Unit:
Department of Communication and Media StudiesLibrary of the Faculties of Political Science and Public Administration, Communication and Mass Media Studies, Turkish and Modern Asian Studies, Sociology
Dissertation committee:
Κωνσταντίνος Μουρλάς, Αναπληρωτής Καθηγητής, Τμήμα Επικοινωνίας και Μέσων Μαζικής Ενημέρωσης, ΕΚΠΑ.
Δημήτριος Γκούσκος, Επίκουρος Καθηγητής, Τμήμα Επικοινωνίας και Μέσων Μαζικής Ενημέρωσης, ΕΚΠΑ.
Μιχαήλ Σπουρδαλάκης, Καθηγητής, Τμήμα Πολιτικής Επιστήμης και Δημόσιας Διοίκησης, ΕΚΠΑ.
Ιωάννης Σταυρακάκης, Καθηγητής, Τμήμα Πολιτικών Επιστημών, Αριστοτέλειο Πανεπιστήμιο Θεσσαλονίκης.
Αριστοτέλης Στυλιανού, Αναπληρωτής Καθηγητής, Τμήμα Πολιτικών Επιστημών, Αριστοτέλειο Πανεπιστήμιο Θεσσαλονίκης.
Ανδρέας Γιαννακουλόπουλος, Αναπληρωτής Καθηγητής, Τμήμα Τεχνών Ήχου & Εικόνας, Ιόνιο Πανεπιστήμιο.
Παναγιώτης Σταματόπουλος, Επίκουρος Καθηγητής, Τμήμα Πληροφορικής και Τηλεπικοινωνιών, ΕΚΠΑ.
Original Title:
Εξατομικευμένη πολιτική επικοινωνία στα μέσα κοινωνικής δικτύωσης
Translated title:
Personalized Political Communication on Social Media Networks
Summary:
A personalized Political Communication model is created designed to work in the environment of a Social Media Network as well as an experimental platform for its evaluation is implemented. The purpose is for the political content that circulates on Social Media Networks to meet the needs of the unique user.
Due to their popularity, Social Media Networks have become an important field for the development of public dialogue, and over time they are considered at least equal - if not reinforced - compared to traditional Media. Political exponents are now particularly active within them.
However, as the excessive amount and variety of content circulating on the Internet, and especially on Social Media Networks, creates conditions for information overload, it becomes increasingly difficult for the user to find the content that concerns him. Personalization environments appear as a solution to this problem. These environments contain recommendation schemes that focus on proposing to the user content related to his personality, interests, goals. A prerequisite for the operation of these systems is the composition of the user's profile, which is based on information collected about him. By "knowing" the user adequately, we "know" the content that interests him.
The research uses personalization techniques for political communication to achieve a two-way process:
a) on the one hand, user can find/receive the political information that he is likely to be interested in,
b) on the other hand, political exponents can identify the appropriate audience, expanding their influence.
This approach is considered beneficial for both parties involved: it enhances user satisfaction and strengthens the influence of political exponents.
The effective orientation of the information becomes feasible through the proposed Political Profile Model, composing of parameters that determine the political identity and the potential political behaviour of the individual. The Political Profile allows the identification of the user's political attitude and the political content's ideology as well. Through their profile, users and political texts are measured in terms of common parameters to achieve a matching between them.
Within the implemented experimental platform, text-mining techniques are utilized for the automatic classification of the political content in terms of its political position. Automatic placement of political texts in the political space becomes the precondition for composing the user's political profile since the political content with which the user interacts will compose implicitly and dynamically her/his political profile. The user's political profile will constantly evolve and will be adjusted as long as the personalization system "perceives" any ideological shifts.
In summary, the proposed Personalised Political Communication Model aims to direct the "proper" political content to the "proper" person, that is, to the person expected to find it interesting and to interact with it, e.g. by reading, approving, commenting, republishing or sharing.
A new perception of the distribution of political content on Social Media Networks can be a step in reheating citizens' preoccupation with public affairs. Besides, it can contribute to the development of a more substantial - but mostly interactive - relationship between political exponents and citizens, with more well-informed users and political exponents monitoring and responding to the "pulse" of public opinion. It is also possible to create favourable conditions for user interconnection, through the possibility of developing dialogue groups, consisting of people with common interests and kinship characteristics, which is essential for strengthening consultation and democratic dialogue. Social Media creates such conditions by default; in the present study, an attempt is made to expand these possibilities.
Finally, it is noted that the implementation of personalized political communication involves the risk of developing enclosed communities that, by "recycling" the same ideas, gradually move away from public affairs. To overcome this risk, a new form of personalization is proposed that focuses on the political content that circulates in the Social Medium as a whole and is not limited to the content created or reproduced exclusively by the user's contacts, aiming to broaden his field of vision. In addition to this, a method for presenting opposite views from those of the user is provided.
Main subject category:
Social, Political and Economic sciences
Keywords:
Political communication, social media, social media networks, personalisation, personalization, text-mining, text mining, text analysis, python
Number of references:
143
