Search For:

Displaying 1-28 out of 28 total
Unbiased Metric Learning: On the Utilization of Multiple Datasets and Web Images for Softening Bias
Found in: 2013 IEEE International Conference on Computer Vision (ICCV)
By Chen Fang,Ye Xu,Daniel N. Rockmore
Issue Date:December 2013
pp. 1657-1664
Many standard computer vision datasets exhibit biases due to a variety of sources including illumination condition, imaging system, and preference of dataset collectors. Biases like these can have downstream effects in the use of vision datasets in the con...
 
Multi-instance Metric Learning
Found in: Data Mining, IEEE International Conference on
By Ye Xu,Wei Ping,Andrew T. Campbell
Issue Date:December 2011
pp. 874-883
Multi-instance learning, like other machine learning and data mining tasks, requires distance metrics. Although metric learning methods have been studied for many years, metric learners for multi-instance learning remain almost untouched. In this paper, we...
 
Exploiting Social Networks for Large-Scale Human Behavior Modeling
Found in: IEEE Pervasive Computing
By Nicholas D. Lane,Ye Xu,Hong Lu,Andrew T. Campbell,Tanzeem Choudhury,Shane B. Eisenman
Issue Date:October 2011
pp. 45-53
The Cooperative Communities (CoCo) learning framework leverages everyday social connections between people to personalize classification models. By exploiting social networks, CoCo spreads the burden of providing training data over an entire community.
 
Cooperative Communities (CoCo): Exploiting Social Networks for Large-scale Modeling of Human Behavior
Found in: IEEE Pervasive Computing
By Nicholas Lane,Ye Xu,Hong Lu,Shane Eisenman,Tanzeem Choudhury,Andrew Campbell
Publication Date: September 2011
pp. N/A
Human behavior modeling at a large-scale and under real-world conditions is still an open problem. Existing classification models do not always perform well on a diverse population. Training personalized models that incorporate different contextual conditi...
 
Power Failure Monitoring and Locating System Based on Google Maps
Found in: 2012 Fourth International Conference on Computational and Information Sciences (ICCIS)
By Jianliang Meng,Haoquan Zhao,Ye Xu
Issue Date:August 2012
pp. 337-340
The power industry is the basis of nation building and the lifeblood of the national economy. With the continuous expansion of the electrical grid size, protecting the user's normal continuous supply of electricity is the top priority of electric service. ...
 
Comparison of Two Algorithms of Attribute Reduction Based on Fuzzy Rough Set
Found in: 2012 Fourth International Conference on Computational and Information Sciences (ICCIS)
By Jianliang Meng,Ye Xu,Junwei Zhang
Issue Date:August 2012
pp. 542-545
Currently, with the large number of data and the increasing importance of it, how to find useful pattern in the large data, has become an important application of data mining. The rough set attribute reduction algorithm, used to study how to contain the sa...
 
The Research and Design of the Indoor Location System Based on RFID
Found in: Computational Intelligence and Design, International Symposium on
By Yongbo Liu,Huaichang Du,Ye Xu
Issue Date:October 2011
pp. 87-90
this paper introduces the concepts of RFID (Radio Frequency Identification) systems and analyses the key technologies about RFID systems, then researches the theory and algorithms of indoor location systems, especially LANDMARC system. In the end, this pap...
 
On a Clustering Method for Handwritten Digit Recognition
Found in: Intelligent Networks and Intelligent Systems, International Workshop on
By Ye Xu, Wei Zhang
Issue Date:November 2010
pp. 112-115
In this paper, a clustering method for handwritten digit recognition is studied. The digit samples, firstly are processed and features are extracted. Based on these features, a clustering method is designed and implemented to cluster the digit samples. Exp...
 
On an IBA Topology Model for Internet Topology Structure
Found in: Networks Security, Wireless Communications and Trusted Computing, International Conference on
By Ye Xu, Zhuo Wang
Issue Date:April 2010
pp. 114-117
Modeling of Internet topology structure is studied in this paper. First, measuring results of Internet topology from CAIDA monitors were handled to produce a complete topology sample. With this sample, research approaches of the frequency-degree power-law,...
 
Approaches of Spectra in Evaluations of Internet Topology Graph
Found in: Networks Security, Wireless Communications and Trusted Computing, International Conference on
By Bin Zhu, Ye Xu
Issue Date:April 2010
pp. 297-300
A way of evaluation of a topology structure model, especially the Internet topology structure model, was studied. SLS (Signless Laplacian Spectra), is used to quantitatively identify the topology properties of the Internet generated by the model and the In...
 
Approaches in Evaluations of Internet Topology Modeling
Found in: Education Technology and Computer Science, International Workshop on
By Ye Xu, Zhuo Wang
Issue Date:March 2010
pp. 31-34
An effective approach to evaluate Internet topology model was mainly studied in this paper. SLS (Signless Laplacian Spectra), is used to quantitatively identify the topology properties of the Internet generated by the model and the Internet out of real mea...
 
A Novel IDS Model Based on a Bayesian Fusion Approach
Found in: Multimedia Information Networking and Security, International Conference on
By Ye Xu, Wen-bo Zhang
Issue Date:November 2009
pp. 546-549
A novel IDS model based on Bayesian fusion approach is studied in this paper. The model comprises several parts such as the pixel level fusion, feature level fusion, decision-making level fusion and so on. Experiments show that the model could result in a ...
 
On a Model of Internet Topology Structure
Found in: Multimedia Information Networking and Security, International Conference on
By Ye Xu, Zhuo Wang
Issue Date:November 2009
pp. 378-382
On the basis of power-law approaches such as the frequency-degree, degree-rank, Internet topology modeling was studied and obvious power-law properties together with corresponding power exponents depicting characters of Internet topology structure were fou...
 
Financial Support and Electronic Communications Industry: An Analysis Based on Chinese Provincial Panel Data
Found in: Management of e-Commerce and e-Government, International Conference on
By Ye Xu, Chengwu Kuang
Issue Date:September 2009
pp. 554-557
Based on spatial panel data models and province data of Chinese electronic communications industry in 1997-2007, this paper gets an estimate of Chinese electronic communications industry’s province level production fuction.We focus on the source of the fun...
 
On Dynamics of Student Networks
Found in: Hybrid Intelligent Systems, International Conference on
By Ye Xu, Zhuo Wang
Issue Date:August 2009
pp. 23-27
An instance of complex networks, Student Networks (SN), is mainly studied in this paper. Dynamic properties of SN were firstly studied and an improved Logistic Model was put foreword and was optimized by float-point GA. Then, fractal methods were applied t...
 
Verifiable Threshold Secret Sharing and Full Fair Secure Two-Party Computation
Found in: Advanced Science and Technology, International e-Conference on
By Jian-wei Ye, Xu-lu Jiao, Yong-zheng Zhang
Issue Date:March 2009
pp. 78-83
Based on the verifiable encryption and zero-knowledge proof protocols of Jarecki and Shmatikov and Pedersen’s verifiable threshold secret sharing scheme, this paper proposes a new full fair secure two-party computation protocols. For getting full fair, the...
 
Research on the Architecture of Policy-Based Satellite Network Management
Found in: Intelligent Systems Design and Applications, International Conference on
By Wen-bo Zhang, Ye Xu, Yue-qiu Jiang
Issue Date:November 2008
pp. 672-676
In Satellite network the nodes are far from the terrestrial station, so the transmission delay between the satellite and the terrestrial station is long, furthermore the nodes mobility is intensive. Based on those characteristics an new network management ...
 
Analysis of Resources Allocation Efficiency of China?s Software Industry ? Based on Empirical Analysis of China?s Eastern Region
Found in: Management of e-Commerce and e-Government, International Conference on
By Ye Xu, Qiuyan Zhang
Issue Date:October 2008
pp. 277-282
First, this paper uses the traditional DEA method to measure the resource allocation efficiency of the eastern part’s software industry in 2005. Then, by using DEA cross-efficiency to make sort of the software industry resources allocation analysis. The re...
 
Research on a Computing Model for Internet Efficiency Index
Found in: International Conference on Internet Computing in Science and Engineering
By Ye Xu, Zhuo Wang, Wen-bo Zhang
Issue Date:January 2008
pp. 153-159
Internet Efficiency Index is a property indicating the performance of Internet and was studied in this paper. First, definitions and analyses of Internet Density and Internet Diameter were performed based on an Internet sample comprising more than 55 milli...
 
Short-Term and Long-Term Computing Models of Internet Diameter
Found in: International Conference on Internet Computing in Science and Engineering
By Ye Xu, Zhuo Wang, Wen-bo Zhang
Issue Date:January 2008
pp. 146-152
Internet Diameter is an Internet property defined in this paper being of great significance in measuring Internet performance. Approaches in short-term and long-term computing of Internet Diameter were mainly discussed. In short-term part, the computing ap...
 
Minimal Condensed Cube: Data Organization, Fast Computation, and Incremental Update
Found in: International Conference on Internet Computing in Science and Engineering
By Zhuo Wang, Ye Xu
Issue Date:January 2008
pp. 60-67
The condensed cube has been proposed to reduce the huge size of data cubes in OLAP system. The intuition of condensed cube is to compress semantically redundant tuples into their representative Base Single Tuples (BSTs). However, previous studies showed th...
 
ESCPP: An Evaluation and Alerting System for the Compatibility between Practice Guidelines and Practicing Medicine Using Arden Syntax
Found in: Information Science and Engineering, International Conference on
By Xingnan Liu, Ye Xu, Zuofeng Li, Lei Liu, Xiaoyan Zhang
Issue Date:December 2009
pp. 3649-3652
The clinical quality is influenced by the disseminating velocity of new medical knowledge and the rate of patients that receive evidence-based clinical service. In this study, we introduced an evaluation and alerting system for the compatibility between pr...
 
Incentivizing Advanced Load Scheduling in Smart Homes
Found in: Proceedings of the 5th ACM Workshop on Embedded Systems For Energy-Efficient Buildings (BuildSys'13)
By Prashant Shenoy, Ye Xu, David Irwin
Issue Date:November 2013
pp. 1-8
In recent years, researchers have proposed numerous advanced load scheduling algorithms for smart homes with the goal of reducing the grid's peak power usage. In parallel, utilities have introduced variable rate pricing plans to incentivize residential con...
     
Preference, context and communities: a multi-faceted approach to predicting smartphone app usage patterns
Found in: Proceedings of the 17th annual international symposium on International symposium on wearable computers (ISWC '13)
By Andrew Campbell, Giuseppe Cardone, Hong Lu, Mu Lin, Nicholas Lane, Tanzeem Choudhury, Ye Xu, Zhenyu Chen
Issue Date:September 2013
pp. 69-76
Reliable smartphone app prediction can strongly benefit both users and phone system performance alike. However, real-world smartphone app usage behavior is a complex phenomena driven by a number of competing factors. In this pa- per, we develop an app usag...
     
Feature selection for link prediction
Found in: Proceedings of the 5th Ph.D. workshop on Information and knowledge (PIKM '12)
By Dan Rockmore, Ye Xu
Issue Date:November 2012
pp. 25-32
Networks that model relationships in the real world have attracted much attention in the past few years. Link prediction plays a central role in the network area. Supervised learning is an important class of algorithms used to address the link prediction p...
     
TAKES: a fast method to select features in the kernel space
Found in: Proceedings of the 20th ACM international conference on Information and knowledge management (CIKM '11)
By Furao Shen, Jinxi Zhao, Wei Ping, Ye Xu
Issue Date:October 2011
pp. 683-692
Feature selection is an effective tool to deal with the "curse of dimensionality". To cope with the non-separable problem, feature selection in the kernel space has been investigated. However, previous study cannot adequately estimate the intrinsic dimensi...
     
Enabling large-scale human activity inference on smartphones using community similarity networks (csn)
Found in: Proceedings of the 13th international conference on Ubiquitous computing (UbiComp '11)
By Andrew T. Campbell, Feng Zhao, Hong Lu, Nicholas D. Lane, Shaohan Hu, Tanzeem Choudhury, Ye Xu
Issue Date:September 2011
pp. 355-364
Sensor-enabled smartphones are opening a new frontier in the development of mobile sensing applications. The recognition of human activities and context from sensor-data using classification models underpins these emerging applications. However, convention...
     
To obtain orthogonal feature extraction using training data selection
Found in: Proceeding of the 18th ACM conference on Information and knowledge management (CIKM '09)
By Jinxi Zhao, Osamu Hasegawa, Shen Furao, Ye Xu
Issue Date:November 2009
pp. 1819-1822
Feature extraction is an effective tool in data mining and machine learning. Many feature extraction methods have been investigated recently. However, few methods can achieve orthogonal components. Non-orthogonal components distort the metric structure of ...
     
 1