检索结果-内蒙古大学图书馆

SHIKSHA: Smart Hybrid Intelligent Knowledge System for Helping Academia 2nd

学校读者我要写书评

暂无评论

SHIKSHA: Smart Hybrid Intelligent Knowledge System for Help...

2nd International Conference on Applied Algorithms, ICAA 2025

作者： Paul, Aishik Shaha, Abhisri Mukherjee, Arijit Guha, Debajit Das, Priyam Das, Ankan Tarafdar, Anurina Sen, Anindya Department of Electronics and Communication Engineering Heritage Institute of Technology West Bengal Kolkata700107 India Department of Computer Science and Engineering Heritage Institute of Technology West Bengal Kolkata700107 India Heritage Institute of Technology West Bengal Kolkata700107 India

ISBN: (纸本)9783031845420

Our Smart Hybrid Intelligent Knowledge System for Helping Academia (SHIKSHA) addresses critical challenges in classroom and examination management by offering a fully automated solution. Key features consist automated attendance tracking with integrated intruder detection which eliminates manual errors and enhances security. The system also improves examination management using the Coordinate and Area-based Row Column Mapping (CAB-RCM) algorithm ensuring accurate seat-matrix mapping via face recognition. SHIKSHA automates student performance monitoring by analyzing behavior with face tracking and seating patterns to help educators make data-driven decisions. It supports resource management by sending maintenance alerts based on projector usage with a brightness-based algorithm. It monitors classroom audio to assess attentiveness and track syllabus progress. Utilizing existing CCTV cameras and microphones, SHIKSHA operates with minimal hardware requirements and offers an eco-friendly, cost-effective and scalable solution that enhances academic efficiency across a wide range of educational institutions. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Knowledge engineering

Deep Sentiment: “An Effective Sentiment Analysis of Twitter Data using BERT+GRU Hybrid Model”

学校读者我要写书评

暂无评论

Deep Sentiment: “An Effective Sentiment Analysis of Twitter...

International Conference on ent Computing and Control Systems (ICICCS)

作者： K. Sudhakar T. Jatin Pratap Singh M. Thanuj B. Praveen Department of Computer Science and Engineering (AI & ML) Vignan Institute of Technology and Science Nalgonda Telangana

ISBN: (数字)9798331512088

ISBN: (纸本)9798331512095

Sentiment Analysis (SA) has been an active area of research for over a decade. It involves an in-depth study of opinions, sentiments, and subjectivity expressed in text. Among various platforms, Twitter is one of the most popular microblogging services and serves as a significant source for sentiment analysis and data exploration. Organizations globally utilize SA on Twitter data to gain insights with applications across various domains. However, while SA is effective for understanding public opinion, bias in tweets can distort analysis, leading to inaccurate results and potentially influencing users to make flawed decisions. Such biased tweets may originate from both authentic but biased individuals and automated social bots that disseminate skewed opinions on specific topics. To address this challenge, this research proposes a statistical model designed to detect users and bots responsible for spreading biased content on Twitter. Using an annotated Twitter dataset, the study evaluates sentiment analysis results with and without biased tweets, analysing the impact of biased users at both micro and macro levels. Experimental results demonstrate that the proposed approach effectively distinguishes biased users and bots from genuine users, thereby enhancing the accuracy of sentiment analysis.

关键词： Sentiment analysis Analytical models Accuracy Social networking (online) Computational modeling Blogs Organizations Control systems Data models Data mining

Internet of Things Enabled Smart Hand Gesture Virtual Mouse System 14

学校读者我要写书评

暂无评论

Internet of Things Enabled Smart Hand Gesture Virtual Mouse ...

14th IEEE International Conference on Communication Systems and Network Technologies, CSNT 2025

作者： Chelekkodan, Danish Muneer Jayapandian, N. Abhinav, S. Shyjan, Aaron Department of Computer Science and Engineering Christ University Kengeri Campus Bangalore India

ISBN: (纸本)9798331531935

This research is aim to focus on IoT based hand gesture model. Mouse is one of the most important input devices of a computer. It works as a pointing device and allows the user to move the pointer as needed by the user. In the early days, a wired mechanical mouse was used for this purpose. In mechanical mouse a ball is fixed underneath the mouse, which rotates as the user moves the mouse. This movement of the ball is used to move the mouse pointer on the screen. Now mostly we use optical mouse which can be wired or wireless. An optical mouse has a high-power laser below it, which takes more than thousand pictures of the surface below the mouse. An image comparator compares the images and sends the signal to move the mouse pointer as the texture of the image changes. Both the types of mouse works based on old technology. As technology leaps to greater heights, the need for simplicity also increases. With the invention of different kind of sensors, microcontrollers and other electronics, we can eliminate the mouse as an input device and instead use our hands to do the work of a mouse. This prototype is an embedded system which runs with the help of an arduino microcontroller. Flex sensors are used to capture the hand gestures. The proposed IoT based hand gesture model is providing high accuracy rate compare to the regular model. The proposed model is analyzed with accuracy level, the average accuracy level of proposed model is more than 90%. © 2025 IEEE.

关键词： Comparators (optical)

Cracks in The Stack: Hidden Vulnerabilities and Licensing Risks in LLM Pre-Training Datasets

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Jahanshahi, Mahmoud Mockus, Audris Department of Electrical Engineering and Computer Science University of Tennessee Knoxville United States

A critical part of creating code suggestion systems is the pre-training of Large Language Models (LLMs) on vast amounts of source code and natural language text, often of questionable origin, quality, or compliance. This may contribute to the presence of bugs and vulnerabilities in code generated by LLMs. While efforts to identify bugs at or after code generation exist, it is preferable to pre-train or fine-tune LLMs on curated, high-quality, and compliant datasets. The need for vast amounts of training data necessitates that such curation be automated, minimizing human intervention. We propose an automated source code autocuration technique that leverages the complete version history of open-source software (OSS) projects to improve the quality of training data. The proposed approach leverages the version history of all OSS projects to: (1) identify training data samples that have ever been modified, (2) detect samples that have undergone changes in at least one OSS project, and (3) pinpoint a subset of samples that include fixes for bugs or vulnerabilities. We evaluate this method using "the Stack" v2 dataset, comprising almost 600M code samples, and find that 17% of the code versions in the dataset have newer versions, with 17% of those representing bug fixes, including 2.36% addressing known CVEs. The clean, deduplicated version of Stack v2 still includes blobs vulnerable to 6,947 known CVEs. Furthermore, 58% of the blobs in the dataset were never modified after creation, suggesting they likely represent software with minimal or no use. Misidentified blob origins present an additional challenge, as they lead to the inclusion of non-permissively licensed code, raising serious compliance concerns. By deploying these fixes and addressing compliance issues, the training of new models can avoid perpetuating buggy code patterns or license violations. We expect our results to inspire process improvements for automated data curation, a critical component of AI engineeri

关键词： Open source software

A Selective Homomorphic Encryption Approach for Faster Privacy-Preserving Federated Learning

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Korkmaz, Abdulkadir Rao, Praveen Department of Electrical Engineering & Computer Science University of Missouri Columbia United States

Federated learning is a machine learning method that supports training models on decentralized devices or servers, where each holds its local data, removing the need for data exchange. This approach is especially useful in healthcare, as it enables training on sensitive data without needing to share them. The nature of federated learning necessitates robust security precautions due to data leakage concerns during communication. To address this issue, we propose a new approach that employs selective encryption, homomorphic encryption, differential privacy, and bit-wise scrambling to minimize data leakage while achieving good execution performance. Our technique FAS (fast and secure federated learning) is used to train deep learning models on medical imaging data. We implemented our technique using the Flower framework and compared with a state-of-the-art federated learning approach that also uses selective homomorphic encryption. Our experiments were run in a cluster of eleven physical machines to create a real-world federated learning scenario on different datasets. We observed that our approach is up to 90% faster than applying fully homomorphic encryption on the model weights. In addition, we can avoid the pretraining step that is required by our competitor and can save up to 46% in terms of total execution time. While our approach was faster, it obtained similar security results as the competitor. Copyright © 2025, The Authors. All rights reserved.

关键词： Federated learning

MaLAware: Automating the Comprehension of Malicious Software Behaviours using Large Language Models (LLMs)

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Saha, Bikash Rani, Nanda Shukla, Sandeep Kumar Department of Computer Science & Engineering Indian Institute of Technology Kanpur Kanpur India

Current malware (malicious software) analysis tools focus on detection and family classification but fail to provide clear and actionable narrative insights into the malignant activity of the malware. Therefore, there is a need for a tool that translates raw malware data into human-readable descriptions. Developing such a tool accelerates incident response, reduces malware analysts’ cognitive load, and enables individuals having limited technical expertise to understand malicious software behaviour. With this objective, we present MaLAware, which automatically summarizes the full spectrum of malicious activity of malware executables. MaLAware processes Cuckoo Sandbox-generated reports using large language models (LLMs) to correlate malignant activities and generate concise summaries explaining malware behaviour. We evaluate the tool’s performance on five open-source LLMs. The evaluation uses the human-written malware behaviour description dataset as ground truth. The model’s performance is measured using 11 extensive performance metrics, which boosts the confidence of MaLAware’s effectiveness. The current version of the tool, i.e., MaLAware, supports Qwen2.5-7B, Llama2-7B, Llama3.1-8B, Mistral-7B, and Falcon-7B, along with the quantization feature for resource-constrained environments. MaLAware lays a foundation for future research in malware behavior explanation, and its extensive evaluation demonstrates LLMs’ ability to narrate malware behavior in an actionable and comprehensive manner. Copyright © 2025, The Authors. All rights reserved.

关键词： Cybersecurity

Real-Time Insights and Dynamic Performance Visualizations for Unprecedented Impact in Ball Badminton 3rd

学校读者我要写书评

暂无评论

Real-Time Insights and Dynamic Performance Visualizations fo...

3rd International Conference on Sports engineering, ICSE 2023

作者： Rohith Gowda, M. Pai, Aditi N. Vishwanatha, C.N. Akash, M.P. Ranganath, G.H. Bhaskar, H.N. Department of Computer Science and Engineering R. V. College of Engineering Bengaluru India Department of Physical Education and Sports R. V. College of Engineering Bengaluru India Department of Physical Education and Sports SJB Institute of Technology Bengaluru India Department of Physical Education and Sports SBRRMFGC Mysuru India

ISBN: (纸本)9789819782963

The absence of advanced analytical tools has been a major obstacle in the world of ball badminton, a traditional sport. The lack of real-time data and valuable insights has long impeded players, coaches, and spectators. This paper introduces an innovative system aimed at transforming the ball badminton experience, enhancing player performance, engaging spectators, and providing valuable tools for coaches and fans through real-time insights and dynamic performance visualizations. This approach involves developing a system, preferably an application, to track the movement of the player and measure position, acceleration, and speed using computer vision and predicting tactics to assist during training, tracking the short trajectory of the ball to assist during practices for improvement, displaying live statistics such as player stats (points won, errors, aces), match progress (scoreboard), and historical data (head-to-head records) to provide context and analysis during the match, besides assisting the umpire in making decisions. Benefits to the viewers are the display of real-time statistics, fan engagement through quizzes on the sport, and result prediction. Coaches can use it to make strategic decisions, players to review their performance, and fans to gain deeper insights into the match. Additionally, it employs a markerless motion capture system, which relies on deep learning for human position estimation and computer vision techniques. To transform pixel data into court coordinates and enable the measurement of parameters like distance covered, court positioning, and average player speeds for ball badminton players, the methodology relies on the inverse perspective mapping method. One of the algorithms applied in this context is the Lucas-Kanade optical flow algorithm, which aids in player tracking and position estimation, contributing to the overall effectiveness of the technology. The technology will track player movements and tactical insights with great preci

关键词： Deep learning

Evaluating the Impacts of Swapping on the US Decennial Census

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Ballesteros, María Dwork, Cynthia King, Gary Olson, Conlan Raghavan, Manish Institute for Quantitative Social Science Harvard University United States Department of Computer Science Harvard University United States Department of Computer Science Columbia University United States Sloan School of Management Department of Electrical Engineering and Computer Science MIT United States

To meet its dual burdens of providing useful statistics and ensuring privacy of individual respondents, the US Census Bureau has for decades introduced some form of "noise" into published statistics. Initially, they used a method known as "swapping" (1990–2010). In 2020, they switched to an algorithm called TopDown that ensures a form of Differential Privacy. While the TopDown algorithm has been made public, no implementation of swapping has been released and many details of the deployed swapping methodology deployed have been kept secret. Further, the Bureau has not published (even a synthetic) "original" dataset and its swapped version. It is therefore difficult to evaluate the effects of swapping, and to compare these effects to those of other privacy technologies. To address these difficulties we describe and implement a parameterized swapping algorithm based on Census publications, court documents, and informal interviews with Census employees. With this implementation, we characterize the impacts of swapping on a range of statistical quantities of interest. We provide intuition for the types of shifts induced by swapping and compare against those introduced by TopDown. We find that even when swapping and TopDown introduce errors of similar magnitude, the direction in which statistics are biased need not be the same across the two techniques. More broadly, our implementation provides researchers with the tools to analyze and potentially correct for the impacts of disclosure avoidance systems on the quantities they study. Copyright © 2025, The Authors. All rights reserved.

关键词： Differential privacy

AM-DGCNN: Leveraging Graph Attention Networks and Edge Attributes for Link Classification in Knowledge Graphs 24

学校读者我要写书评

暂无评论

AM-DGCNN: Leveraging Graph Attention Networks and Edge Attri...

Proceedings of the SC '24 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis

作者： Dhroov Pandey Tong Shu Department of Computer Science and Engineering University of North Texas Denton TX USA

ISBN: (纸本)9798350355543

Graph-based representations are increasingly popular for storing and managing information through knowledge graphs, which capture entities and their relationships. However, these knowledge graphs often suffer from incomplete link information. To address this issue, link classification methods can be used to predict and verify missing connections. Traditional methods use heuristics to guess the associativity patterns of the graph, but this results in a loss of generalizability. Recently, supervised heuristic learning methods have improved the link classification accuracy by learning the heuristic that fits best for a particular graph. Specifically, the SEAL framework [37], as a state-of-the-art supervised heuristic learning tool, excels in learning associativity patterns by analyzing local enclosing subgraphs to classify links. However, DGCNN, a graph neural network (GNN) model in this framework, lacks the capability to process edge attributes, leading to poor classification accuracy in knowledge graphs with rich link information. Hence, this paper proposes an Augmented Model of the DGCNN (AM-DGCNN) by replacing Graph Convolution Networks with Graph Attention Networks to better incorporate link information. With extensive experiments, we demonstrate that our AM-DGCNN in the SEAL framework can achieve up to 99% AUC and 97% precision for classifying links in knowledge graphs.

关键词： Graph attention networks