scispace - formally typeset

Institution

Delft University of Technology

EducationDelft, Zuid-Holland, Netherlands
About: Delft University of Technology is a(n) education organization based out in Delft, Zuid-Holland, Netherlands. It is known for research contribution in the topic(s): Population & Catalysis. The organization has 37681 authors who have published 94404 publication(s) receiving 2741710 citation(s). The organization is also known as: TU-Delft & Technische Hogeschool Delft.


Papers
More filters
Journal ArticleDOI
TL;DR: The objective of this review paper is to summarize and compare some of the well-known methods used in various stages of a pattern recognition system and identify research topics and applications which are at the forefront of this exciting and challenging field.
Abstract: The primary goal of pattern recognition is supervised or unsupervised classification. Among the various frameworks in which pattern recognition has been traditionally formulated, the statistical approach has been most intensively studied and used in practice. More recently, neural network techniques and methods imported from statistical learning theory have been receiving increasing attention. The design of a recognition system requires careful attention to the following issues: definition of pattern classes, sensing environment, pattern representation, feature extraction and selection, cluster analysis, classifier design and learning, selection of training and test samples, and performance evaluation. In spite of almost 50 years of research and development in this field, the general problem of recognizing complex patterns with arbitrary orientation, location, and scale remains unsolved. New and emerging applications, such as data mining, web searching, retrieval of multimedia data, face recognition, and cursive handwriting recognition, require robust and efficient pattern recognition techniques. The objective of this review paper is to summarize and compare some of the well-known methods used in various stages of a pattern recognition system and identify research topics and applications which are at the forefront of this exciting and challenging field.

6,278 citations

Journal ArticleDOI
TL;DR: A common theoretical framework for combining classifiers which use distinct pattern representations is developed and it is shown that many existing schemes can be considered as special cases of compound classification where all the pattern representations are used jointly to make a decision.
Abstract: We develop a common theoretical framework for combining classifiers which use distinct pattern representations and show that many existing schemes can be considered as special cases of compound classification where all the pattern representations are used jointly to make a decision. An experimental comparison of various classifier combination schemes demonstrates that the combination rule developed under the most restrictive assumptions-the sum rule-outperforms other classifier combinations schemes. A sensitivity analysis of the various schemes to estimation errors is carried out to show that this finding can be justified theoretically.

5,535 citations

Journal ArticleDOI
01 May 1998-Nature
Abstract: The use of individual molecules as functional electronic devices was first proposed in the 1970s (ref 1) Since then, molecular electronics2,3 has attracted much interest, particularly because it could lead to conceptually new miniaturization strategies in the electronics and computer industry The realization of single-molecule devices has remained challenging, largely owing to difficulties in achieving electrical contact to individual molecules Recent advances in nanotechnology, however, have resulted in electrical measurements on single molecules4,5,6,7 Here we report the fabrication of a field-effect transistor—a three-terminal switching device—that consists of one semiconducting8,9,10 single-wall carbon nanotube11,12 connected to two metal electrodes By applying a voltage to a gate electrode, the nanotube can be switched from a conducting to an insulating state We have previously reported5 similar behaviour for a metallic single-wall carbon nanotube operated at extremely low temperatures The present device, in contrast, operates at room temperature, thereby meeting an important requirement for potential practical applications Electrical measurements on the nanotube transistor indicate that its operation characteristics can be qualitatively described by the semiclassical band-bending models currently used for traditional semiconductor devices The fabrication of the three-terminal switching device at the level of a single molecule represents an important step towards molecular electronics

5,147 citations

Journal ArticleDOI
Abstract: We present the final data release of observations of ?21-cm emission from Galactic neutral hydrogen over the entire sky, merging the Leiden/Dwingeloo Survey (LDS: Hartmann & Burton 1997, Atlas of Galactic Neutral Hydrogen) of the sky north of ? = ?30? with the Instituto Argentino de Radioastronomia Survey (IAR: Arnal et al. 2000, AA and Bajaja et al. 2005, A&A, 440, 767) of the sky south of ? = ?25?. The angular resolution of the combined material is HPBW ? 0. ?6. The LSR velocity coverage spans the interval ?450 km s?1 to +400 km s?1, at a resolution of 1.3 kms?1. The data were corrected for stray radiation at the Institute for Radioastronomy of the University of Bonn, refining the original correction applied to the LDS. The rms brightness-temperature noise of the merged database is 0.07?0.09 K. Residual errors in the profile wings due to defects in the correction for stray radiation are for most of the data below a level of 20?40 mK. It would be necessary to construct a telescope with a main beam efficiency of ?MB >? 99% to achieve the same accuracy. The merged and refined material entering the LAB Survey of Galactic HI is intended to be a general resource useful to a wide range of studies of the physical and structural characteristices of the Galactic interstellar environment. The LAB Survey is the most sensitive Milky Way HI survey to date, with the most extensive coverage both spatially and kinematically.

4,122 citations

Journal ArticleDOI
Abstract: Classical least squares regression consists of minimizing the sum of the squared residuals. Many authors have produced more robust versions of this estimator by replacing the square by something else, such as the absolute value. In this article a different approach is introduced in which the sum is replaced by the median of the squared residuals. The resulting estimator can resist the effect of nearly 50% of contamination in the data. In the special case of simple regression, it corresponds to finding the narrowest strip covering half of the observations. Generalizations are possible to multivariate location, orthogonal regression, and hypothesis testing in linear models.

3,487 citations


Authors

Showing all 37681 results

NameH-indexPapersCitations
Albert Hofman2672530321405
Charles M. Lieber165521132811
Ad Bax13848697112
George C. Schatz137115594910
Georgios B. Giannakis137132173517
Jaap S. Sinninghe Damsté13472661947
Avelino Corma134104989095
Mark A. Ratner12796868132
Jing Kong12655372354
Robert J. Cava125104271819
Reza Malekzadeh118900139272
Jinde Cao117143057881
Mike S. M. Jetten11748852356
Liquan Chen11168944229
Oscar H. Franco11182266649
Network Information
Related Institutions (5)
Georgia Institute of Technology

119K papers, 4.6M citations

95% related

École Polytechnique Fédérale de Lausanne

98.2K papers, 4.3M citations

94% related

Technical University of Denmark

66.3K papers, 2.4M citations

94% related

ETH Zurich

122.4K papers, 5.1M citations

94% related

Hong Kong University of Science and Technology

52.4K papers, 1.9M citations

93% related

Performance
Metrics
No. of papers from the Institution in previous years
YearPapers
2022151
20215,389
20205,525
20195,229
20185,054
20174,724