Advancing Machine Comprehension of the Georgian Language
Tea Munjishvili () and
Teona Shugliashvili ()
Additional contact information
Tea Munjishvili: Tbilisi State University
Teona Shugliashvili: Ludwig Maximilian University of Munich
A chapter in Digital Management and Artificial Intelligence, 2025, pp 243-251 from Springer
Abstract:
Abstract The complexity of the Georgian language poses challenges for machines. AI efforts, like Schmidt’s (2019) use of POS tagging with Bi-LSTMs and AI Translate chatbots, aim to teach machines Georgian. Integration of Georgian Language into models like ChatGPT, using NLP techniques and diverse machine learning models, has occurred, but outcomes have limitations. Companies pay money for premium access to ChatGPT, While many companies invest in premium access to ChatGPT, its performance can vary across languages. In the case of Georgian, the model might occasionally lack precision due to limited training data. In this paper we outline our AI strategy to enhance machine understanding of Georgian language. Our unique method, “predictive grammar” tackles the limitations of reliance on statistical patterns. It combines semantic analysis algorithms with beyond-system engineering techniques grounded in the axiom: “Rules for understanding semantics in each problem area can be detected to develop a model of logical-probability semantics.” To determine if a candidate sentence conveys the same meaning as a unique sentence or bag-of-words from the knowledge base, we First, analyze the influence of each word in the candidate sentence on the knowledge base of the unique sentence by calculating multiple information indices. Next, we construct a logical-probability semantic model and establish semantic connections using these influence measures. The meaning of the candidate sentence is derived from these semantic connections. We identify each sentence corresponding to the unique sentence or bag-of-words by understanding its meaning and matching it with the candidate sentences. This process is detailed in Munjishvili et al. (2023). Over 11 years, its application across diverse fields, from digital health to financial analytics, “predictive grammar” method showcases the robustness, with no instances of miscomprehension recorded. Machine Understanding of the Georgian Language has wide applications in digital management, spanning business intelligence, data management, CRM, chatbots, content management, customer behavior research, education, supply chain, and human resources. It’s beneficial for data-driven AI systems, improving decision-making, automating tasks, enhancing customer interactions, and optimizing digital content and resource management.
Keywords: Georgian language; AI-driven comprehension; Predictive grammar methodology; Machine understanding enhancement; Digital management applications (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:
There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.
Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
HTML/Text
Persistent link: https://EconPapers.repec.org/RePEc:spr:prbchp:978-3-031-88052-0_19
Ordering information: This item can be ordered from
http://www.springer.com/9783031880520
DOI: 10.1007/978-3-031-88052-0_19
Access Statistics for this chapter
More chapters in Springer Proceedings in Business and Economics from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().