Samarjeet Borah

Object of the study is the transformation of models of citizen–state interaction in the context of public sector digital transformation, particularly within the Government-to-Citizen (G2C) paradigm in the European Union. Problem statement. The digital transformation of the public sector leads to significant changes in the ways citizens interact with the state. In European Union countries, these processes are increasingly associated with the transition from traditional e-government models to integrated human-centric digital governance ecosystems. Unresolved issues. Despite the active implementation of digital governance tools, the evolution of citizen–state interaction models and the systemic development of human-centric G2C ecosystems remain insufficiently conceptualized, especially in the context of their adaptation to Ukrainian realities. The purpose of the study is to analyze the evolution of citizen–state interaction models in the context of digital transformation, to examine the development of G2C ecosystems in the European Union, and to identify opportunities for applying this experience in Ukraine. Research methodology and main results. The study employs methods of comparative analysis, generalization of European digital governance policies, and conceptual modeling of G2C ecosystem development. The article systematizes the evolution of citizen–state interaction models –from e-government to human-centric digital governance—and provides a preliminary mapping of G2C ecosystems across EU countries. Key development trends are identified, including the expansion of proactive public services, the emergence of platform-based governance models, the use of artificial intelligence for the personalization of public services, and the development of interoperable digital infrastructures. Particular attention is given to the analysis of key challenges, such as the fragmentation of government information systems, uneven development of electronic identification mechanisms, the digital divide among population groups, and regulatory challenges related to the use of artificial intelligence in the public sector. Scientific novelty lies in the conceptualization of the evolution of citizen–state interaction models in the context of public sector digital transformation and in the systematization of European experience in the development of human-centric G2C ecosystems. Conclusions. The findings indicate that the transition to human-centric digital governance requires not only technological modernization but also deep institutional transformation, integration of public services, and adherence to the principles of interoperability and proactivity. The results define practical directions for applying European experience in Ukraine, particularly in the integration of digital public services, the development of proactive services, and the enhancement of interoperability among government information systems.

Social media is a platform for individuals to connect, share, and create information. Social bots produce automated content and interact with humans; in the process, they learn and mimic humans’ behaviour. This research study addresses the challenge of identifying social media bots (SMB) that can rapidly disseminate information or misinformation on platforms like Twitter. It contributes to the field by reviewing literature to define bot behaviours and exploring advanced machine learning classifiers for effective bot detection using account-level data. The study employed Spearman's rank correlation coefficient to select relevant features for SMB classification, then trained six different machine learning models: decision tree (DT), random forest (RF), logistic regression (LR), support vector machine (SVM), and k-nearest neighbour (KNN). To further improve accuracy, a classifier stacking technique was applied. Key findings revealed that while individual classifiers performed variably, with RF leading at 89% accuracy, the stacked classifier approach outperformed all single-classifier methods with an impressive 90% accuracy rate. The results underscore the potential of combining multiple classifiers to enhance the precision of social media bot detection efforts.

Significant issues in tokenization and stemming in natural language processing are addressed in this work, focusing primarily on the Limbu language. Two essential preprocessing procedures that work to normalize words by condensing them into compact content and their origin are stemming and tokenization. We introduce a novel stemmer for the Limbu language, which achieves 93.5% accuracy rates using the given word bank. The stemmer is preceded by tokenization with a 100% accuracy rate on the current available corpus. The Limbu language presents peculiar issues due to its constrained computational resources and composite nature. This work is a first step toward overcoming the lack of necessary resources for effective stemming in Limbu language computational work. Advanced algorithms with morphologically structured rules especially chosen for the Limbu language are used in stemming techniques to increase accuracy and speed. The discoveries significantly improve natural language processing activities, providing strong tools for search engines, sentiment analysis, and automatic translation systems, as well as marking a first step in computational development for the Limbu language.