Page 3 - Abstract

1. The author proposes a definition of public opinion, combining the
developments within the framework of various existing (delete) theoretical and
methodological approaches (empirical, psychological, sociological and
psychological, activity-based): public opinion is a socio-psychological
phenomenon with a systemic nature, which can be formed spontaneously or
deliberately, reflects opinions on current issues that exist in society or a certain
stratum of it through the actions and/or statements of individuals as separate
representatives of society and at the same time has an impact on society as a whole
and on its individual representatives or their groups and, in connection with this,
has a transformational potential.
2. The specifics and nature of public opinion formation are considered
through the prism of evolutionary changes in public policy models. The stages
and problems of the formation of the public sphere of Kazakhstan are studied.
The analysis revealed that along with positive changes, there is a predominance
of administration on the part of the state and a lack of control on the part of civil
society. This can be overcome through the implementation of the "listening state"
model, which involves the constant improvement of public opinion monitoring
tools and increased responsiveness to public inquiries with the active use of
modern technologies, including artificial intelligence.
3. The effectiveness of modern political technologies for shaping public
opinion in the Republic of Kazakhstan depends on their adaptation to specific
conditions of use. The activities of government agencies and the strengthening of
dialogue with the population are currently directly linked to the development of
information technologies and their integration into everyday work;
4. Digital technologies are revolutionizing traditional tools for managing
public opinion, and managing the masses with the help of artificial intelligence
has become a global trend. Based on the analysis of international experience,
successful practices and technologies used in various countries have been
identified, and proposals for their implementation have been developed taking
into account domestic specifics.
5. As specific recommendations, an applied model for artificial intelligence
Data Science was developed in the Phyton programming language in Jupyter
Notebook, designed to predict the evaluation of comments on the selected dataset.

Optimal hyperparameters were determined and model metrics were built.
6. Recommendations have been developed for working with public opinion,
as well as for regulating and protecting digital privacy for implementation in the
areas of public administration and public policy.
Compliance with scientific development directions or state programs.
Scientific research complies with the Concept of Artificial Intelligence
Development for 2024-2029 (Resolution of the Government of the Republic of
Kazakhstan dated July 24, 2024 No. 592), the Concept of Cybersecurity ("Cyber
Shield of Kazakhstan") (Resolution of the Government of the Republic of
Kazakhstan dated June 30, 2017 No. 407), the Concept of Civil Society