检索结果-内蒙古大学图书馆

arXiv 2018年

作者： He, Jing Li, Xin Liao, Lejian Cheung, William K. Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Application School of Computer Science and Technology Beijing Institute of Technology Beijing10081 China

In this paper, we address the problem of personalized next Point-of-interest (POI) recommendation which has become an important and very challenging task for location-based social networks (LBSNs), but not well studied yet. With the conjecture that, under different contextual scenarios, human exhibits distinct mobility pattern, we attempt here to jointly model the next POI recommendation under the influence of user's latent behavior pattern. We propose to adopt a third-rank tensor to model the successive check-in behaviors. By integrating categorical influence into mobility patterns and aggregating user's spatial preference on a POI, the proposed model deal with the next new POI recommendation problem by nature. By incorporating softmax function to fuse the personalized Markov chain with latent pattern, we furnish a Bayesian Personalized Ranking (BPR) approach and derive the optimization criterion accordingly. Expectation Maximization (EM) is then used to estimate the model parameters. We further develop a personalized model by taking into account personalized mobility patterns under the contextual scenario to improve the recommendation performance. Extensive experiments on two large-scale LBSNs datasets demonstrate the significant improvements of our model over several state-of-the-art methods. Copyright © 2018, The Authors. All rights reserved.

关键词： Maximum principle

来源：评论

学校读者我要写书评

暂无评论

Enhancing ability of fault detection for component systems based on object interactions graph

Enhancing ability of fault detection for component systems b...

引用

2012 International Conference on Emerging Technologies for information Systems, computing, and Management, ICM 2012

作者： Sun, Fuzhen Liao, Lejian Du, Jianguang Li, Guoqiang Beijing Eng. Res. Ctr. of High Vol. Language Information Processing and Cloud Computing Applications Beijing Key Laboratory of Intelligent Information Technology School of Computer Science Beijing China School of Computer Science and Technology Shandong University of Technology Zibo China

ISBN: (纸本)9781461470090

Test case prioritization is a technique to schedule the test case in order to maximize some objective function. Early fault detection can provide a faster feedback generating a scope for debuggers to carry out their task at an early stage. In this paper, a method is proposed to prioritize the test cases for testing component dependency in a Component Based Software Development (CBSD) environment using Greedy Approach. The OIG (Object Interaction Graph) is traversed to calculate the total number of inter component object interactions and intra component object interactions. Depending upon the number of interactions, the objective function is calculated and the test cases are ordered accordingly. This technique is applied to the components developed in Java for a software system and found to be very effective in early fault detection as compared with non-prioritize approach. © 2013 Springer Science+Business Media New York.

关键词： Fault detection

来源：评论

学校读者我要写书评

暂无评论

A NEW ALGORITHM OF EXTRACTING MAXIMAL-LENGTH EVALUATION PHRASE AND COLLOCATION OF EVALUATION

A NEW ALGORITHM OF EXTRACTING MAXIMAL-LENGTH EVALUATION PHRA...

引用

2012 IEEE 2nd International Conference on cloud computing and Intelligence Systems

作者： Quanchao Liu Heyan Huang Changzhi Wang Chong Feng Beijing Engineering Applications Research Center of High Volume Language Information Processing and Cloud Computing Beijing Institute of Technology

We propose an algorithm of automatically identifying maximal-length evaluation (MEP) based on statistics and rules. The principle of the algorithm is that transform the problem of evaluation phrase recognition into a sequence tagging problem, use conditional random fields model for recognizing simple structure phrase of evaluation, and then automatically identify MEP with complex structure based on the establishment and application of the rule base. The testing precision and recall rate reached 72.38%. Based on the work above, we proposed a new algorithm of automatically extracting collocation of evaluation based on rules and make the automatic extraction of evaluation object and the MEP become reality. Experiment was conducted at Netease car portal for test, and got a higher accuracy rate.

关键词： Maximal-length evaluation phrase Conditional random fields Sentiment evaluation unit Collocation of evaluation Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Erasure Correcting Code to Optimize Partial Stripe Write in Raid-6

Erasure Correcting Code to Optimize Partial Stripe Write in ...

引用

2nd International Conference on Communication Technology(ICCT 2015)

作者： Ghulam Muhammad Shaikh Jiaqing Qu Quanxin Zhang Yu-an Tan Beijing Engineering Research Center of Massive Language Information Processing and Cloud Computing Application School of Computer Science and Technology Beijing Institute of Technology Shanghai Radio Equipment Research Institute

In Large Scale database system storage system, Raid 6 is to be used to tolerate failure disks and provide high reliability with the support of different codes. In large system there is no any risk of data loss or disk failure but in case of disaster or suddenly something happened with the risk of data failure Raid can be tolerate more than three disk failure. In this paper we optimize the partial stripe write with the erasure correcting code. Moreover, some erasure correcting can be evaluate or optimize on different codes in Raid, here we focus on partial stripe write where data can be written on disk either fully or partially.

关键词： RAID Erasure Correcting Code Partial Stripe Write Algorithm of partial stripe

来源：评论

学校读者我要写书评

暂无评论

SSD lifespan-oriented cache optimization in SSD/HDD hybrid storage system

SSD lifespan-oriented cache optimization in SSD/HDD hybrid s...

引用

2014 International Conference on Environmental Engineering and Computer application (ICEECA 2014)

作者： X.Wang Q.X.Zhang Y.A.Tan J.Q.Qu Beijing Engineering Research Center of Massive Language Information Processing and Cloud Computing Application School of Computer Science and Technology Beijing Institute of Technology Shanghai Radio Equipment Research Institute

Layered SSD/HDD hybrid storage system has been widely used. However, most traditional designs are performance oriented, and not optimized for SSD lifespan. In this paper, SSD lifespan oriented layered model and cache algorithm are proposed. To decrease SSD wear rate, HDD-RAM switch tunnel is appended to the traditional layered hybrid structure. Cache window size is considered in cache algorithm to filter data, avoiding unnecessary cache cost. Results show optimized designs significantly reduced erase times on SSD, and slightly influenced original performance.

关键词： SSD lifespan-oriented cache optimization in SSD/HDD hybrid storage system HDD RAM cache

来源：评论

学校读者我要写书评

暂无评论

Patent literatures translation system based on hadoop

引用

9th FTRA InternationalConference on Future information Technology, FutureTech 2014

作者： Zhang, Di Huang, Heyan Huang, Yonggang Beijing Eng. Res. Ctr. of High Vol. Language Information Processing and Cloud Computing Applications Beijing Institute of Technology Beijing 100081 China School of Computer Science and Technology Beijing Institute of Technology Beijing 100081 China

ISBN: (纸本)9783642550379

In order to tackle the slow response caused by massive patent literatures, a patent literatures translation system based on Hadoop is proposed in this paper. The paper presents a hybrid storage structure and a parallel translation model for massive patent literatures. The hierarchical storage structure is based on HDFS (Hadoop Distributed File System), which stores the patent documents and HBase where directories of such data are stored. This hybrid structure enables faster retrieval through the distributed file system. In translation, The Hadoop MapReduce framework is utilized. The MapReduce computation model not only can translate the patent literatures in highly parallel, but also can process multiple documents simultaneously. The experimental results show that the proposed machine translation system in this paper has better translation performance than the conventional machine translation approach. © 2014 Springer-Verlag Berlin Heidelberg.

关键词： MapReduce

来源：评论

学校读者我要写书评

暂无评论

Noisy smoothing image source identification 1

引用

9th International Symposium on Cyberspace Safety and Security, CSS 2017

作者： Liu, Yuying Huang, Yonggang Zhang, Jun Liu, Xu Shen, Hualei School of Computer Science and Technology Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Applications Beijing Institute of Technology Beijing100081 China School of Information Technology Deakin University MelbourneVIC3125 Australia College of Computer and Information Engineering Henan Normal University Xinxiang453007 China

ISBN: (数字)9783319694719

ISBN: (纸本)9783319694702

Feature based image source identification plays an important role in the toolbox for forensics investigations on images. Conventional feature based identification schemes suffer from the problem of noise, that is, the training dataset contains noisy samples. To address this problem, we propose a new Noisy Smoothing Image Source Identification (NS-ISI) method. NS-ISI address the noise problem in two steps. In step 1, we employ a classifier ensemble approach for noise level evaluation for each training sample. The noise level indicates the probability of being noisy. In step 2, a noise sensitive sampling method is employed to sample training samples from original training set according to the noise level, producing a new training dataset. The experiments carried out on the Dresden image collection confirms the effectiveness of the proposed NS-ISI. When the noisy samples present, the identification accuracy of NS-ISI is significantly better than traditional methods. © 2017, Springer International Publishing AG.

关键词： Sampling

来源：评论

学校读者我要写书评

暂无评论

Automatic Verification of Data-centric Web Service Specifications

引用

IERI Procedia 2013年 4卷 93-98页

作者： Zhi Fang Lejian Liao Ruoyu Chen Beijing Engineering Research Centre of High Volume Language Information Processing & Cloud Computing Applications School of Computer Science Beijing Institute of TechnologyBeijing 100081 China

Amodel are proposed for modeling data-centric Web services which are powered by relational databases and interact with users according to logical formulas specifying input constraints, control-flow constraints and state/output/action rules. The Linear Temporal First-Order Logic (LTL-FO) formulas over inputs, states, outputs and actions are used to express the properties to be *** have proven that automatic verification of LTL-FO properties of data-centric Web services under input-bounded constraints is decidable by reducing Web services to data-centric Web applications. Thus, we can verify Web service specifications using existing verifier designed for Web applications.

关键词： Web Service Automatic Verification First-Order Linear Temporal Logic

来源：评论

学校读者我要写书评

暂无评论

How to protect query and report privacy without sacrificing service quality in participatory sensing

How to protect query and report privacy without sacrificing ...

引用

IEEE International Conference on Performance, computing and Communications (IPCCC)

作者： Meng Li Fan Wu Guihai Chen Liehuang Zhu Zijian Zhang Beijing Engineering Research Center of Massive Language Information Processing and Cloud Computing Application Beijing Institute of Technology China Department of Computer Science and Engineering Shanghai Jiao Tong University China

The ubiquity of mobile devices has brought forth the concept of participatory sensing, whereby people can collect and share data from ambient environment for the benefit of themselves or community. To encourage participation of all stakeholders and guarantee system functionality, a privacy-preserving participatory sensing system should be established to hide querier's and participant's sensitive information(e.g., interest, location and content). Meanwhile, it is also imperative for server to provide an accurate and quick service(match the “need” with “supply” and retrieve the desired result from collected data set) for queriers when queries and reports are encrypted to protect privacy. In this paper, we propose Query and Report privacy-preserving protocol(QueRe) in participatory sensing system aiming to protect the query privacy and report privacy without sacrificing the service quality. Security analysis and performance simulation show our method achieves superior performance in privacy protection and service quality. To the best of our knowledge, our work is first attempt for protecting query privacy and report privacy while considering server's service quality.

关键词： Privacy Servers Sensors Data privacy Cryptography Protocols Measurement

来源：评论

学校读者我要写书评

暂无评论

System Report for CCL23-Eval Task 9: Improving MRC Robustness with Overlapping Segments Generation for GCRC advRobust 22

System Report for CCL23-Eval Task 9: Improving MRC Robustnes...

引用

22nd Chinese National Conference on Computational Linguistics, CCL 2023

作者： He, Suzhe Yang, Chongsheng Shi, Shumin Beijing Institute of Technology Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Applications China

ISBN: (纸本)9781713889908

There are many challenges in machine reading comprehension when it comes to extracting semantically complete evidence for specific statement. Existing works on unsupervised evidence extraction can be mainly divided into two categories. The first category utilizes static word vectors and employs beam search iteratively to extract relevant sentences. The second category uses instance-level supervised methods, including independent evidence extraction and end-to-end evidence extraction. The former category involves a more complex process, while the latter category suffers from instability during joint training, which directly affects the stable improvement of model performance. In Task 9 of CCL23-Eval, this paper proposes an adaptive end-to-end evidence extraction method based on overlapping segments generation. This method addresses the uncertainty in evidence sentence boundaries by dividing the document into multiple overlapping sentence segments and extracting key parts as evidence to achieve overall semantic extraction. Simultaneously, the evidence extraction embedding module is optimized to achieve automatic adjustment of evidence segment confidence. Experimental results demonstrate that the proposed method greatly eliminates interference from redundant content, and with just one hyperparameter, it can stably improve the performance of the reading comprehension model, improving model robustness. © 2023 China National Conference on Computational Linguistics.

关键词： Semantics

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

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

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：