Adaptive Committees of Neural Classifiers
نویسنده
چکیده
It is obvious that combination of several classifiers might improve overall classification performance. In this paper, on the contrary to the ordinary approach of utilising all neural networks available to make the committee decision, we propose to create adaptive committees, which are specific for each input data point. A prediction neural network is used to identify classifiers to be fused for making a committee decision about the given input data. The proposed technique is tested in three aggregation schemes and the effectiveness of the approach is demonstrated on the three real data sets.
منابع مشابه
Validation of Voting Committees
This article contains a method to bound the test errors of voting committees with members chosen from a pool of trained classifiers. There are so many prospective committees that validating them directly does not achieve useful error bounds. Because there are fewer classifiers than prospective committees, it is better to validate the classifiers individually than use linear programming to infer...
متن کاملCombining Neural Network Voting Classifiers and Error Correcting Output Codes
Papers published in this report series are preliminary versions of journal articles and not for quotations. Abstract We show that error correcting output codes (ECOC) can further improve the eeects of error dependent adaptive resampling methods such as arc-lh. In traditional one-inn coding, the distance between two binary class labels is rather small, whereas ECOC are chosen to maximize this di...
متن کاملAdaptive Committees of Feature-Specific Classifiers for Image Classification
We present a system for image classification based on an adaptive committee of five classifiers, each specialized on classifying images based on a single MPEG-7 feature. We test four different ways to set up such a committee, and obtain important accuracy improvements with respect to a baseline in which a single classifier, working an all five features at the same time, is employed.
متن کاملAdaptive combinations of classifiers with application to on-line handwritten character recognition
Classifier combining is an effective way of improving classification performance. User adaptation is clearly another valid approach for improving performance in a user-dependent system, and even though adaptation is usually performed on the classifier level, also adaptive committees can be very effective. Adaptive committees have the distinct ability of performing adaptation without detailed kn...
متن کاملAn Adaptive Thr esholding Multiple Classifiers System for Remote Sensing Image Classification
A multiple classifiers system which adopts an effective weighting policy to combine the output of several classifiers, generally leads to a better performance in image classification. The two most commonly used weighting policies are Bagging and Boosting algorithms. However, their performance is limited by high levels of ambiguity among classes. To overcome this difficulty, an adaptive threshol...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2008