Filters








372,332 Hits in 9.1 sec

Sensor Selection for IT Infrastructure Monitoring [chapter]

Gergely János Paljak, Imre Kocsis, Zoltán Égel, Dániel Tóth, András Pataricza
2010 Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering  
The selection of a compact, but sufficiently characteristic set of control variables is one of the core problems both for design and run-time complexity.  ...  Supervisory control is the main means to assure a high level performance and availability of large IT infrastructures.  ...  We use the systematic, well-tunable mRMR algorithm for variable selection. It is also based on mutual information and has been shown to scale well for large problem spaces [10, 11] .  ... 
doi:10.1007/978-3-642-11482-3_9 fatcat:kiwsyjgukfhltjsthd4qjf4ksu

A Data-analytics Approach for Enterprise Resilience

Donna Xu, Ivor W. Tsang, Eng K. Chew, Cosimo Siclari, Varun Kaul
2019 IEEE Intelligent Systems  
For example, a model can be trained on the features designed as system performance metrics, to detect an incident in the system.  ...  However, such generated alarms result in a loss of information, such as the actual value of a performance metric that is used to generate the alarms, and other performance metrics that imply bad system  ... 
doi:10.1109/mis.2019.2918092 fatcat:jvk4bgzutfgqrohwtqw7cf7sn4

Performance Issue Diagnosis for Online Service Systems

Qiang Fu, Jian-Guang Lou, Qing-Wei Lin, Rui Ding, Dongmei Zhang, Zihao Ye, Tao Xie
2012 2012 IEEE 31st Symposium on Reliable Distributed Systems  
Given a detected performance issue and collected system metrics for an online service system, engineers usually need to make great efforts to conduct diagnosis by first identifying performance issue beacons  ...  In order to reduce the manual efforts, in this paper, we propose a new approach to effectively detecting performance issue beacons to help with performance issue diagnosis.  ...  These approaches fail to detect the metrics of "CPU usage" and "blocking query", which contain diagnosis information in fine granularity.  ... 
doi:10.1109/srds.2012.49 dblp:conf/srds/FuLLDZYX12 fatcat:zcnnr5to2raobf7osa4loncf7y

Search-Based Refactoring Detection Using Software Metrics Variation [chapter]

Rim Mahouachi, Marouane Kessentini, Mel Ó Cinnéide
2013 Lecture Notes in Computer Science  
To this end, we have developed an approach to automate the detection of source code refactorings using structural information extracted from the source code.  ...  In applying our approach to several versions of four open source projects we find the average Precision and Recall to be over 90%, thus confirming the effectiveness of our detection approach.  ...  Several approaches are tailored to detect refactorings in program code. For instance, Dig et al. [2] propose an approach to detect applied refactorings in Java code.  ... 
doi:10.1007/978-3-642-39742-4_11 fatcat:b5qoxql3v5gytewhyxa2mqmgbe

Hardware performance counter-based problem diagnosis for e-commerce systems

Keith A. Bare, Soila Kavulya, Priya Narasimhan
2010 2010 IEEE Network Operations and Management Symposium - NOMS 2010  
Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining  ...  Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for  ...  Finally, the target system and an overview of the approach to solving the problem were presented.  ... 
doi:10.1109/noms.2010.5488457 dblp:conf/noms/BareKN10 fatcat:bx2kpbaqyfclhhopelaqo67ere

Performance-Aware Management of Cloud Resources: A Taxonomy and Future Directions [article]

Sara Kardani-Moghaddam, Rajkumar Buyya, Kotagiri Ramamohanarao
2018 arXiv   pre-print
Therefore, a continuous monitoring of system attributes and performance metrics provide the raw data for the analysis of problems affecting the performance of the application.  ...  Obtained knowledge form the data analysis process helps to find out about the changes in the workloads, faulty components or problems that can cause system performance to degrade.  ...  An advantage of a master node approach is to have all system related performance information in one place; therefore, one can analyze the interactions and relation among the metrics at different layers  ... 
arXiv:1808.02254v1 fatcat:iw4t5vp43jgo3b4kz3j3drk3ia

The Importance of Metric Learning for Robotic Vision: Open Set Recognition and Active Learning [article]

Benjamin J. Meyer, Tom Drummond
2019 arXiv   pre-print
Further to detecting novel examples, we propose an open set active learning approach that allows a robot to efficiently query a user about unknown observations.  ...  We show how a deep metric learning classification system can be applied to such open set recognition problems, allowing the classifier to label novel observations as unknown.  ...  In addition to detecting observations belonging to N , our system should select the most informative examples in U for labelling by a user. Obtaining a label is referred to as a query.  ... 
arXiv:1902.10363v1 fatcat:3id4abhr7jaz5l2dq6m435646u

An information theoretic approach for tracker performance evaluation

K Kao Edward, P Daggett Matthew, B Hurley Michael
2009 2009 IEEE 12th International Conference on Computer Vision  
Existing evaluation approaches assess tracker performance through measures of correspondence between ground truth tracks and system tracks using metrics such as track detection rate, track completeness  ...  Towards this end, this paper presents a pair of information theoretic metrics with similar behavior to the Receiver Operating Characteristic (ROC) curves of signal detection theory.  ...  Acknowledgement The authors would like to thank several Lincoln Laboratory colleagues: Gary Condon, Peter Jones, and Luke Skelly for technical discussions and feedback; Luke Skelly, Robert Garnier, and  ... 
doi:10.1109/iccv.2009.5459275 dblp:conf/iccv/KaoDH09 fatcat:siwzdkg3jncnjf7zvdaon2vlnm

Analyzing IoT Attack Feature Association with Threat Actors

Muhammad Shafiq, Zhaoquan Gu, Shah Nazir, Rahul Yadav, Barbara Guidi
2022 Wireless Communications and Mobile Computing  
Feature selection is an important part of problem formulation in machine learning.  ...  To overcome the above problems, this paper proposes a novel feature selection framework RFS for IoT attack detection using machine learning (ML) techniques.  ...  Acknowledgments This work is supported in part by the Guangdong Province Key Research and Development Plan (Grant No. 2019B010136003), the National Natural Science Foundation of China (61902082), the Guangdong  ... 
doi:10.1155/2022/7143054 fatcat:i7vpb6kkjvb6dajoc3hccevetu

Semi-Supervised Self-Training of Object Detection Models

C. Rosenberg, M. Hebert, H. Schneiderman
2005 2005 Seventh IEEE Workshops on Applications of Computer Vision (WACV/MOTION'05) - Volume 1  
In this work we present a semi-supervised approach to training object detection systems based on self-training.  ...  fully labeled data, and that a training data selection metric that is defined independently of the detector greatly outperforms a selection metric based on the detection confidence generated by the detector  ...  approaches to object detection problems.  ... 
doi:10.1109/acvmot.2005.107 dblp:conf/wacv/RosenbergHS05 fatcat:dm6laopk6rhdziylbesafjg5xy

Sensor Selection and Optimization for Health Assessment of Aerospace Systems

William A. Maul, George Kopasakis, Louis M. Santi, Thomas S. Sowers, Amy Chicatelli
2008 Journal of Aerospace Computing Information and Communication  
Furthermore, the safety and reliability requirements are met through sensor suite augmentation in an ad hoc, heuristic manner, rather than any systematic approach.  ...  Furthermore, the safety and reliability requirements are met through sensor suite augmentation in an ad hoc, heuristic manner, rather than any systematic approach.  ...  Project under the NASA Aviation Safety Program for their interest in and support of this effort.  ... 
doi:10.2514/1.34677 fatcat:mbhegzbignckbhbf6f3svd75tm

Software Metrics and tree-based machine learning algorithms for distinguishing and detecting similar structure design patterns

Mohammad Y. Mhawish, Manjari Gupta
2019 SN Applied Sciences  
In this paper, we propose a design pattern detection approach based on tree-based machine learning algorithms and software metrics to study the effectiveness of software metrics in distinguishing between  ...  Design patterns are general reusable solutions for recurrent occurring problems.  ...  They developed an approach as a tool for detecting design patterns-PAT. Keller et al. [10] proposed an approach for design pattern detection that splits the detection process into two phases.  ... 
doi:10.1007/s42452-019-1815-3 fatcat:gso7znyqfjbvfdhnxpfrawvsvm

Performance anomaly detection using isolation‐trees in heterogeneous workloads of web applications in computing clouds

Sara Kardani‐Moghaddam, Rajkumar Buyya, Kotagiri Ramamohanarao
2019 Concurrency and Computation  
In order to efficiently manage the resources, a continuous analysis of the operational state of the system is required to be able to detect the performance degradations and malfunctioned resources as soon  ...  In this paper, we propose Isolation-Forest based anomaly detection (IFAD) framework based on the unsupervised Isolation technique for anomaly detection in a multi-attribute space of performance indicators  ...  We like to thank Adel Nadjaran Toosi for his comments on improving the paper.  ... 
doi:10.1002/cpe.5306 fatcat:vy3xxsdbwraudahehpb2kprq2y

Ranking of Social Media Alerts with Workload Bounds in Emergency Operation Centers

Hemant Purohit, Carlos Castillo, Muhammad Imran, Rahul Pandey
2018 2018 IEEE/WIC/ACM International Conference on Web Intelligence (WI)  
This paper presents a novel problem and a model to quantify the relationship between the performance metrics of ranking systems (e.g., recall, NDCG) and the bounds on the user workload.  ...  Existing ranking systems can provide a baseline for selecting which updates or alerts to push to emergency responders.  ...  We formulate a novel problem of how to create an alert ranking system that is adaptive to the bounds on user performance, for deciding how many alerts to generate and when.  ... 
doi:10.1109/wi.2018.00-88 dblp:conf/webi/Purohit00P18 fatcat:7hdg7rettrbcrkpjxbanqc2way

Ranking of Social Media Alerts with Workload Bounds in Emergency Operation Centers [article]

Hemant Purohit, Carlos Castillo, Muhammad Imran, Rahul Pandey
2018 arXiv   pre-print
This paper presents a novel problem and a model to quantify the relationship between the performance metrics of ranking systems (e.g., recall, NDCG) and the bounds on the user workload.  ...  Existing ranking systems can provide a baseline for selecting which updates or alerts to push to emergency responders.  ...  We formulate a novel problem of how to create an alert ranking system that is adaptive to the bounds on user performance, for deciding how many alerts to generate and when.  ... 
arXiv:1809.08489v1 fatcat:e3nncq7sbnh77nr726igggkm6q
« Previous Showing results 1 — 15 out of 372,332 results