this paper, we will describe a simple rule-based approach to automated learning of linguistic knowledge. This approach has been shown for a number of tasks to capture information in a clearer and more direct fashion without a compromise in performance. We present a detailed case study of this learning method applied to part of speech tagging

In this paper we describe an unsupervised learning algorithm for automatically training a rule-based part of speech tagger without using a manually tagged corpus. We compare this algorithm to the Baum-Welch algorithm, used for unsupervised training of stochastic taggers. Next, we show a method for combining unsupervised and supervised rule-based training algorithms to create a highly accurate tagger using only a small amount of manually tagged text.

Abstract. This work compares different metaheuristics techniques applied to an important problem in natural language: tagging. Tagging amounts to assigning to each word in a text one of its possible lexical categories (tags) according to the context in which the word is used (thus it is a disambiguation task). Specifically, we have applied a classic genetic algorithm (GA), a CHC algorithm, and a Simulated Annealing (SA). The aim of the work is to determine which one is the most accurate algorithm (GA, CHC or SA), which one is the most appropriate encoding for the problem (integer or binary) and also to study the impact of parallelism on each considered method. The work has been highly simplified by the use of MALLBA, a library of search techniques which provides generic optimization software skeletons able to run in sequential, LAN and WAN environments. Experiments show that the GA with the integer encoding provides the more accurate results. For the CHC algorithm, the best results are obtained with binary coding and a parallel implementation. SA provides less accurate results than any of the evolutionary algorithms. 1