Elements of a Good Feature Story

2019 ◽  
Vol 17 (6) ◽  
pp. 5-5
Keyword(s):  
1956 ◽  
Vol 7 (3) ◽  
pp. 193-220
Author(s):  
D. Williams

SummaryThe mathematical theory of nosewheel shimmy is given, with particular reference to twin nosewheel assemblies. It is shown that a sovereign remedy for shimmy is to make the castor length greater than what is here called the “ creep distance,” which in practice is found to be approximately equal to the tyre radius. Lateral flexibility of the oleo leg is disadvantageous but elastic constraint at the pivot is a good feature. The one necessitates an increased castor for stability while the other allows a smaller castor. It is also shown how, by the use of a compact linkage mechanism, the effective castor length can be made independent of the wheel-leg offset and can have any desired value. Model experiments that confirm the theoretical conclusions are described.


2010 ◽  
Vol 9 ◽  
pp. CIN.S4020 ◽  
Author(s):  
Chen Zhao ◽  
Michael L. Bittner ◽  
Robert S. Chapkin ◽  
Edward R. Dougherty

When confronted with a small sample, feature-selection algorithms often fail to find good feature sets, a problem exacerbated for high-dimensional data and large feature sets. The problem is compounded by the fact that, if one obtains a feature set with a low error estimate, the estimate is unreliable because training-data-based error estimators typically perform poorly on small samples, exhibiting optimistic bias or high variance. One way around the problem is limit the number of features being considered, restrict features sets to sizes such that all feature sets can be examined by exhaustive search, and report a list of the best performing feature sets. If the list is short, then it greatly restricts the possible feature sets to be considered as candidates; however, one can expect the lowest error estimates obtained to be optimistically biased so that there may not be a close-to-optimal feature set on the list. This paper provides a power analysis of this methodology; in particular, it examines the kind of results one should expect to obtain relative to the length of the list and the number of discriminating features among those considered. Two measures are employed. The first is the probability that there is at least one feature set on the list whose true classification error is within some given tolerance of the best feature set and the second is the expected number of feature sets on the list whose true errors are within the given tolerance of the best feature set. These values are plotted as functions of the list length to generate power curves. The results show that, if the number of discriminating features is not too small—that is, the prior biological knowledge is not too poor—then one should expect, with high probability, to find good feature sets. Availability: companion website at http://gsp.tamu.edu/Publications/supplementary/zhao09a/


Sensors ◽  
2021 ◽  
Vol 21 (5) ◽  
pp. 1894
Author(s):  
Jiangzhong Cao ◽  
Yunfei Huang ◽  
Qingyun Dai ◽  
Wing-Kuen Ling

Aiming at the high cost of data labeling and ignoring the internal relevance of features in existing trademark retrieval methods, this paper proposes an unsupervised trademark retrieval method based on attention mechanism. In the proposed method, the instance discrimination framework is adopted and a lightweight attention mechanism is introduced to allocate a more reasonable learning weight to key features. With an unsupervised way, this proposed method can obtain good feature representation of trademarks and improve the performance of trademark retrieval. Extensive comparative experiments on the METU trademark dataset are conducted. The experimental results show that the proposed method is significantly better than traditional trademark retrieval methods and most existing supervised learning methods. The proposed method obtained a smaller value of NAR (Normalized Average Rank) at 0.051, which verifies the effectiveness of the proposed method in trademark retrieval.


1913 ◽  
Vol 10 (2) ◽  
pp. 73-86 ◽  
Author(s):  
A. R. Horwood

This district is bounded on the north by the Coalville and Ashby line, on the west and south by the county boundaries, on the east by the Shackerstone and Market Bosworth lines. In the north are exposures of Coal-measures, Permian breccias, and Bunter, which the Trias in turn rests upon unconformably. The Lower Keuper forms a long tract on the west not more than two miles in breadth, forming a good feature, the sandstones giving rise to scarps, whilst the Red Mail occupies the rest of the district to the east. On the south-west beds of sandstone form marked features, which also give rise to bold escarpments, whilst the Red Marl itself constitutes a uniform plateau with little or no variation in heights. There are few exposures in the marls, which on the extreme east are covered by a mantle of Boulder-clay and sands. The River Sence and the Sence Brook, however, cut down to the lower parts of the Red Marl, and a good deal of alluvium fills the valleys to the south. The altitude over most of this ground rises uniformly above 300 feet, and in some parts to over 400, rarely sinking below 250. A ridge of hills is formed by the Orton Sandstone striking north-west and south-east, and another ridge meets it at right angles from Market Bosworth.


Author(s):  
Chang-Hong Fu ◽  
Ya-Wen Zhao ◽  
Hong-Bin Zhang ◽  
Yui-Lam Chan ◽  
Wan-Chi Siu

Sign in / Sign up

Export Citation Format

Share Document