Machine Learning and Neural Networks Tools to Address Noisy Data Issues
Machine Learning and Neural Networks Tools to Address Noisy Data Issues
Author(s): Maria Teresa Artese, Isabella GagliardiSubject(s): Library and Information Science, Information Architecture, Library operations and management, Electronic information storage and retrieval, Education and training
Published by: Институт по математика и информатика - Българска академия на науките
Keywords: Digital Library; Unsupervised Tools; Noisy Data; Tags; Content Based Retrieval.
Summary/Abstract: In this paper, we present tools for addressing noisy keyword issues in digital libraries. Two tasks, language detection and misspelling detection and correction, are addressed using both machine learning and deep learning techniques. To train and validate the models, different datasets were used/created/scraped. Encouraging preliminary results are presented and discussed.
Journal: Digital Presentation and Preservation of Cultural and Scientific Heritage
- Issue Year: 2021
- Issue No: XI
- Page Range: 89-98
- Page Count: 10
- Language: English