检索结果-内蒙古大学图书馆

computers and Communications (ACCC), Asia Conference on

作者： Yasir Giani Luo Ping Syed Asad Shah Software School Tsinghua University Beijing China Dept.of Computer Engineering Bilkent University Ankara Turkey

Code clones make software maintenance more challenging. Detecting bugs in large systems may significantly increase maintenance costs. Despite the fact that several techniques for clone identification have been proposed over the years, the accuracy and scalability of clone detection techniques remain hot research areas. Previously, Akram et al. proposed the DroidCC hybrid technique, where tokens were encoded into MD5 hash values by encoding them into 128-bit fingerprints, and clones were identified by matching identical hash values. Encoding tokens into MD5 hash values take more time due to the large fingerprint size of MD5 hash values. Due to the enormous chunk size, DroidCC cannot achieve higher accuracy. To overcome the weakness of the DroidCC technique, We proposed a novel AYAT a lightweight hybrid technique to detect clones at the fragment level. To speed up the detection process, we converted tokens into 32-bit polynomial values, and we set the chunk size to 5 lines per chunk to improve accuracy. We tested our technique on 10,968 java projects against 4.98 million lines of code. In comparison to the well-known DroidCC technique, it is significantly faster and more efficient. Our examination demonstrates that precision is significantly improved despite sacrificing scalability. AYAT code cloning detection technique has outscored DroidCC in every aspect.

关键词： computers software maintenance Java Codes Costs Scalability Cloning

来源：评论

学校读者我要写书评

暂无评论

Deep Neural Networks in Social Media Forensics: Unveiling Suspicious Patterns and Advancing Investigations on Twitter

Deep Neural Networks in Social Media Forensics: Unveiling Su...

引用

Intelligent Cybersecurity Conference (ICSC)

作者： Yousef Sharrab Dimah Al-Fraihat Mohammad Alsmirat Dept. of Data Science and Artificial Intelligence Isra University Amman Jordan Dept.of Software Engineering Isra University Amman Jordan Dept. of Computer Science University of Sharjah Sharjah UAE

Text data forensics, a rapidly developing field that focuses on analyzing textual content to identify criminal or suspicious activities, is becoming increasingly important due to the popularity and the huge number of text posts on social media platforms. This study aims to improve the detection of suspicious text on social media using deep neural networks. Suspicious text is defined as any text that is likely to be associated with criminal activity or is unusual or out of the ordinary. This study could make a significant contribution to the field of text data forensics by helping to improve the detection of such text. We leveraged the “CIC Truth Seeker Dataset 2023” [1], which is widely recognized as a comprehensive and representative dataset for text data forensics research. The dataset contains over 180,000 tweets related to 700 real and 700 fake pieces of news, labeled by experts. In this study, we enhance text data forensics in social media by leveraging the powerful analytical capabilities of deep neural networks. More specifically, we investigate the effectiveness of Long Short-Term Memory (LSTM) in the detection of suspicious text. The results are very promising as we achieved an accuracy of 96% during preliminary evaluations. We plan to explore future work on the model's potential applications, including criminal activity identification, misinformation detection, and online harassment prevention.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Vote Bank Forecast Based on Social Media Sentiment Analysis

Vote Bank Forecast Based on Social Media Sentiment Analysis

引用

International Conference on Open Source Systems and Technologies (ICOSST)

作者： Zubair Zaland Sibghat Ullah Bazai Muhammad Imran Ghafoor Saira Akram Muhammad Waseem Muhammad Qaseem Dept. of Software Engineering Balochistan University of Information Technology Engineering and Management Sciences (BUITEMS) Quetta Pakistan Dept.of Computer Engineering Balochistan University of Information Technology Engineering and Management Sciences (BUITEMS) Quetta Pakistan Dept. of Engineering Pakistan Television Corporation Lahore Pakistan

Sentiment Analysis of human behavior from social media revolutionized many fields, such as health, business, and election prediction. Sentiment analysis is the process of analyzing an individual’s opinions, attitudes, views, sentiments, and emotions. In order to predict sentiments, predicting the polarity of words and categorizing them as positive or negative are the major goals. A quick and effective way to determine what the public thinks about one’s own political party, the opposition party, and political figures is through Twitter. prior sentiment analysis techniques have primarily struggled with classification accuracy because biases in the training data cause a sizable portion of tweets to be incorrectly classified. As a remedy, we set out to collect real-time data in order to forecast Pakistan’s next election results. Users can quickly share real-time updates on a variety of activities and events on Twitter, promoting the rapid diffusion of information and news. To do this, we accessed the Twitter API through the “Tweepy’’ Python package and retrieved real-time data from the service. Through keyword searches, we were able to find more than 5,000 tweets. We attributed feelings to particular tweets using Python’s “TextBlob’’ package. We are able to generate prediction graphics by determining the frequency of positive attitudes with respect to a particular candidate.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Evaluating machine learning techniques on human activity recognition using accelerometer data

Evaluating machine learning techniques on human activity rec...

引用

2020 International Conference on UK-China Emerging Technologies, UCET 2020

作者： Khan, Rimsha Abbas, Muhammad Anjum, Rubia Waheed, Fatima Ahmed, Sheeraz Bangash, Fahad Dept.of Computer and Software Engineering Islamabad Pakistan Iqra National University Dept.of Computer Science Peshawar Pakistan Bahria University Dept.of Electrical Engg Islamabad Pakistan

ISBN: (数字)9781728194882

ISBN: (纸本)9781728194882

Human activity recognition is gaining increasing importance because of its implication in remote monitoring application including security, health and fitness apps. This paper provides an analysis of different machine learning techniques for recognizing human activity. Firstly, all the recent work related to human activity recognition using accelerometer data is analyzed and presented in the paper. In this study the accelerometer used in smartphones as well as those embedded in wearable devices are compared and recognition methodologies applied on both the devices are presented. The dataset used in this project is a transformed version of 'Activity Recognition using Cell Phone Accelerometers,' by the Wireless Sensor Data Mining WSDM. Some important features were extracted from the data and based on it different models were assessed using Matlab Classification Learner App. Four distinct machine learning techniques were applied on the dataset, namely, linear regression, logistic regression, support vector machine and neural network. For the purposed of applying classifier Weka tool is used. The results of these algorithms are compared and presented in the form of tables and graphs and Bagged Tree is identified to be the best algorithm based on accuracy results. © 2020 IEEE.

关键词： Accelerometers

来源：评论

学校读者我要写书评

暂无评论

An Integrated Approach to Enhance Legal Information Retrieval of Sri Lankan Supreme Court Verdicts

An Integrated Approach to Enhance Legal Information Retrieva...

引用

Advancements in Computing (ICAC), International Conference on

作者： Asel Jayasooriya Akeel Ahamed Yomal Bandara Chamod Gavindya Dharshana Kasthurirathna Lakmini Abeywardhana Department of Information Technology Sri Lanka Institute of Information Technology Malabe Sri Lanka Dept.of Computer Science & Software Engineering Sri Lanka Institute of Information Technology Malabe Sri Lanka

Legal professionals and law students often grapple with the arduous task of locating prior Supreme Court rulings for argument preparation and academic study, leading to a laborious and resource-intensive process. This study introduces an innovative approach to streamline the retrieval of legal information from Sri Lankan Supreme Court verdicts. The method centers on developing a Custom Named Entity Recognition (NER) model, boasting a remarkable accuracy exceeding 90%. This model efficiently extracts crucial legal entities from Supreme Court rulings, readily available on the Ministry of Justice's Supreme Court website. The high accuracy ensures precise entity extraction, followed by systematic organization within a dedicated database. Subsequently, a knowledge graph is formed by linking recorded legal entities, reducing information retrieval time to a mere 179 milliseconds, significantly outperforming existing methods. Moreover, a BART summarization model is crafted to generate concise, accurate, and insightful summaries of Supreme Court decisions, boasting an impressive ROGUE1 score of 85%. This approach revolutionizes legal information retrieval, delivering a user-friendly platform that enhances the identification of cases and fosters a deeper understanding, ultimately elevating the quality of legal research and practice.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Classification System Based on Data Mining Technique for Flat-Colored Textiles

SSRN

引用

SSRN 2025年

作者： Zhou, Min Lin, Yukai Hu, Wei Dai, Chai Tang, Guangfa He, Fei School of Information and Intelligent Engineering Guangzhou xinhua University Dongguan China School of Intelligent Control Hunan Railway Professional Technology College Zhuzhou China College of Electrical and Electrical Engineering Hunan University Changsha China Dept. of Pingdingshan Branch of China Mobile Communication Group Henan co. LTD Henan Pingdingshan China Dept.of Physics Electrical and Intelligent Manufacturing Huaihua University Changsha China College of Computer Science and Software Engineering Shenzhen University Shenzhen China

Manual visual inspection for defect detection in manufacturing can lead to high false positive rates. This paper compares data mining techniques for the binary classification of defects in flat products in a relatively open scenario. The background provides context on research into data mining-based detection. The methodology is divided into five sections: data collection, selection, preprocessing and transformation, classification, and model evaluation. Each section details the techniques applied, tailored to the application context, with specifying examples. Experiments apply real-world data sets self-collected from production lines. Statistics like prediction accuracy are generated by established data mining evaluation algorithms. Results indicate the efficacy of data mining for automated defect detection over manual approaches. Finally, conclusions summarize findings and discuss the limitations of the current research and planned future work to address the drawbacks. Overall, the formalized methodology and supportive experimental results demonstrate both the viability and potential areas of improvement for data mining in manufacturing quality control. © 2025, The Authors. All rights reserved.

关键词： Data accuracy

来源：评论

学校读者我要写书评

暂无评论

Graph exploration by energy-sharing mobile agents

arXiv

引用

arXiv 2021年

作者： Czyzowicz, J. Dobrev, S. Killick, Ryan Kranakis, E. Krizanc, D. Narayanan, L. Opatrny, J. Pankratov, D. Shende, S. Département d’Informatique Université du Québec en Outaouais Canada Slovak Academy of Sciences Bratislava Slovakia School of Computer Science Carleton University Ottawa Canada Dept.of Mathematics & Computer Science Wesleyan University MiddletownCT United States Department of Computer Science and Software Engineering Concordia University Canada Department of Computer Science Rutgers University United States

We consider the problem of collective exploration of a known n-node edge-weighted graph by k mobile agents that have limited energy but are capable of energy transfers. The agents are initially placed at an arbitrary subset of nodes in the graph, and each agent has an initial, possibly different, amount of energy. The goal of the exploration problem is for every edge in the graph to be traversed by at least one agent. The amount of energy used by an agent to travel distance x is proportional to x. In our model, the agents can share energy when co-located: when two agents meet, one can transfer part of its energy to the other. For an n-node path, we give an O(n+k) time algorithm that either finds an exploration strategy, or reports that one does not exist. For an n-node tree with ` leaves, we give an O(n+k2) algorithm that finds an exploration strategy if one exists. Finally, for the general graph case, we show that the problem of deciding if exploration is possible by energy-sharing agents is NP-hard, even for 3-regular graphs. In addition, we show that it is always possible to find an exploration strategy if the total energy of the agents is at least twice the total weight of the edges;moreover, this is asymptotically optimal. © 2021, CC BY.

关键词： Mobile agents

来源：评论

学校读者我要写书评

暂无评论

Faster RCNN Detection Based OpenCV CSRT Tracker Using Drone Data

Faster RCNN Detection Based OpenCV CSRT Tracker Using Drone ...

引用

2019 International Conference on Information Science and Communications Technologies, ICISCT 2019

作者： Farhodov, Xurshedjon Kwon, Oh-Heum Kang, Kyung Won Lee, Suk-Hwan Kwon, Ki-Ryong Pukyong National University Dept. of IT Convergence and Application Engineering Busan Korea Republic of Dongeui University Dept.of Computer Software Engineering Busan Korea Republic of Tongmyong University Dept. of Information Communication Engineering Busan Korea Republic of Tongmyong University Dept.of Information Security Busan Korea Republic of

ISBN: (纸本)9781728125640

Object tracking is one of the well-known topic in computer Vision sphere with challenging and problematic tasks that frames come with problems like overlapping, camera motion blur, changing object appearance, environmental variations and so on. Using OpenCV based CSRT (Channel and Spatial Reliability Tracking) tracker itself, without integration object detection part may face problems which mentioned above. To overcome that challenges we integrated deep learning based object detection with CSRT tracker and got better tracking results. © 2019 IEEE.

关键词： Aircraft detection

来源：评论

学校读者我要写书评

暂无评论

Integrated software Development Methodology for Real-time Precise Control of UAV

Integrated Software Development Methodology for Real-time Pr...

引用

作者： Jung Kyu Park Eun Young Park Heung Seok Jeon Dept.of Computer Software Engineering Changshin Univ. Dept.Biomedical Laboratory science Shinhan University Dept.of Computer Engineering Konkuk University

Recently, drones have been widely used in education and industry. In this paper, we propose a navigation and control algorithm to develop unmanned drones quickly. To limit the proposed framework, we analysed the existing drone solution and simulation framework. We also used Hardware-in-the-Loop(HIL) simulation for rapid development. HIL will simulate physical systems that interface with embedded control equipment on real-time hardware. The code generated in this process can be experimented on a hardware device so that the control performance of the drones can be verified without draining the drones in the outdoor environment. Using the proposed method, you save time and money on drone development while providing stability.

关键词： Integrated software Development Methodology for Real-time Precise Control of UAV UAV

来源：评论

学校读者我要写书评

暂无评论

Faster RCNN Detection Based OpenCV CSRT Tracker Using Drone Data

Faster RCNN Detection Based OpenCV CSRT Tracker Using Drone ...

引用

International Conference on Information Science and Communications Technologies (ICISCT)

作者： Xurshedjon Farhodov Oh-Heum Kwon Kyung Won Kang Suk-Hwan Lee Ki-Ryong Kwon Dept. of IT Convergence and Application Engineering Pukyong National University Busan South Korea Dept.of Computer Software Engineering Dongeui University Busan South Korea Dept. of Information & Communication Engineering Tongmyong University Busan South Korea Dept.of Information Security Tongmyong University Busan South Korea

ISBN: (数字)9781728125640

ISBN: (纸本)9781728125657

关键词： Target tracking Object detection Object tracking Training Machine learning Reliability

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：