Symbol recognition with a new autonomously evolving classifier autoclass

Abstract

A new algorithm for symbol recognition is proposed in this paper. It is based on the AutoClass classifier [1], [2], which itself is a version of the evolving fuzzy rule-based classifier eClass [3] in which AnYa[1] type of fuzzy rules and data density are used. In this classifier, symbol recognition task is divided into two stages: feature extraction, and recognition based on feature vector. This approach gives flexibility, allowing us to use various feature sets for one classifier. The feature extraction is performed by means of gist image descriptors[4] augmented by several additional features. In this method, we map the symbol images into the feature space, and then we apply AutoClass classifier in order to recognise them. Unlike many of the state-ofthe-art algorithms, the proposed algorithm is evolving, i.e. it has a capability of incremental learning as well as ability to change its structure during the training phase. The classifier update is performed sample by sample, and we should not memorize the training set to provide recognition or further update. It gives a possibility to adapt the classifier to the broadening and changing data sets, which is especially useful for large scale systems improvement during exploitation. More, the classifier is computationally cheap, and it has shown stable recognition time during the increase of training data set size that is extremely important for online applications.

Evolving clustering, classification and regression with TEDA

“…As it was stated in the previous work on TEDAClass [ 18], and contrary to the formulations of eClass [ 16] and AulaClass [ 17], we do not define any global statistical characteristics over the data.…”

“…Nowadays, a lot of fuzzy systems are serving to solve classification, regression and clustering problems. Recently introduced algorithms include DE C [ 24], AutoClass [ 17], eClass [ 16], FLEXF I SClass [ 25]. Some of these algorithms are included in comparison in the experimental section.…”

“…I t is evolving, as it is capable of adding and removing the clusters during the data stream processing that is significantly use ful for various applications. Basically, the fuzzy systems follow the same paradigm as previously proposed eClass [ 16] and AutoClass [ 17] frameworks.…”

Parsimonious random vector functional link network for data streams

“…This concept was defined in AnYa [1], [3] and was patented in [2]. This concept is also the underlying component of AutoClass and TEDA-Class [4]. All of which come from Angelov's sound work of RDE [3].…”

“…It is concluded that random regions should be carefully selected and should be chosen close enough to true operating regions of a system being modelled. The MATLAB codes of pRVFLN is made publicly available in 4) to help further study.…”