检索结果-内蒙古大学图书馆

2nd International Conference on Computational Intelligence, Communication Technology and Networking, CICTN 2025

作者： Garg, Sourav Mittal, Aashu Sathiyasuntharam, V. Department of Computer Science and Engineering Sharda University Greater Noida India

ISBN: (纸本)9798331530389

The level of crime rates in practically every part of the world has risen to unprecedented levels. These are tough times that call for extreme methods to protect individuals, particularly those who must travel to unknown and known locations on a daily basis. Regardless of the mode of transportation used by the victim walking, driving one's own car, using a public transportation vehicle, autorickshaws, or taxis the majority of these crimes happen when the victim is traveling. This research suggests a Deep Learning Based Model To Recommend Safe Route Navigation System that shows the user a safe path shown on maps and based on the area's historical criminal history. Our approach is implemented in a hierarchical manner: a Decision Network is used to determine the user-specific attributes in the first place, and then that same Decision Network is used to activate the creation of safe routes via Geospatial Data Analysis. Users can view the navigated routes of the effective system as a colour coded map. Acknowledging that the real consequences which our research strive for should produce, we validated our model with specific data of San Francisco city for demonstrative purposes. © 2025 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

SDN-Driven Delay Minimization for Task Offloading in VECN Using Particle Swarm Optimization 6

SDN-Driven Delay Minimization for Task Offloading in VECN Us...

引用

6th International Conference on Sustainable Technologies for Industry 5.0, STI 2024

作者： Musfiq, Md. Kowshik, Aktaruzzaman Utsha, Md. Mahdizzaman Ahmed, Jargis Tareq, Md. Ahsan Habib Green University of Bangladesh Department of Computer Science and Engineering Bangladesh

ISBN: (纸本)9798331531973

In Vehicular Edge Computing (VEC), efficient task offloading is essential for reducing latency and enhancing system performance, especially in resource-constrained environments. In this regard, Roadside Units (RSUs) can provide computation resources to vehicle users to execute offloaded tasks. However, RSUs are resource-constrained and could fail if many vehicular users offload tasks to certain RSUs. Hence balancing the load and distributing users among RSUs is required. In this paper, we considered a Mobile Edge Computing server (MEC) acting as an SDN that can observe the situation of the subordinate RSUs and map RSUs and users for computation offloading considering the delay constraint of the users. Due to the NP-hard nature of the formulated problem, we presented a meta-heuristic algorithm called Delay-aware Particle Swarm Optimization for task offloading (DAPSO). Where MEC as SDN will solve the task placement problem for vehicular users and direct users to offload tasks to the best possible RSU or the MEC server. Our experimental results demonstrate that the proposed DAPSO approach significantly reduces overall system delay and improves overall system efficiency compared to existing literature works. © 2024 IEEE.

关键词： Heuristic algorithms

来源：评论

学校读者我要写书评

暂无评论

A Multimodal Named Entity Recognition Approach Based on Multi-Perspective Contrastive Learning 7

A Multimodal Named Entity Recognition Approach Based on Mult...

引用

7th International Conference on Machine Learning and Natural Language Processing, MLNLP 2024

作者： Liu, Huafu Wang, Yongli Liu, Dongmei School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China

ISBN: (纸本)9798350354973

Multimodal Named Entity Recognition (MNER) is a vision-language task aimed at detecting the scope of entities and determining their types in given text-image pairs. Although existing MNER models have made some progress, they often encounter issues like mismatched feature dimensions and inconsistent feature representations when interacting with text sequences. This makes it difficult to bridge the semantic gap between images and text. Therefore, this paper proposes a multimodal named entity recognition method based on multiperspective contrastive learning (MPCL-NER). Initially, the ALBERT pretrained language model and the ViT visual model are used to extract features from text and images, respectively. Based on this, image feature representations are enhanced through multi-perspective contrastive learning, which also promotes the integration of image and text information. Finally, the two types of features are merged through a multi-head adaptive attention mechanism and a gated fusion mechanism. This is followed by entity tagging with a CRF (Conditional Random Field) decoder. Experimental results show that this method outperforms other baseline methods on the public dataset Twitter-2017, achieving an F1 score of 86.11%, demonstrating the effectiveness of this approach. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

A Color Histogram-Based Approach for Scene Segmentation in a Video 1

A Color Histogram-Based Approach for Scene Segmentation in a...

引用

1st IEEE International Conference on Smart Power Control and Renewable Energy, ICSPCRE 2024

作者： Majumdar, Soumya Rao, Krothapalli Sreenivasa Iit Kharagpur Department of Computer Science & Engineering Kharagpur India

ISBN: (纸本)9798350377002

A scene in a video consists of several shots and difference between shot and scene is that shot is captured from different camera locations whereas scene consists of shots that are captured at same location but from different camera angles. Simi-larity measurements between shots for detecting scene boundary are done in existing work using a sliding window of variable length. This method can lead to poor result if the sliding window size is too large or too small because success of this method depends upon video content. Our proposed method for video scene boundary detection and segmentation does clustering of keyframes instead of applying sliding window. Proposed method has four stages: shot segmentation with the help of tool, selection of keyframes with highest entropy from each shot, clustering of consecutive frames with most similarity, generation of scenes from clustered keyframes. Dataset used for experiment is four episodes of the first season of an American sitcom named 'The Bing Bang Theory' and the result is compared with existing state-of-the-art literature. Proposed method is time-consuming because of lesser number of computations. This method also has less time complexity because entropy differences of keyframes are computed instead of all frames. However, the presence of ads in the datasets used in the experiment affects the result in some cases. © 2024 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

An Online Spike Pattern Learning Algorithm Based on Local Variable Driven STDP 15

An Online Spike Pattern Learning Algorithm Based on Local Va...

引用

15th IEEE Annual Computing and Communication Workshop and Conference, CCWC 2025

作者： Liu, Jiahua Lin, Xianghong Zhang, Zhengyi Li, Chun College of Computer Science and Engineering Northwest Normal University Lanzhou China

ISBN: (纸本)9798331507695

Online learning algorithms adjust the synaptic weights when the neurons fire spikes during the learning process of spiking neural networks, which are shown to be more suitable and effective for processing spatio-temporal information in the real world. This paper presents an online spike pattern supervised learning algorithm for spiking neural networks based on local variable driven STDP (LVDSTDP), which combines low-pass filtering of spike trains fired by pre- and postsynaptic neurons to adjust synaptic weights in real time. The LVDSTDP algorithm has been successfully applied to spike trains learning tasks, and the effects of synaptic parameters are analyzed in detail. Experimental results show that the proposed algorithm can effectively adjust the synaptic weights to improve learning performance. This demonstrates effective for solving complex pattern learning problems using the LVDSTDP rule, and many further optimizations are possible. We also applied the proposed method in training SNNs, and obtained comparably high recognition accuracies on the MNIST datasets. © 2025 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

An effective abstract text summarisation using shark smell optimised bidirectional encoder representations from transformer

引用

International Journal of Business Intelligence and Data Mining 2023年第1期23卷 50-72页

作者： Nafees Muneera, M. Sriramya, P. Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Saveetha University Tamilnadu Chennai602105 India Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Chennai India

Recently, a vast amount of text data has increased rapidly and therefore information must be summarised to retrieve useful knowledge. First, the preprocessing module utilises the fixed-length stemming method, and then the segmentation module makes use of a pre-trained bidirectional encoder representations from transformers (BERT). The text of input is segmented with the utilisation of feedforward and multi-head attention layer. This BERT segmentation paradigm is adjoined alongside shark smell optimisation (SSO) methodology, and thus, the phrases that are extricated are employed to prepare the document stage of a dataset of Amazon merchandise assessment. This study aspires to create a concise summary and invigorating headlines, which grab the focus of the readers. This paper demonstrates that it performs by amalgamating the duo extractive and abstractive procedures employing a pipelined technique for creating a succinct summary that is later utilised for headline creation. Experimentation was executed on publicly accessible datasets – CNN/Daily Mail. Copyright © 2023 Inderscience Enterprises Ltd.

关键词： Signal encoding

来源：评论

学校读者我要写书评

暂无评论

Network Summarization Through the Lens of Latent Semantic Analysis and Fuzzy Techniques

Network Summarization Through the Lens of Latent Semantic An...

引用

International Conference on Internet of Everything and Quantum Information Processing, IEQIP 2023

作者： Vanathi, M. Lakshmanan, L. Department of Computer Science and Engineering Sathyabama Institute of Science and Technology Chennai India

ISBN: (纸本)9783031619281

Extracting text summarization involves the retention of only the most important sentences in a document. Finding the appropriate ranking order of the phrases in the document according to their relevance is a vital step in extractive text summarization. Singular value decomposition or SVD algorithm based on latent semantic analysis focuses on recognizing the sections in the document which are related in terms of their semantic nature. Fuzzy algorithms involve reasoning of the priority order of the sentences using fuzzy logic unlike the use of discrete values. While significant work has been done for extractive text summarization in English and other foreign languages, there is ample scope for improving the performance of systems when dealing with Marathi text. An analysis of the characteristics of both these techniques is conducted to compare their benefits and shortcomings. The performance of both the algorithms is evaluated on a document dataset using standard performance metrics including the ROUGE metric. Both of these approaches are objectively compared in order to determine their applicability, particularly when dealing with non-English text in general or Marathi in particular. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Multi-view session-based social recommendation with self-supervised learning 10

Multi-view session-based social recommendation with self-sup...

引用

10th IEEE Smart World Congress, SWC 2024

作者： Li, Jiamian Liu, Xiao Guo, Yixing Shanghai University School of Computer Engineering and Science Shanghai China

ISBN: (纸本)9798331520861

Session-based social recommendation (SSR) is an important branch of session-based recommendation that aims to generate more accurate recommendations by combining users' historical sessions with their social information. Existing SSR models are usually based on graph neural networks. Although they have shown better performance compared to traditional SSR models, they ignored users' temporal behavior patterns and data sparsity issues. Therefore, we propose a multi-view SSR model MVSSR that effectively constructs multiple views to model the temporal and global user preferences in sparse SSR scenarios. On one hand, we construct user data into multiple views to thoroughly capture users' temporal preferences and global preferences. On the other hand, we constructed a view-aware self-supervised learning method. Through self-supervised learning within and between views, we enhance the user's session representation to alleviate the data sparsity is-sues in SSR. We performed ex-tensive experiments on three real-world datasets to confirm the effectiveness of MVSSR. The results indicate that our model surpasses the current state-of-the-art models in terms of hit rate and mean reciprocal rank. © 2024 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Dynamic Parameterized Problems on Unit Disk Graphs 35

Dynamic Parameterized Problems on Unit Disk Graphs

引用

35th International Symposium on Algorithms and Computation, ISAAC 2024

作者： An, Shinwoo Cho, Kyungjin Jang, Leo Jung, Byeonghyeon Lee, Yudam Oh, Eunjin Shin, Donghun Shin, Hyeonjun Song, Chanho Department of Computer Science and Engineering POSTECH Pohang Korea Republic of

ISBN: (纸本)9783959773546

In this paper, we study fundamental parameterized problems such as k-Path/Cycle, Vertex Cover, Triangle Hitting Set, Feedback Vertex Set, and Cycle Packing for dynamic unit disk graphs. Given a vertex set V changing dynamically under vertex insertions and deletions, our goal is to maintain data structures so that the aforementioned parameterized problems on the unit disk graph induced by V can be solved efficiently. Although dynamic parameterized problems on general graphs have been studied extensively, no previous work focuses on unit disk graphs. In this paper, we present the first data structures for fundamental parameterized problems on dynamic unit disk graphs. More specifically, our data structure supports 2O(√k) update time and O(k) query time for k-Path/Cycle. For the other problems, our data structures support O(log n) update time and 2O(√k) query time, where k denotes the output size. © Shinwoo An, Kyungjin Cho, Leo Jang, Byeonghyeon Jung, Yudam Lee, Eunjin Oh, Donghun Shin, Hyeonjun Shin, and Chanho Song.

关键词： Parameterization

来源：评论

学校读者我要写书评

暂无评论

An Evaluation of Quality Stability in NeRF Object Modeling

An Evaluation of Quality Stability in NeRF Object Modeling

引用

2024 International Conference on Artificial Intelligence and Mechatronics System, AIMS 2024

作者： Luo, Shihao Flaherty, Jean Yang, Haoyi Onuoha, Chibuike Thang, Truong Cong The University of Aizu Department of Computer Science and Engineering Aizuwakamatsu Japan

ISBN: (纸本)9798350350524

Neural Radiance Field (NeRF) has become an increasingly popular approach for modeling the geometry and appearance of scenes/objects. Various datasets for training and testing the rendered view quality have been introduced. However, there is still a lack in evaluating quality variations for NeRF models. This is especially crucial in interactive multimedia services like e-learning and Virtual Reality, where users may change the viewing direction frequently. To address this limitation, we carry out quality stability assessments with different NeRF models and 3D objects. Rather than using commonly used spiral sampling or random sampling on the target area, we employ parallel contour lines as the trajectory of cameras. The empirical results show that quality variations strongly depend on models, objects, and the local location (e.g. angle and height) of rendered images. Also, it is important to report statistic measures of quality variations in NeRF modeling. © 2024 IEEE.

关键词： Virtual reality

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：