171,358 research outputs found
Decision confidence-based multi-level support vector machines
No full textSupport vector machines (SVM) have been showing high accuracy of prediction in many applications. However, as any statistical learning algorithm, SVM's accuracy drops if some of the training points are contaminated by an unknown source of noise. The choice of clean training points is critical to avoid the overfitting problem which occurs generally when the model is excessively complex, which is reflected by a high accuracy over the training set and a low accuracy over the testing set (unseen points). In this paper we present a new multi-level SVM architecture that splits the training set into points that are labeled as 'easily classifiable' which do not cause an increase in the model complexity and 'non-easily classifiable' which are responsible for increasing the complexity. This method is used to create an SVM architecture that yields on average a higher accuracy than a traditional soft margin SVM trained with the same training set. The architecture is tested on the well known US postal handwritten digit recognition problem, the Wisconsin breast cancer dataset and on the agitation detection dataset. The results show an increase in the overall accuracy for the three datasets. Throughout this paper the word confidence is used to denote the confidence over the decision as commonly used in the literature. © 2013 Elsevier Ltd. All rights reserved.Aronszajn N., 1950, INTRO THEORY HILBERT; Burges CJC, 1998, DATA MIN KNOWL DISC, V2, P121, DOI 10.1023-A:1009715923555; CORTES C, 1995, MACH LEARN, V20, P273, DOI 10.1023-A:1022627411411; Deniz O, 2003, PATTERN RECOGN LETT, V24, P2153, DOI 10.1016-S0167-8655(03)00081-3; Devijver P., 1982, PATTERN RECOGNITION; Dumais S., 1998, Proceedings of the 1998 ACM CIKM International Conference on Information and Knowledge Management, DOI 10.1145-288627.288651; Gammerman A., 1998, Uncertainty in Artificial Intelligence. Proceedings of the Fourteenth Conference (1998); Joachims T., 1998, MACH LEARN ECML 98, P137, DOI DOI 10.1007-BFB0026683; Kecman V., 2001, LEARNING SOFT COMPUT; KOLMOGOR.AN, 1968, INT J COMPUT MATH, V2, P157, DOI 10.1080-00207166808803030; Le Cun BB, 1990, ADV NEURAL INFORM PR; Li L, 2006, LECT NOTES COMPUT SC, V4282, P437; Mitra P, 2004, IEEE T PATTERN ANAL, V26, P413, DOI 10.1109-TPAMI.2004.1262340; Platt J. C., 1999, ADV LARGE MARGIN CLA, P61; Sakr GE, 2010, IEEE T AFFECT COMPUT, V1, P98, DOI 10.1109-T-AFFC.2010.2; Saunders C., 1999, INT JOINT C ART INT, V2, P722; Scholkopf B., 2002, LEARNING KERNELS; Street N., 1992, NUCL FEATURE EXTRACT; Vapnik V., 1998, STAT LEARNING THEORY; Vapnik V.N., 1995, NATURE STAT LEARNING; Wu YC, 2008, PATTERN RECOGN, V41, P2874, DOI 10.1016-j.patcog.2008.02.0101
Multi level SVM for subject independent agitation detection
No full textThe need to automate the detection of agitation for dementia patients is a major requirement for caregivers. This research aims at detecting the agitation status of the subjects using soft computing techniques that does not require supervision beyond the training phase. An autonomous multi-sensory device has been developed to achieve automatic assessment of agitation and to control stimulation that will reduce the agitation level automatically. The focus of this paper is the agitation detection algorithm. Three vital signs are monitored for agitation detection: the Heart Rate (HR) the Galvanic Skin Response (GSR) and Skin Temperature (ST). These measures are fed into a new SVM architecture: The Multi level SVM learning machine. Results show very high detection accuracy of agitation, quick adaptation to the subject and a strong correlation between the physiological signals monitored and the emotional states of the subjects. The result is a learning algorithm that is Subject-Independent. ©2009 IEEE.American Psychiatric Association, 1994, DIAGN STAT MAN MENT; CULTER NR, 1996, UNDERSTANDING ALZHEI, P65; FOOK VFS, 2007, E HLTH NETW APPL SER, P68; Kecman V., 2001, LEARNING SOFT COMPUT; Kistler A, 1998, INT J PSYCHOPHYSIOL, V29, P35, DOI 10.1016-S0167-8760(97)00087-1; LIAO WH, 2005, IEEE COMP SOC C, V3, P70; Murray DR, 2003, CHEST, V123, P664, DOI 10.1378-chest.123.3.664; ROD K, 2000, INT J PSYCHOPHYSIOL, V37, P121; Rosenblatt Adam, 2005, Cleve Clin J Med, V72 Suppl 3, pS3; SAKR GE, 2008, ADV INT MECH IEEE AS; Tamura T., 1997, P 19 ANN INT C IEEE, V3, P999, DOI 10.1109-IEMBS.1997.756513; TULEN JHM, 1989, PHARMACOL BIOCHEM BE, V32, P9, DOI 10.1016-0091-3057(89)90204-9; *US BUR CENS, 2006, 65 US; Vapnik V.N., STAT LEARNING THEORY; Zhai J., 2006, FLAIRS C, P39532
Dataset for Tubular anti-resonant hollow core fiber for visible Raman spectroscopy
No full textThis dataset supports the publication:
AUTHORS:Ian A. Davidson, Matthew Partridge, John R. Hayes, Yong Chen, Thomas D. Bradley, Hesham Sakr, Shuichiro Rikimi, Gregory T. Jasion, Eric Numkam Fokoua, Marco Petrovich, Francesco Poletti, David J. Richardson and Natalie V. Wheeler
TITLE: Tubular Anti-resonant Hollow Core Fiber for Visible Raman Spectroscopy
JOURNAL: Sixth International Workshop on Specialty Optical Fibers and Their Applications (WSOF 2019) Conference Digest
PAPER DOI: 10.1117/12.2557110</span
Support vector machines to define and detect agitation transition
No full textThe need to automate the detection of agitation and the detection of agitation transition for dementia patients is a significant facilitator for caregivers. This research aims at detecting the transitional phase toward agitation, as well as agitation detection of subjects, using soft computing techniques that do not require supervision beyond the training phase. Three vital signs are monitored: Heart Rate (HR), Galvanic Skin Response (GSR), and Skin Temperature (ST). These measures are fed into two proposed SVM architectures which are based on the definition of a new confidence measure: Confidence-Based SVM and Confidence-Based Multilevel SVM. Results show very high detection accuracy of agitation and agitation transition, a quick adaptation to the subject, and a strong correlation between the physiological signals monitored and the emotional states of the subjects. Another challenge that is successfully addressed in this paper is the ability to train the classifier on a limited group of subjects, and then test it on subjects not belonging to the training group. The result is a learning algorithm that is Subject-Independent. © 2010 IEEE.[Anonymous], 1994, DSM 4 DIAGN STAT MAN; Aronszajn N., 1950, INTRO THEORY HILBERT; Burges CJC, 1998, DATA MIN KNOWL DISC, V2, P121, DOI 10.1023-A:1009715923555; Dishman RK, 2000, INT J PSYCHOPHYSIOL, V37, P121, DOI 10.1016-S0167-8760(00)00085-4; Ekman P., 1984, APPROACHES EMOTION, V3, P319; Fook V., 2007, P 9 INT C E HLTH NET, P68; HE W, 2005, CURRENT POPULATION R, pP23; Kecman V., 2001, LEARNING SOFT COMPUT; Kim KH, 2004, MED BIOL ENG COMPUT, V42, P419, DOI 10.1007-BF02344719; Kistler A, 1998, INT J PSYCHOPHYSIOL, V29, P35, DOI 10.1016-S0167-8760(97)00087-1; Lee J., 2010, EVALUATION CAPACITIV; Li L, 2006, LECT NOTES COMPUT SC, V4282, P437; Liao W., 2005, P IEEE COMP SOC C CO, P70; Malik M., 1996, EUROPEAN HEART J, V17, P354; Murray DR, 2003, CHEST, V123, P664, DOI 10.1378-chest.123.3.664; Murugappan M, 2008, IFMBE PROC, V21, P262; Pearson K, 1901, PHILOS MAG, V2, P559; Picard RW, 2001, IEEE T PATTERN ANAL, V23, P1175, DOI 10.1109-34.954607; Rosenblatt A., 2005, CLEV CLIN J MED, V72, P3; Sakr GE, 2009, IEEE ASME INT C ADV, P538, DOI 10.1109-AIM.2009.5229958; Sakr GE, 2008, IEEE ASME INT C ADV, P200, DOI 10.1109-AIM.2008.4601659; Spielberger C. D., 1970, STAI MANUAL STATE TR; Stroop JR, 1935, J EXP PSYCHOL, V18, P643, DOI 10.1037-0096-3445.121.1.15; Tamura T., 1997, P 19 ANN INT C IEEE, V3, P999, DOI 10.1109-IEMBS.1997.756513; Vapnik V., 1998, STAT LEARNING THEORY; Wenhui L., 2005, P IEEE CS C COMP VIS; Zhai J., 2006, P 19 INT FLOR ART IN, P395109
Dataset for: Interband Short Reach Data Transmission in Ultrawide Bandwidth Hollow Core Fiber
No full textThis dataset supports the publication: Sakr, H. et a. (2019). Interband Short Reach Data Transmission in Ultrawide Bandwidth Hollow Core Fiber. Journal of Lightwave Technology.
We report a Nested Antiresonant Nodeless hollow-core Fiber (NANF) operating in the first antiresonant passband. The fiber has an ultrawide operational bandwidth of 700 nm, spanning the 1240–1940 nm wavelength range that includes the O-, S-, C- and L- telecoms bands.</span
Efficient forest fire occurrence prediction for developing countries using two weather parameters
No full textForest fire occurrence prediction plays a major role in resource allocation, mitigation and recovery efforts. This paper compares two artificial intelligence based methods, artificial neural networks (ANN) and support vector machines (SVM), utilizing a reduced set of weather parameters. Using a reduced set of parameters results in an efficient and reduced cost prediction system especially for developing countries. In this paper the aim is to predict forest fire occurrence by reducing the number of monitored features, and eliminating the need for weather prediction mechanisms. The reason is to reduce errors due to inaccuracies in weather prediction. The challenge is to choose a limited number of easily measurable features in the aim of reducing the cost of the system and its maintenance. At the same time, the chosen features must have a high correlation with the risk of fire occurrence. A literature review of forest fire prediction methods divided into systems-indices, and artificial intelligence is provided. The two fire danger prediction algorithms utilize relative humidity and cumulative precipitation to output a risk estimate. The assessment of these algorithms, using data from Lebanon, demonstrated their ability to accurately predict the risk of fire occurrence on a scale of four levels. © 2010 Elsevier Ltd. All rights reserved.Alonso-Betanzos A, 2003, EXPERT SYST APPL, V25, P545, DOI 10.1016-S0957-4174(03)00095-2; Aronszajn N., 1950, INTRO THEORY HILBERT; Brillinger DR, 2003, INST MATH S, V40, P177; Cheng Tao, 2008, Transactions in GIS, V12, P591; Cheng T., 2006, ISPRS TECHN COMM 7 M, P148; CLARKE KC, 1994, PHOTOGRAMM ENG REM S, V60, P1355; Cortez P., 2007, P 13 EPIA 2007 PORT, P512; Deeming J.E., 1977, INT39 USDA FOR SERV, P63; Dunn A, 2004, LECT NOTES COMPUT SC, V3305, P395; Fiorucci P, 2008, ENVIRON MODELL SOFTW, V23, P690, DOI 10.1016-j.envsoft.2007.05.008; *FOR CAN FDGC, 1992, STX3 FOR CAN; Hagan M. T., 1996, NEURAL NETWORK DESIG; HAGAN MT, 1994, IEEE T NEURAL NETWOR, V5, P989, DOI 10.1109-72.329697; HAN J, 2003, LECT NOTES COMPUTER, P370; Iliadis LS, 2005, ENVIRON MODELL SOFTW, V20, P613, DOI 10.1016-j.envsoft.2004.03.006; Jaiswal R. K., 2002, International Journal of Applied Earth Observation and Geoinformation, V4, P1, DOI 10.1016-S0303-2434(02)00006-5; Kecman V., 2001, LEARNING SOFT COMPUT; Li Zhanqing, 2001, P199; Mazloumi E, 2011, ENG APPL ARTIF INTEL, V24, P534, DOI 10.1016-j.engappai.2010.11.004; McArthur A.G, 1966, WEATHER GRASSLAND FI, P100; Mitri GH, 2004, INT J WILDLAND FIRE, V13, P367, DOI 10.1071-WF03079; Muzy A., 2001, ACT C ESS 2001 C, P641; Ntaimo L., 2006, SIMULATION SERIES, V38, P103; Rothermel R. C., 1972, MATH MODEL PREDICTIN; Sakr GE, 2010, IEEE T AFFECT COMPUT, V1, P98, DOI 10.1109-T-AFFC.2010.2; Scales L.E., 1985, INTRO NONLINEAR OPTI; STOCKS BJ, 1989, FOREST CHRON, V65, P450; Ubeyli ED, 2008, ENG APPL ARTIF INTEL, V21, P1196, DOI 10.1016-j.engappai.2008.03.012; van Wagner C., 1987, DEV STRUCTURE CANADI; Van Wagner CE, 1985, EQUATIONS FORTRAN PR; VANNEST T, 1999, NAT INT FIR BEH WORK; Vapnik V., 1998, STAT LEARNING THEORY; VIEGAS DX, 2000, INT J WILDLAND FIRE, V9, P235; Wiering M., 1998, P 12 INT S COMP SCI, P37843
Red cell transfusion triggers in critically ill patients:time for some new TRICCs?
No full textCurrent evidence suggests that critically ill patients tolerate anaemia well and that blood transfusions may increase the risk of adverse outcomes. Dr Sakr and colleagues present a contradictory analysis of a surgical ICU cohort, finding an association between blood transfusions and lower hospital mortality after adjustment for a range of potential confounders. Analyses of this kind are interesting and provocative, but are limited by residual confounding and bias by indication. The data emphasise the need for additional high quality trials of transfusion practice in critical care
Going Beyond Counting First Authors in Author Co-citation Analysis
No full textThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Evolutionary dynamics of phenotypic plasticity in metastatic colorectal cancer
No full textWINTER MEETING Pathology and precision medicine - Abstract RF4Mossner, M., Househam, J., Kyle, P., Donato-Brown, D., Woods, S.L., Heide, T., Baker, A.M., Sakr, C., Stafford, A.T., Flood, M., Murphy, J., Sottoriva, A., Graham, T.A
- …
