Repozitoriul Instituțional al Universității de Stat din Moldova (RI USM)

Institutional Repository of Moldova State University (IR MSU)

Repozitoriul Instituțional al Universității de Stat din Moldova reprezintă arhiva digitală cu acces deschis a rezultatelor cercetărilor științifice și științifico-didactice efectuate în cadrul universității. Conținutul arhivei este multidisciplinar și include lucrări din domeniile științelor exacte și socio-umanistice.
The Institutional Repository of the State University of Moldova is an open access digital archive of the results of scientific and scientific-didactic research carried out within the university. The archive's content is multidisciplinary and includes works from the fields of exact sciences, social sciences and humanities.

Communities in DSpace

Select a community to browse its collections.

Now showing 1 - 5 of 17

Recent Submissions

Item

Digitization of Moldovan folklore texts using the HeDy platform [Articol]

(2024) Colesnicova, Vlada; Caftanatov, Olesea; Cojocaru, Svetlana; Colesnicov, Alexandru; Malahov, Ludmila

This paper discusses the use of the digitization platform HeDy to produce electronic resources of Moldavian proverbs and sayings. It will promote the development and enrichment of annotated corpora and other electronic resources of philological data on regional folklore and nonstandard texts from Moldova. The results should be uniform in terms of formats and standards, in particular, to be incorporated into the Universal Dependency (UD) repository.

Item

Some Applications of the Depth-First Search Algorithm for Undirected Graphs [Articol]

(2024) Ciubotaru, Constantin

Applications of the deep first search algorithm (DFS) have been developed that allow at a single traversal of the graph: a) to check the connectivity/biconnectivity of the graph, b) to build the spanning tree, c) to highlight the cut vertices, d) to calculate the biconnected components, e) to build the fundamental set of cycles.The proposed algorithms were programmed and tested.

Item

Comparison of Large Language Models and Traditional Neural Networks in Optical Character Recognition for Old Alphabets [Articol]

(2024) Cerescu, Marius; Bumbu, Tudor

This study compares large language models (LLMs) and traditional neural networks (TNNs) in Optical Character Recognition (OCR) for historical alphabets. While deep learning has advanced OCR technology, recognizing old scripts remains challenging due to their complexity. LLMs, with vision capabilities, offer a novel approach by integrating visual and linguistic understanding. This research evaluates the accuracy and robustness of both models on a dataset of an ancient alphabet, highlighting the potential of LLMs to improve OCR in historical linguistics and digital preservation. The findings provide valuable insights for applying modern AI to the preservation of historical texts.

Item

The Virtual GPT Assistant: Emulating the Teaching Style of a Real Professor [Articol]

(2024) Caftanatov, Olesea; Parahonco, Alexandr

This paper discusses the development of the Virtual GPT Assistant, an AI-driven educational tool designed to emulate the teaching style of a real professor. Utilizing OpenAI’s GPT-4 architecture, the assistant aims to provide personalized, contextually relevant support to students, contributing to the evolving landscape of educational technology marked by Intelligent Tutoring Systems.

Item

Our Approach to Digitizing Handwritten Mathematical Text in Cyrillic Containing Formulas and Drawings [Articol]

(2024) Caftanatov, Olesea; Demidova, Valentina; Verlan, Tatiana

This paper describes the steps through which the authors passed during the process of digitization of manually written mathematical texts with formulas and figures. Some diffculties met are also discussed. Our project highlighted the challenges associated with working with handwritten, non-homogeneous texts stored on outdated records, but it also demonstrated the effectiveness of combining modern technology with traditional manual methods.