Download presentation
Published byArron Cummings Modified over 9 years ago
1
ENN: Extended Nearest Neighbor Method for Pattern Recognition
This lecture notes is based on the following paper: B. Tang and H. He, "ENN: Extended Nearest Neighbor Method for Pattern Recognition," IEEE Computational Intelligence Magazine, vol.10, no.3, pp , Aug. 2015 ENN: Extended Nearest Neighbor Method for Pattern Recognition Prof. Haibo He Electrical Engineering University of Rhode Island, Kingston, RI 02881 Computational Intelligence and Self-Adaptive Systems (CISA) Laboratory
2
Extended Nearest Neighbor for Pattern Recognition
Limitations of K-Nearest Neighbors (KNN) “Two-way communication”: Extended Nearest Neighbors (ENN) Experimental Analysis Conclusion
3
Pattern Recognition Parametric Classifier
Class-wise density estimation, including naive Bayes, mixture Gaussian, etc. Non-Parametric Classifier Nearest Neighbors Neural Network Support Vector Machine Nonparametric nature Easy implementation Powerfulness Robustness Consistency
4
Limitations of traditional KNN
Scale-Sensitive Problem: The class 1 samples dominate their near neighborhood with higher density (i.e., more concentrated distribution). The class 2 samples are distributed in regions with lower density (i.e., more spread out distribution). Those class 2 samples which are close to the region of class 1 may be easily misclassified.
5
ENN: A New Approach Define generalized class-wise statistic for each class: Si denotes the samples in class i, and NNr(x, S) denotes the r-th nearest neighbor of x in S. Ti measures the coherence of data from the same class. 0 ≤ Ti ≤ 1 with Ti = 1 when all the nearest neighbors of class i data are also from the same class i, and with Ti = 0 when all the nearest neighbors are from other classes.
6
Intra-class coherence:
Given an unknown sample Z to be classified, we iteratively assign it to class 1 and class 2, respectively, to obtain two new generalized class-wise statistics Tij, where j=1,2. Then, the sample Z is classified according to: ENN Classification Rule: Maximum Gain of Intra-class Coherence. For N-class classification:
7
To avoid the recalculation of generalized class-wise statistics in testing stage, an Equivalent Version of ENN is proposed:
8
The equivalent version has the same result as the original one, but avoids the recalculation of Tij
9
How this simple rule works better than KNN
The ENN method makes a prediction in a “two-way communication” style: it considers not only who are the nearest neighbors of the test sample, but also who consider the test sample as their nearest neighbors.
10
Experimental Results and Analysis
11
Sampling methods Synthetic Data Set:
A 3-dimensional Gaussian data with 3 classes: Considering the following four models, their error rates are: Model 2 Class 1 Class 2 Class 3 KNN ENN k = 3 32 31.9 39.3 34.4 31.4 30.5 k = 5 31.2 29.7 40.5 33.7 28.6 26.7 k = 7 28.5 28.3 40.8 33.6 25 24.3 Model 3 33.2 31 27 26.8 38.8 30.3 27.3 24 23.2 40.2 33.5 25.1 20.8 40.6 33
12
Sampling methods Real-life Data Sets:
MNIST Handwritten Digit Recognition Data Examples
13
Sampling methods Real-life Data Sets:
20 data sets from UCI Machine Learning Repository t-test shows that ENN can significantly improve the classification performance in 17 out of 20 datasets, in comparison with KNN.
14
ENN Summary: Three versions of ENN
B. Tang and H. He, "ENN: Extended Nearest Neighbor Method for Pattern Recognition," IEEE Computational Intelligence Magazine, vol.10, no.3, pp , Aug. 2015
15
ENN.V1 Summary: Three versions of ENN
B. Tang and H. He, "ENN: Extended Nearest Neighbor Method for Pattern Recognition," IEEE Computational Intelligence Magazine, vol.10, no.3, pp , Aug. 2015
16
ENN.V2 Summary: Three versions of ENN
B. Tang and H. He, "ENN: Extended Nearest Neighbor Method for Pattern Recognition," IEEE Computational Intelligence Magazine, vol.10, no.3, pp , Aug. 2015
17
Online Resources Summary: Three versions of ENN
Supplementary materials and Matlab source code implementation available at: B. Tang and H. He, "ENN: Extended Nearest Neighbor Method for Pattern Recognition," IEEE Computational Intelligence Magazine, vol.10, no.3, pp , Aug. 2015
18
Conclusion A new ENN classification methodology based on the maximum gain of intra-class coherence. “Two-way communication”: ENN considers not only who are the nearest neighbors of the test sample, but also who consider the test sample as their nearest neighbors. Important and useful for many other machine learning and data mining problems, such as density estimation, clustering, regression, among others. B. Tang and H. He, "ENN: Extended Nearest Neighbor Method for Pattern Recognition," IEEE Computational Intelligence Magazine, vol.10, no.3, pp , Aug. 2015
Similar presentations
© 2024 SlidePlayer.com. Inc.
All rights reserved.