Search

In this article, the application of Ant-Colony Optimization (ACO) to a morphological segmentation task is described, where the aim is to analyse a set of words into their constituent stem and ending. A number of criteria for determining the optimal segmentation are evaluated comparatively while at the same time investigating more comprehensively the effectiveness of the ACO system in defining appropriate values for system parameters. Owing to the characteristics of the task at hand, particular emphasis is placed on studying the ACO process for learning sessions of a limited duration. Morphological segmentation becomes hardest in highly inflectional languages, where each stem is associated with a large number of distinct endings. Consequently, the present article investigates morphological segmentation of words from a highly inflectional language, specifically Ancient Greek, by combining pattern-recognition principles with limited linguistic knowledge. To weigh these sources of knowledge, a set of weights is used as a set of system parameters, to be optimized via ACO. ACO-based experimental results are shown to be of a higher quality than those achieved by manual optimisation or ‘randomised generate and test’ methods. This illustrates the applicability of the ACO-based approach to the morphological segmentation task. [ABSTRACT FROM AUTHOR]

Given a text or collection of texts involving unconstrained language, a basic task in a multitude of applications is the identification of stems and endings for each word form, which is termed morphological analysis. In this paper, the use of an ant colony optimization (ACO) metaheuristic is proposed for a linguistic task that involves the automated morphological segmentation of Ancient Greek word forms into stem and ending. The task of morphological analysis is essential for implementing text-processing applications such as semantic analysis and information retrieval. The difficulty of the morphological analysis task differs depending on the language chosen, being hardest in the case of highly-inflectional languages, where each stem may be associated with a large number of different endings. In this paper, focus is placed on the morphological analysis of Ancient Greek, which has been shown to be a particularly hard task. To perform this task, a system for the automated morphological processing has been proposed, which implements the morphological analysis of words by coupling an iterative pattern-recognition algorithm with a modest amount of linguistic knowledge, expressed via a set of interactions associated with weights. In an earlier version of the system, these weights were determined by combining the input from specialized scientists with a lengthy manual optimization process. In this paper, the ACO metaheuristic is applied to the task of defining near-optimal system weights using an automated process based on a set of training data. The experiments performed indicate that the segmentation quality achieved by ACO is equivalent to or in several cases substantially higher than that achieved using manually optimized weights. [ABSTRACT FROM AUTHOR]

A data analysis task is described, which is focused on the clustering of high-dimensional meteorological data collected long term (more than 43 years) at 128 weather stations in Greece. The proposed hybrid method combines (a) the assignment of the stations to two-dimensional grids of nodes via self-organizing maps (SOMs) of various sizes and (b) statistical clustering of the SOM nodes. The areas resulting from clustering have well-defined meteorological profiles; they are also described by distinct combinations of morphological and geographical characteristics, indicating that morphology and geographical location largely affect the meteorological measurements. The most salient data parameters per area as well as over the entire map are determined, whereby the parameters and parameter ranges that shape the various meteorological profiles are exposed. The classification of stations with missing and noise-contaminated meteorological measurements into their expected areas demonstrates the prediction capability and robustness of the proposed hybrid method. © 2008 Wiley Periodicals, Inc. [ABSTRACT FROM AUTHOR]

This article reports on experiments performed with a large corpus, aiming at separating texts according to the author style. The study initially focusses on whether the classification accuracy regarding the author identity may be improved, if the text topic is known in advance. The experimental results indicate that this kind of information contributes to more accurate author recognition. Furthermore, as the diversity of a topic set increases, the classification accuracy is reduced. In general, the experimental results indicate that taking into account knowledge regarding the text topic can lead to the construction of specialized models for each author with higher classification accuracy. For example, by focussing on a specific topic, the accuracy with which the author identity is determined increases, the exact amount depending on the specific topic. This also applies when the topic of the text is more broadly determined, as a set of topic categories. In an associated task, the most salient parameters within an 85-parameter vector are studied, for a number of subsets of the corpus, where each subset contains speeches from a single topic. These studies indicate that the salient parameters are the same for the different subsets. Two fixed data vectors have been defined, using 16 and 25 parameters, respectively. The classification accuracy obtained, even with the smallest data vector, is only 5% less than with the complete vector. This indicates that the parameters retained in the reduced vectors bear a large amount of discriminatory information and suffice for an accurate classification of the corpus. [ABSTRACT FROM AUTHOR]

The article investigates the effectiveness of the self-organizing map (SOM) in categorizing texts according to the stylistic differences among authors. This task is of great significance for information retrieval applications using huge databases of documents. The results of statistical methods portray the clustering capability of isolated parameter groups and the potential of SOM in managing high-dimensional data vectors.

This article describes a method for discriminating among authors within a given register of Modern Greek. The focus here is to determine to what extent the stylistic differences among authors can be detected with a high degree of accuracy for a set of texts belonging to a well-defined register. To that end, the chosen register is characterized by a well-defined sub-language, from which a corpus of more than 1,000 documents has been created. To discriminate the texts according to author style, a series of experiments have been performed using statistical techniques. Each text has been represented by a vector covering several linguistic aspects, in an effort to determine the most effective style markers. The experimental results indicate that the proposed approach can successfully separate the author styles for a given register. An extensive study of the effectiveness of the different variable categories has been performed. For instance, diglossia information on its own is not sufficient for author discrimination. Instead, a systematic evaluation process indicates that part-of-speech, structural and algorithmically derived lemma-frequency variables are the most important style markers, their use leading to an author discrimination accuracy exceeding 90%. [ABSTRACT FROM AUTHOR]

This article describes a method for discriminating among registers of Modern Greek and among authors within a given register. Two issues have been investigated: (a) whether register discrimination can successfully exploit linguistic information reflecting the evolution of a language (such as diglossia features of the Modern Greek language) and (b) what kind of linguistic information and which statistical techniques may be applied to author discrimination within one register. Using clustering techniques and variables reflecting the diglossia situation, we have successfully discriminated registers in Modern Greek. However, diglossia information on its own has not been shown sufficient for author discrimination within one register. Instead, other linguistic features, including PoS distribution and discourse tendencies, have been combined with methods such as discriminant analysis in order to obtain a high degree of accuracy. [ABSTRACT FROM AUTHOR]

We report on the application of the Self-Organizing Map (SOM) classification method to the task of categorizing texts according to their register and the style of their author. The SOM has been selected as its performance in various data-mining applications has been found to be highly successful. Here, the method is evaluated against the task of clustering textual data which are corpora of texts written in the Greek language; the parameters used depict linguistically important structural properties of the texts. The experiments reported indicate that the SOM results are equivalent to those generated by statistical methods. [ABSTRACT FROM AUTHOR]

Presents information on a study that examined the scanning n-tuple technique in pattern recognition tasks. Decision margin and classification confidence of n-tuple systems; Application of the class-spitting algorithm to the ESSEX and NIST data set; Methods for improving the scanning n-tuple classification accuracy.