Eleni Galiotou

A word spotting framework for historical machine-printed documents

In this paper, we propose a word spotting framework for accessing the content of historical machine-printed documents without the use of an optical character recognition engine. A preprocessing step is performed in order to improve the quality of the document images, while word segmentation is accomplished with the use of two complementary segmentation methodologies. In the proposed methodology, synthetic word images are created from keywords, and these images are compared to all the words in the digitized documents. A user feedback process is used in order to refine the search procedure. The methodology has been evaluated in early Modern Greek documents printed during the seventeenth and eighteenth century. In order to improve the efficiency of accessing and search, natural language processing techniques have been addressed that comprise a morphological generator that enables searching in documents using only a base word-form for locating all the corresponding inflected word-forms and a synonym dictionary that further facilitates access to the semantic context of documents.

Accessing the content of Greek historical documents

In this paper, we propose an alternative method for accessing the content of Greek historical documents printed during the 17th and 18th centuries by searching words directly in digitized documents based on word spotting, without the use of an optical character recognition engine. We describe a methodology according to which synthetic word images are created from keywords. These images are compared to all the words in the digitized documents while user feedback is used in order to refine the search procedure. In order to improve the efficiency of accessing and searching, we have used natural language processing techniques that comprise (i) a morphological generator for early Modern Greek which provides the users with the ability to search documents using only a word stem and locate all the corresponding inflected word forms and (ii) a synonym dictionary which facilitates access to the semantic context of documents and enriches the results of the search process.

Using neural networks and SVMs for automatic medical diagnosis: A comprehensive review

In this paper we make a comprehensive review regarding the use of neural networks in automated medical diagnosis, with a special emphasis in Support Vector Machines (SVMs), which are specialized types of neural functions. Through the study, we see that, in many cases, symptoms and diseases can be efficiently predicted by neural systems, while SVMs are increasingly used in medical diagnosis due to their accurate classification characteristics.

Structuring a Multimedia Tri-Dialectal Dictionary

This paper deals with the problem of designing and implementing a multimedia electronic dictionary of three Greek dialects in Asia Minor (Pontic, Cappadocian, Aivaliot). At first, we present the linguistic and lexicographic approach adopted, as well as the principles for designing the macro/microstructure of the dictionary. Next, we present and describe the conceptual model of the tri-dialectal dictionary. Finally, we discuss the relational schema and implementation issues. Although, in its current state, the system hosts three dialects, it is designed so as to be able to incorporate other dialectal dictionaries in the future.

Towards a Unified Exploitation of Electronic Dialectal Corpora: Problems and Perspectives

In this paper, we deal with the problem of storing and retrieving dialectal data in a unified framework. In particular, we discuss issues concerning the design and implementation of a multimedia database which will contain written and oral data from three Greek dialects in Asia Minor. At first, we describe the overall architecture of a system aiming at providing the user with the possibility to store audio recordings, text transcripts, and other annotations. Then we discuss the possibilities and limitations of a retrieval module aiming at combining different linguistic levels for a unified exploitation of oral and written corpora.

On the effect of stemming algorithms on extractive summarization: a case study

In this paper, we discuss the efficiency of stemming algorithms and their contribution to the improvement of shallow summarization methods. We describe a benchmarking experiment on the use of two stemming algorithms and two different sets of stopwords in combination with two approaches to extractive summarization in Greek texts. The results of our experimentation show the limits of extractive summarization and stemming algorithms.

A Workbench for Extractive Summarizing Methods

We present a software workbench for testing available/well known methods for text summarization. The software was designed and implemented in order to incorporate alternative proposed methodologies of extractive summarization. Therefore, the provided methods do not aim at understanding and condensing the meanings of an amount of original sentences into a smaller number of sentences not occurring in the original text. It simply extracts a subset of the original sentences which are the most (promising as being) relevant for expressing the meaning of the text. Our purpose is twofold: a) to provide a utility for benchmarking of well known summarization methods to researchers or/and professionals who are interested in implementing summarization products, b) to perform our own evaluations of the well known extractive summarization methods.

Towards a computational processing of oral dialectal data

In this paper, we discuss issues concerning the computational processing of oral data in a unified framework for the exploitation of oral and written dialectal corpora. We describe the analysis and design of a multimedia database and software for storing and retrieving dialectal data, focusing on the subsystem of oral resources from three Greek dialects in Asia Minor. We discuss problems concerning the archiving and exploitation of such an oral corpus and we propose solutions.

Aligning Greek-English parallel texts

In this paper, we discuss issues concerning the alignment of parallel texts written in languages with different alphabets based on an experiment of aligning texts from the proceedings of the European Parliament in Greek and English. First, we describe our implementation of the k-vec algorithm and its application to the bilingual corpus. Then the output of the algorithm is used as a starting point for an alignment procedure at a sentence level which also takes into account mark-ups of meta-information. The results of the implementation are compared to those of the application of the Church and Gale alignment algorithm on the Europarl corpus. The conclusions of this comparison can give useful insights as for the efficiency of alignment algorithms when applied to the particular bilingual corpus.

A Preliminary Investigation into the Automatic EuroVoc Indexing of Greek Documents

In this paper, we present an automatic indexing experiment of greek documents. In particular, we describe an attempt to use JEX, the JRC-developed indexing tool, in order to assign EuroVoc descriptors to a collection of Greek open data. We discuss the results and limitations of this approach and we propose solutions which take into account the particularities of the Greek language.