In the research prepared by Bora Candan and Onur Yıldırım, students of Kabataş Boys' High School, with the support of artificial intelligence and published in an international peer-reviewed magazine, elements of violence against women were detected in approximately one-third of the most re-watched scenes of Turkish television series.
The study titled "Sociological Analysis of Violence Against Women in Turkish Society (With the Help of Artificial Intelligence)", prepared under the guidance of high school project coordinator and philosophy teacher Özgür Aydemir, was published in the International Journal of Social Science Research in June 2026.
The research was awarded a place among nearly a thousand projects at the 10th Üsküdar University Science and Ideas Festival.
In the research, which brings together sociology, statistics and artificial intelligence, the representation of violence in television series was examined with artificial intelligence-supported media analysis, while teachers' awareness of violence against women was evaluated by survey method.
In the media analysis section of the research, popular TV series scenes were determined by using YouTube's data on the moments that users rewind and rewatch the most. 90-second sections of the selected scenes were converted into text with the help of artificial intelligence and classified in terms of psychological, physical and economic violence elements.
In the manual verification performed on 108 scenes selected from the first 20 episodes of the "Kingfisher" series, the accuracy rate of the artificial intelligence-supported classification system was measured as 89 percent.
In the study examining the most rewatched scenes of four popular television series, elements of violence against women were found in approximately one-third of the scenes. It was determined that this rate reached 48 percent in one of the TV series.
Researchers pointed out that the presence of violent elements in scenes that viewers showed great interest in and watched repeatedly should be evaluated in terms of the social effects of media content.
Teachers' awareness of violence was examined
In the second part of the research, a 15-item "Violence Against Women Scale" was developed to measure teachers' awareness of violence against women.
With the permission of the Ministry of National Education, the data obtained from the scale applied to 109 randomly selected teachers in Beşiktaş district were analyzed with the SPSS 27 program.
According to the research results, it was seen that teachers generally had high awareness about violence against women. Findings were obtained that female teachers were more sensitive to violence and male teachers were more affected by television series.
The study concluded that as professional seniority increases, teachers' level of influence from TV series increases, and their sensitivity to violence against women decreases.
The artificial intelligence methodology of the research was developed by İstinye University Data Science Master's Program Head and MIT Computer Science and Artificial Intelligence Laboratory (CSAIL) researcher Assoc. Dr. It was shaped in line with the interviews held with Şebnem Özdemir.
Guidance teacher Özgür Aydemir stated that the study shows that artificial intelligence can be used effectively in the analysis of social problems and creates a new basis for discussion regarding the visibility of violence against women in media content.
Aiming for the findings of the research to be evaluated by relevant institutions and decision makers, the students also launched a signature campaign on "change.org".
Bora Candan and Onur Yıldırım argued that using scenes of violence against women in television series as a rating and attention-grabbing element could lead to the normalization and commercialization of violence.
Young researchers, who suggested imposing restrictions on violent scenes in TV series, expanding artificial intelligence-supported content controls and providing training to screenwriters, stated that their aim was to increase social awareness by transforming the narrative language on the screens rather than punishing.
