genetic-based fusion approach, we will extract 54 weights. We have 27 languages in our database and 2 individual phonetic LID systems;therefore, we will achieve 54 weights for our fusion. The first 27 weights correspond to our system using a universal phone recognizer and the second 27 weights are related to our system with the Farsi phone recognizer. In the end, we use these weights to combine the results of each of our individual phonetic LID systems. The experimental results conducted on 27 languages within the NIST-LRE09 corpus demonstrated that the proposed fusion approach could greatly increase the classification accuracy of target languages. It should also be noted that we separate the files of each speaker and place them only in one set (train set, development set, or test set) to prevent speaker-related biases. © 2021 IEEE.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

A Systematic Embedded Software Design Flow for Robotic Applications 11

A Systematic Embedded Software Design Flow for Robotic Appli...

引用

11th International Conference on computer engineering and Knowledge, ICCKE 2021

作者： Mahdian, Navid Attarzadeh-Niaki, Seyed-Hosein Salimi-Badr, Armin Shahid Beheshti University Faculty of Computer Science and Engineering Tehran Iran

ISBN: (纸本)9781665402088

This paper presents an end-to-end high-level design flow for development and testing of robotic embedded software. Our design flow's well-organized steps aim to minimize the potential issues of the software before the final implementation as well as the total development time. The proposed design flow devises a clear interface between the mechanics/hardware and the controller/software teams and is capable of being used in a wide range of robotic applications. Using our design flow, controllers can be developed thoroughly without the necessity of code adjustment. We demonstrate the applicability of our design flow using a synthetic case study and a real-world in-house mobile robot. © 2021 IEEE.

关键词： Robotics

来源：评论

学校读者我要写书评

暂无评论

Short-Term Traffic Flow Prediction Based on a Recurrent Deep Neural Network: a Study in Tehran 12

Short-Term Traffic Flow Prediction Based on a Recurrent Deep...

引用

12th International Conference on Information and Knowledge Technology, IKT 2021

作者： Abdoos, Monireh Vajedsamiei, Taha Shahid Beheshti University Faculty of Computer Science and Engineering Tehran Iran

ISBN: (纸本)9781665408493

With growing of population, the issue of optimal mobility between two points of the city has become one of the most important problems. There are various tools to suggest the optimal route, but due to the momentary changes in traffic in cities, especially large cities, providing the optimal route without predicting the traffic load will not be accurate. In this regard, it can be noted that one of the most widely used up-to-date methods is the use of deep neural networks to predict the future. In this paper, while examining some of the most widely used deep neural networks to predict traffic sequence, a method is presented based on one of recurrent neural networks. The method has been evaluated on real traffic data on a part of Tehran. The results show that the proposed method outperforms the other similar neural networks. © 2021 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Text Classification Accuracy Enhancement Using Deep Neural Networks

Text Classification Accuracy Enhancement Using Deep Neural N...

引用

2023 Al-Sadiq International Conference on Communication and Information Technology, AICCIT 2023

作者： Najm, Mobajjal Kadhim Aljumaily, Mustafa S. Al-Khaleefa, Ahmed Ayal, A.M. Mohammed, Dheyaa Abdulameer Challoob, Aqeel Luaibi Shahid Chamran University of Ahvaz Faculty of Engineering Computer Engineering Department Ahvaz Iran University of Tennessee Electrical Engineering and Computer Science Department Tennessee United States University of Misan Faculty of Education Department of Physics Maysan Iraq University of Misan Maysan Iraq Imam Ja'afar Al-Sadiq University College of Information Technology Department of Computer Technical Engineering Al-Muthanna Iraq Department of Computer Techniques Engineering1 Baghdad Iraq

ISBN: (纸本)9798350341881

Machine-learning systems are part of the field of artificial intelligence that automatically learn models from data to make better decisions. Natural language processing (NLP), using object and learning approaches, performs well in statistical tasks, such as text classification or sentiment mining. Currently, in the field of natural language processing (NLP), various models, methods, and architectures have been proposed and a wide range of NLP tasks have been developed. Most natural language processing (NLP) problems can be formulated as classification problems (decide on the class of this object depending on the object and its context). In this field, machine learning methods and especially using neural network models are the most used. In this research, we propose a deep learning architecture for text classification that can influence model performance. The basic steps of this proposed method include text pre-processing (including sentence segmentation, punctuation, rooting, speech tagging, and vector conversion), text feature extraction, and LSTM algorithm-based classification model construction. The dataset used in this study is 20NewsGroup and IMDB. The evaluation results are provided in the form of MSE, Precision, Recall, Accuracy, and F-Measure evaluation criteria. © 2023 IEEE.

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

RFBES at SemEval-2024 Task 8: Investigating Syntactic and Semantic Features for Distinguishing AI-Generated and Human-Written Texts

arXiv

引用

arXiv 2024年

作者： Rad, Mohammad Heydari Farsi, Farhan Bali, Shayan Etezadi, Romina Shamsfard, Mehrnoush Computer Engineering Department Amirkabir University of Technology Tehran Iran School of Electrical Engineering and Computer Science University of Ottawa Ottawa Canada Faculty of Computer Science and Engineering Shahid Beheshti University Tehran Iran

Nowadays, the usage of Large Language Models (LLMs) has increased, and LLMs have been used to generate texts in different languages and for different tasks. Additionally, due to the participation of remarkable companies such as Google and OpenAI, LLMs are now more accessible, and people can easily use them. However, an important issue is how we can detect AI-generated texts from human-written ones. In this article, we have investigated the problem of AI-generated text detection from two different aspects: semantics and syntax. Finally, we presented an AI model that can distinguish AI-generated texts from human-written ones with high accuracy on both multilingual and monolingual tasks using the M4 dataset. According to our results, using a semantic approach would be more helpful for detection. However, there is a lot of room for improvement in the syntactic approach, and it would be a good approach for future work. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Improving Enterprise Architecture Mining Methods

Improving Enterprise Architecture Mining Methods

引用

Advances in Enterprise Architecture (NCAEA), National Conference on

作者： Mohsen Mohammadi Nezhad Fereidoon Shams Aliee Faculty of Computer Science and Engineering SOEA Lab Shahid Beheshti University Tehran Iran

Enterprise architecture takes a step in the path of digital transformation by providing a comprehensive system with an integrated view of the organization. In traditional methods, architects manually prepare enterprise architecture products. The manual methods of producing enterprise architecture products have many challenges. One of the main challenges is the production of volumes of documentation, the complexity and difficulty of making enterprise architecture artifacts, and their costly and error-prone nature. In response to these challenges, enterprise architecture mining is proposed, automatically producing outputs and artifacts of enterprise architecture instead of manual modeling. Surveys show that enterprise architecture analysis still needs to be mature, and full automation is still far away. Some methods may be successful in some layers, but reaching a method that automatically provides an integrated enterprise architecture model with sufficient accuracy remains the *** research will use model analysis methods to improve enterprise architecture mining. At first, the presented method extracts enterprise architecture models automatically. Detecting architecture smells reveals enterprise architecture debts in the next step. Ultimately, it improves the extracted models using refactoring methods. The proposed method can improve the accuracy and quality of the extracted models.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Knowledge-Based Approach for Recognizing Textual Entailments with a Focus on Causality and Contradiction

SSRN

引用

SSRN 2023年

作者： Rahimi, Zeinab ShamsFard, Mehrnoush NLP lab Faculty of computer science and engineering Shahid Beheshti University Tehran Iran

In natural language texts, some information, is not explicitly observed in the sentence and can be deduced from the text using the background or commonsense knowledge. Recognition of textual entailments is used to extract such information which receives two pieces of texts and decides if the second one is an entailment or contradiction to the first one or there is no relation between them. One of the entailment types is *** event or situation in a text that can cause the event or situation in another *** paper is dedicated to enhancing textual entailment by recognizing causality and contradiction relations among pieces of texts, employing knowledge ***, we use a knowledge-based method that recognizes textual entailments by the use of linguistic and common-sense knowledge (ontologies), with a focus on contradiction and causal relations including implicit or ambiguous *** this purpose, we first developed two ontologies including causality and contradiction relations among concepts. Then two concurrent paths in a pipeline are traversed. In the first path a rule-based system is used to recognize simple contradiction *** rules are extracted semi automatically using datamining methods. In the second path the provided knowledge is injected into a language model using prompt engineering. The enriched language models are then fine-tuned by entailment datasets to determine textual *** results show that the combination of the rule-based method and the enriched models is the best. Our best system has accuracy of 87.5% on RTE tasks trained and tested on Farstail with 3.5% enhancement over other systems. © 2023, The Authors. All rights reserved.

关键词： Ontology

来源：评论

学校读者我要写书评

暂无评论

Improving Energy Consumption In IoT Networks: Reducing Sensors Energy By Timing Control

Improving Energy Consumption In IoT Networks: Reducing Senso...

引用

2023 Al-Sadiq International Conference on Communication and Information Technology, AICCIT 2023

作者： Al-Taie, Ali Hameed Luaibi Aljumaily, Mustafa S. Ayal, A.M. Al-Khaleefa, Ahmed Dakhil, Adel Manaa Alsalamy, Ali Shahid Chamran University of Ahvaz Faculty of Engineering Computer Engineering Department Ahvaz Iran University of Tennessee Electrical Engineering and Computer Science Department KnoxvilleTN United States University of Misan Maysan Iraq University of Misan Faculty of Education Department of Physics Maysan Iraq University of Misan College of Engineering Department of Electrical Engineering Maysan Iraq Imam Ja'afar Al-Sadiq University College of Information Technology Department of Computer Technical Engineering Al-Muthanna Iraq

ISBN: (纸本)9798350341881

The number of sensors is constantly increasing as a result of the urgent need for smart homes, factories, hospitals, military facilities, etc. Consequently, the energy consumption of these mostly battery-driven devices has increased, and due to their continuous work without stopping, it became necessary to think about and study to reduce energy consumption. In this paper, we are suggesting reducing energy consumption by establishing a schedule to turn the IoT sensors on and off according to the user demand and under their control. This research proposed a different On/Off time scenario when sensors are distributed as groups, each connected to gateways (Microcontrollers), which in turn are connected to the main server via a wireless connection, the Microcontrollers continuously check the readings sent by these sensors. The results showed the proposed algorithm increased the residual energy of the network by assigning sleeping intervals depending on the covered area and the number of neighboring nodes. © 2023 IEEE.

关键词： Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

χ perbp: a cloud-based lightweight mutual authentication protocol

引用

Peer-to-Peer Networking and Applications 2023年第4期16卷 1785-1802页

作者： Adeli, Morteza Bagheri, Nasour Sadeghi, Sadegh Kumari, Saru Faculty of Electrical and Computer Engineering Malek Ashtar University of Technology Tehran1774-15875 Iran Electrical Engineering Department Shahid Rajaee Teacher Training University Tehran16788-15811 Iran Zanjan45137-66731 Iran Zanjan45137-66731 Iran Department of Mathematics Chaudhary Charan Singh University Meerut250004 India School of Computer Science Institute for Research in Fundamental Sciences IPM 19395-5746 Tehran Iran

Cloud-based RFID is gaining popularity in tandem with the growth of cloud computing and the Internet of Things (IoT). The cloud-based RFID system is developed with the intent of providing real-time data that can be sent into the cloud for easy access and interpretation. The security and privacy of constrained devices in these systems is a challenging issues for many applications. To deal with this problem, we first introduce χ per, as a new hardware/software friendly component that can be implemented using bit-wise operations and extensively analyze its security. Next, we propose χ perbp, a lightweight authentication protocol based on χ per component. To evaluate the performance efficiency of our proposed scheme, we implement the χ perbp scheme on an FPGA module Xilinx Kintex-7 using the hardware description language VHDL. Our security and cost analysis of the proposed protocol shows that the proposed protocol provides desired security against various attacks, at a reasonable cost. Also, formal security evaluation using BAN logic and the Scyther tool indicates its security correctness. Besides, we analyze the security of a related protocol which has been recently proposed by Fan et al. It is a cloud-based lightweight mutual authentication protocol for RFID devices in an IoT system. The authors have claimed that their scheme is secure against active and passive attacks, however, our detailed security analysis in this paper demonstrates the major drawbacks of this protocol. More precisely, the proposed attack discloses the tag’s secrets efficiently. Given the tag’s secrets, any other attack will be trivial. © 2023, The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature.

关键词： Authentication

来源：评论

学校读者我要写书评

暂无评论

Uncovering Secrets of Microbursts in Datacenter Network Traffic

Uncovering Secrets of Microbursts in Datacenter Network Traf...

引用

International Conference on Network and Service Management

作者： Mohammad Hosseini Sina Darabi Mohammad Nakhjiri Patrick Eugster Faculty of Computer Science and Engineering Shahid Beheshti University Tehran Iran The Faculty of Informatics Università Della Svizzera Italiana (USI) Lugano Switzerland

ISBN: (数字)9783903176669

ISBN: (纸本)9798331505158

Designing efficient methods and policies for mitigating microbursts requires a thorough understanding of microburst characteristics and behaviors. However, the lack of detailed studies on microburst characteristics and comprehensive tools for measuring and analyzing them has been a significant challenge for researchers in this field. We introduce BurstVision, a tool that extracts various characteristics of microbursts from traffic traces. Using BurstVision, we analyze several traffic traces from various cloud datacenter applications and report on the diverse characteristics of microbursts observed. Our analysis reveals that microburst characteristics significantly vary across applications. Moreover, we discuss how these varying characteristics can influence the effectiveness of different microburst mitigation solutions. Our findings highlight the importance of considering the specific type and characteristics of microbursts in traffic when adopting a microburst mitigation solution.

关键词： Prevention and mitigation Telecommunication traffic Tuning

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：