检索结果-内蒙古大学图书馆

9th Annual ACM International conference on Web Search and Data Mining (WSDM)

作者： You, Quanzeng Luo, Jiebo Jin, Hailin Yang, Jianchao Univ Rochester Rochester NY 14623 USA Adobe Res San Jose CA 95110 USA Snapchat Inc Venice CA 90291 USA

ISBN: (纸本)9781450337168

Sentiment analysis of online user generated content is important for many social media analytic's tasks. Researchers have largely relied on textual sentiment analysis to develop systems to predict political elections, measure economic indicators, and so on. Recently, social media users are increasingly using additional images and videos to express their opinions and share their experiences. Sentiment analysis of such large-scale textual and visual content can help better extract user sentiments toward events or topics. Motivated by the needs to leverage large-scale social multimedia content for sentiment analysis, we propose a cross-modality consistent regression (CCR) model, which is able to utilize both the state-of-the-art visual and textual sentiment analysis techniques. We first, tine-tune a convolutional neural network (CNN) for image sentiment analysis and train a paragraph vector model for textual sentiment analysis. On top of them, we train our multi-modality regression model. We use sentimental queries to obtain half a million training samples from Getty Images. We have conducted extensive experiments on both machine weakly labeled and manually labeled image tweets. The results show that the proposed model can achieve better performance than the state of the art textual and visual sentiment analysis algorithms alone.

关键词： sentiment analysis cross-modality regression multimodality analysis

来源：评论

学校读者我要写书评

暂无评论

Harnessing Cloud and Edge Synergies: Toward an Information Theory of Fog Radio access Networks

引用

IEEE COMMUNICATIONS MAGAZINE 2016年第8期54卷 44-50页

作者： Tandon, Ravi Simeone, Osvaldo Univ Arizona Dept Elect & Comp Engn Tucson AZ 85721 USA NJIT CWiP Newark NJ USA

Edge caching and the centralization of base-band processing by means of the C-RAN architecture are among the most promising and transformative trends in the evolution of wireless networks. A key advantage of C-RAN is the possibility to perform cooperative transmission across multiple edge nodes, such as small cell base stations, thanks to centralized cloud processing. Cloud processing, however, comes at the cost of the potentially large delay entailed by fronthaul transmission between edge and cloud. In contrast, edge caching enables the low-latency transmission of popular multimedia content, but at the cost of constraining the operation of the edge nodes to decentralized transmission strategies with limited interference management capabilities. In order to accommodate the broad range of quality of service requirements of mobile broadband communication, in terms of spectral efficiency and latency, that are envisioned to be within the scope of 5G systems and beyond, this article considers a hybrid architecture, referred to as fog RAN (F-RAN), that harnesses the benefits of, and the synergies between, edge caching and C-RAN. In an F-RAN, edge nodes may be endowed with caching capabilities, while at the same time being controllable from a central cloud processor as in a C-RAN. In this article, an information-theoretic framework is presented that aims to characterize the main trade-offs between performance of an F-RAN, in terms of worst case delivery latency, and its resources: caching and fronthaul capacities.

关键词： Baseband Cache storage Cloud computing Computer architecture Encoding Process control Wireless communication

来源：评论

学校读者我要写书评

暂无评论

content-based Image Retrieval System for Marine Invertebrates 3

Content-based Image Retrieval System for Marine Invertebrate...

引用

3rd International conference on Information Retrieval and Knowledge Management (CAMP)

作者： Mustaffa, Mas Rina Norowi, Noris Mohd Yee, Sim May Univ Putra Malaysia Fac Comp Sci & Informat Technol Dept Multimedia Upm Serdang 43400 Selangor Malaysia

ISBN: (纸本)9781509029549

There are many marine life around the world where it is essential to have proper documentation for future records. Many information retrieval systems for marine science today require text input from user and can only be accessed online. Therefore, people who do not know the name of the marine species or do not have Internet access cannot search using the systems. Responding to this important need, this work aims to develop a content-based Image Retrieval (CBIR) system for marine invertebrates based on colour and shape features. With the CBIR system for marine invertebrates, users can use the system to look for marine invertebrates' species instead of using traditional methods of searching such as using books and encyclopedias. Users can easily upload the image of marine invertebrate that they want to search into the system and the system will retrieve all the other similar images of marine invertebrates along with their description. All the system interface's buttons, icons and text were designed in a way where any user can easily understand and further learn to operate the system themselves. Based on the retrieval effectiveness experiment and questionnaire-based survey, the proposed CBIR system for marine invertebrates is shown to be effective, help users search similar images of marine invertebrates, provide concise information on marine invertebrate's species for learning purposes, and is reliable and user-friendly.

关键词： content-based Image Retrieval marine invertebrates colour feature shape feature

来源：评论

学校读者我要写书评

暂无评论

Hybrid Swarm Intelligence Method for Post Clustering content Based Image Retrieval 7

Hybrid Swarm Intelligence Method for Post Clustering Content...

引用

7th International conference on Communication, Computing and Virtualization (ICCCV)

作者： Meshram, Shubhangi P. Thakare, Anuradha D. Gudadhe, Santwana Pimpri Chinchwad Coll Engn Dept Comp Engn Pune 411044 Maharashtra India

content Based Image Retrieval is one of the most promising method for image retrieval where searching and retrieving images from large scale image database is a critical task. In content Based Image Retrieval many visual feature like color, shape, and texture are extracted in order to match query image with stored database images. Matching the query image with each image of large scale database results in large number of disc scans which in turns slows down the systems performance. The proposed work suggested an approach for post clustering content Based Image Retrieval, in which the database images are clustered into optimized clusters for further retrieval process. Various clustering algorithms are implemented and results are compared. Among all, it is found that hybrid ACPSO algorithm performs better over basic algorithms like k-means, ACO, PSO etc. Hybrid ACPSO has the capability to produce good cluster initialization and form global clustering. This paper discusses work-in-progress where we have implemented till clustering module and intermediate results are produced. These resulted clusters will further be used for effective content Based Image Retrieval. (C) 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://***/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the Organizing Committee of ICCCV 2016

关键词： content Based Image Retrieval (CBIR) Image Clustering Ant Colony Optimization (ACO) Particle Swarm Optimization (PSO) Hybrid ACPSO

来源：评论

学校读者我要写书评

暂无评论

Proceedings of the 43rd International conference on Very Large Data Bases, VLDB February 2017

Proceedings of the 43rd International Conference on Very Lar...

引用

43rd International conference on Very Large Data Bases, VLDB 2017

This proceedings contains 7 papers. VLDB is a premier annual international forum for data management, database researches and application developments. This conference covers issues in data management, database and information systems research. It also aims at the newest improvements in Data, Cloud Computing, Big Data, Data Management, Information systems, Database and Data Mining. The themes and topics include: access Methods, Concurrency Control, Recovery, Transactions, Indexing and Search, In-memory Data Management, Hardware Accelerators, Query Processing and Optimization, Storage Management;Privacy and Security in Data Management;Graph Data Management, Social Networks, Recommendation systems;Data Mining and Analytics, Warehousing;Crowdsourcing, Embedded and Mobile Databases, Real-time Databases, Sensors and IoT, Stream Databases;Data Models and Query Languages, Schema Management and Design, Database Usability, User Interfaces and Visualization;Tuning, Benchmarking, Performance Measurement, Database Administration and Manageability;Distributed Database systems, Cloud Data Management, NoSQL, Scalable Analytics, Distributed Transactions, Consistency, P2P and Networked Data Management, Database-as-a-Service, content Delivery Networks;Provenance and Workflows, Spatial, Temporal, and multimedia Databases, Scientific and Medical Data Management, Profile-based or Context-Aware Data Management;Data Cleaning, Information Filtering and Dissemination, Information Integration, Metadata Management, Data Discovery, Web Data Management, Semantic Web, Heterogeneous and Federated Database systems;Fuzzy, Probabilistic and Approximate Databases, Information Retrieval, Text in Databases, etc. The key terms of this proceedings include sparse vector technique, OLAK, social networks, data tweening, query processing, NED, spatial graphs, NAM architecture, secure multiparty computation, SAC search algorithms.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of Video Recommender systems in Big Data Era 8

A Comparative Study of Video Recommender Systems in Big Data...

引用

8th International conference on Ubiquitous and Future Networks (ICUFN)

作者： Hong, Seong-Eun Kim, Hwa-Jong Kangwon Natl Univ Dept Comp & Commun Engn Chunchon South Korea

ISBN: (纸本)9781467399913

Recently, due to the wide spread of high-bandwidth access to the Internet, and abundant generation of various kinds video contents, we live in a big data era, especially in video contents. There are too much videos already, but we are even unable to know which video is good for me now. In the coming big data era, video contents providers should develop efficient recommendation system to be competitive and survive. In the paper, we compared video recommendation technologies of four famous companies: Netflix, Google (YouTube), Hulu, and Amazon in order to understand the basic differences between their recommendation algorithms and investigate the pros and cons. Most recommendation algorithms adopted collaborative filtering, but there are some differences. These days, as the data of user behavior can be gathered more easily, meta data play more important roles than recommendation algorithms.

关键词： Video content Service Recommendation System Collaborative Filtering

来源：评论

学校读者我要写书评

暂无评论

Seamless Transitions Between Filter Schemes for Location-based Mobile Applications 41

Seamless Transitions Between Filter Schemes for Location-bas...

引用

41st IEEE conference on Local Computer Networks (LCN)

作者： Richerzhagen, Bjoern Richerzhagen, Nils Zobel, Julian Schoenherr, Sophie Koldehofe, Boris Steinmetz, Ralf Tech Univ Darmstadt KOM Multimedia Commun Lab Darmstadt Germany

ISBN: (纸本)9781509020546

With a plethora of sensors and ubiquitous access to the Internet, modern smartphones have enabled a broad range of context-based applications. Most applications make use of the user's physical location to filter relevant content. However, filtering based on dynamic contextual information results in high complexity of the filtering process. This limits the applicability of existing publish/subscribe systems, as they rely on aggregation of filters and fast decentralized matching and forwarding. In this work, we propose a mechanism for transitions between different filter schemes for location-based services. Our mechanism adapts the filtering process to the dynamics in user behavior and resulting load by trading computational complexity at the broker against communication overhead and computational complexity at the mobile client. We integrate our mechanism into an existing publish/subscribe system and evaluate transitions between a context-based filter scheme and two channel-based filter schemes, showing the applicability of our approach.

关键词： Location-based applications mobile applications publish/subscribe

来源：评论

学校读者我要写书评

暂无评论

"Stubborn" Strategy to Mitigate Remaining Cache Misses 34

"Stubborn" Strategy to Mitigate Remaining Cache Misses

引用

34th IEEE International conference on Computer Design (ICCD)

作者： Nomura, Hayato Katchi, Hiroyuki Irie, Hidetsugu Sakai, Shuichi Univ Tokyo Grad Sch Informat Sci & Technol Tokyo Japan

ISBN: (纸本)9781509051427

The capacity of cache memory has reached the megabytes magnitude and various smart cache algorithms have been introduced. However, cache systems still suffer from plenty of cache misses, especially for memory-intensive applications. In this paper, we reveal that a vast majority of the remaining cache misses on LLC are caused by long interval, unpredictable re-reference accesses. Based on the analysis, we propose a "Stubborn" strategy to mitigate misses caused by long re-reference interval accesses. Our strategy simply enhances existing smart algorithms by mixing the cache lines that do not evict their holding content for longer than a 10 to 100 mega-instruction interval. We evaluate the strategy that is mixed with 2MB LLC based on LRU or DRRIP. The results show that our Stubborn strategy achieves a reduction in long interval re-reference misses, as expected. It outperforms LRU and DRRIP on the IPC metric showing increases of 24% and 6%, respectively.

关键词： Cache Algorithm Replacement Policy Re-reference Interval Cache access Patterns

来源：评论

学校读者我要写书评

暂无评论

A data access prediction unit for multimedia applications

A data access prediction unit for multimedia applications

引用

International conference on Microelectronics, ICM

作者： Tareq Alawneh Ahmed Elhossini Technical University of Berlin Embedded Systems Architectures (AE) Berlin Germany Department of Computers and Systems Al Azhar University Cairo Egypt

A large number of algorithms, which work on multimedia data, such as images and videos, perform data processing over rectangular regions of pixels. If this distinctive data access and other data accesses are exploited properly, it can yield significant application performance improvements. For the purpose of achieving efficient exploitation for these data accesses, their information should be monitored and detected at runtime. For this purpose, we propose a data access prediction unit for multimedia applications. Our simulation results reveal that the accuracy of predicting the data access information achieved by the proposed unit is, on average, 87.4%, for the evaluated workloads, when the proposed unit utilizes a history table with 64 entries.

关键词： History Two dimensional displays multimedia communication Prefetching Streaming media Mathematical model Prediction algorithms

来源：评论

学校读者我要写书评

暂无评论

Improving the RDP based applications by using HTML5 content representation

Improving the RDP based applications by using HTML5 content ...

引用

Mobile Devices and multimedia: Enabling Technologies, algorithms, and Applications 2016

作者： Ganji, Rama Rao Mitrea, Mihai Panovski, Dancho Joveski, Bojan Institut Mines-Telecom Telecom-SudParis UMR 8145 - MAP5 France Ustartapp France

Despite the large variety of "off the shelf" solutions and academic research studies, application virtualization for cloud distribution is still an open research topic, with significant issues to be solved, ranging from bridging the gap between users expectation of simple, intuitive and user-friendly access from any type of terminal and the fragmented landscape of SaaS offer, with peculiarities related to the hardware/software configurations and the optimization of the technical resources consumption. Our study investigates the possibility of using HTML5 a virtualization tool for RDP-based applications. Architectural modules related to the RDP content interception, conversion, adaptation, remote rendering and interaction are specified, designed and implemented. This architecture is validated under the framework of the MEDUSA European project, in partnership with medical institutions. The testbed considers a server and 5 mobile users, with heterogeneous devices (tablets, smartphones, laptops) running under iOS, Android and Windows operating systems. The objective/subjective evaluations demonstrated that: (1) the user experience is not reduced by the virtualization, (2) the network consumption is reduced by a factor of 1.8 with respect to state-ofthe- Art solutions. © 2016 Society for Imaging Science and Technology.

关键词： Windows operating system

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：