In this paper, an adaptive on-line recognizer is described. The recognizer is based on k nearest neighbor rule. It is used for recognizing isolated alphanumeric characters including both the upper and lower case versions of the letters and some Scandinavian diacriticals. The recognition system has six different dissimilarity measures which all are based on dynamic time warping (DTW), and the best one of these was chosen. The main focus of the work is on on-line adaptation. The adaptation is carried out by modifying the prototype set of the classier according to its recognition performance and the user's writing style. These modications include: 1) adding new prototypes, 2) inactivating confusing prototypes, and 3) reshaping existing prototypes. The reshaping algorithm is based on Learning Vector Quantization (LVQ). Four different adaptation strategies, according to which the modications of the prototype set are performed, have been studied both off- and on-line. The purpose of the adap...

This work describes a prototype-based online handwritten character recognition system and a two-phase recognition scheme aimed to speed up the recognition. In the first phase, the prototype set is pruned and ordered on the basis of preclassification performed with heavily down-sampled characters and prototypes. In the second phase, the final classification is performed without down-sampling by using the reduced set of prototypes. Two down-sampling methods, a linear and nonlinear one, have been analyzed to see their properties regarding the recognition time and accuracy. 1.

this paper. The strategies are based on first

We have developed an adaptive handwriting recognizer for isolated Latin characters in which the adaptive behavior is based on the Dynamically Expanding Context (DEC) algorithm. In our current system, the outputs of a set of static classifiers are combined in a committee machine, whose rules are adapted. Every misclassified character gives rise to adding a new DEC rule to the rule set of the committee. When the existing rules fail to produce correct recognition output, more and more context information is utilized in forming the new DEC rules. Not only the first-ranking outputs from the member classifiers but also the secondranking ones can be taken into account when forming the DEC rules. In the experiments described in this paper, various options in the implementation of the DEC committee classifier are evaluated. The results of the experiments show that the system is capable of fast adaptation to the user's handwriting and leads to lowered recognition error rates.

Methods for controlling the adaptation process of an on-line handwritten character recognizer are studied. The classifier is based on the k-nearest neighbor rule and it is adapted to a new writing style by adding new prototypes, inactivating confusing prototypes, and reshaping existing prototypes in a self-supervised fashion. The dissimilarity measure used for the comparison of characters is a nonlinear curve matching method based on dynamic time warping algorithm. Time needed for the evaluation of the dissimilarity measure for a single character depends linearly on the size of the prototype set. The purpose of the control methods is to increase the classifier's tolerance to malformed or mislabeled learning samples and to limit the growth of the prototype set. The control methods either set an upper limit for the number of prototypes per class or switch the adaptation of a particular character class on or off depending on the earlier performance of the classifier.

We have considered problems involved in the self-supervised learning process of an online handwriting recognition system. Our system is able to recognize isolated characters by comparing them to prototype characters with a method based on the Dynamic Time Warping algorithm. The recognition system is adapted by adding new prototypes, inactivating confusing or erroneous ones, and reshaping existing prototypes with a method based on the Learning Vector Quantization. We have analyzed the sources of erroneous learning samples and studied the influence of such samples on the performance of the recognizer via simulations. In these simulations, two adaptation strategies combined with four methods for inactivating prototypes were applied. The results of the simulations showed that the adaptation strategies are able to improve the system's recognition rate and the prototype inactivation methods do reduce the harmful effects of erroneous learning samples.

There are two main approaches... In the experiments of this paper the feasibility of using an adaptive committee classifier is explored and tested with on-line character recognition. The results clearly show that the use of adaptive committees can improve on the recognition results, both in comparison to the individual member classifiers and the non-adaptive reference committee.

The purpose of this paper is to summarize our work on adaptive on-line recognition methods for handwritten characters. Reports on the work have been published in various conference proceedings and book chapters. As each publication covers only some specific part of our work, it is hard to see the whole picture and get a good overview of the whole work. Instead of trying to explain in detail all the techniques and experiments, we compare them with each other and give more general results. By adaptation we mean that the system is able to learn new writing styles and thus improve its performance. We have had two different approaches to the adaptation: experiments have been carried out with both individually adaptive classifiers and adaptive committees of static classifiers. The main techniques applied in our work include the k-Nearest Neighbor and the Local Subspace Classification rules, Dynamic Time Warping and Levenshtein distances, Learning Vector Quantization, and Dynamic...