6. INTERNATIONAL CONGRESS ON CONTEMPORARY SCIENTIFIC RESEARCH
PROBLEMS AND SOLUTIONS IN SENTIMENT ANALYSIS OF TEXTS FROM DIFFERENT HISTORICAL/LITERARY ERAS
This study addresses the main challenges and proposed solutions related to sentiment analysis of historical and literary texts. Key issues include the physical inaccessibility of texts, the lack of digital versions, difficulties in automatically removing in-text dictionaries, and the inadequacies of OCR programs. Particularly in Ottoman texts, identifying word roots and the absence of period-specific sentiment dictionaries complicate the analysis. Proposed solutions involve the use of AI-based translation tools, digitization efforts, and deep learning methods for word root detection.
