BİLDİRİLER

BİLDİRİ DETAY

Burak BATIBAY, Atabey KAYGUN
MACHINE LEARNING DRIVEN TV SERIES RATING WEEKLY FORECAST: STRATEGIC INSIGHTS INTO VIEWER BEHAVIOR
 
In broadcasting, viewership ratings which are shaped by audience demographics, habits, and channel-specific seasonal trends, are crucial to revenue-related decisions and strategic planning. In competitive environments, strategic planning relies on foreseeing outcomes, particularly revenue generation-related ones. So, with burgeoning digital consumption habits, an industry adaptation to proactive realignments in strategic planning has become necessary. This study introduces a comprehensive machine learning approach in integrating daily rating data, along with external data such as weather conditions, sports events, and IMDB data. We aim to significantly aid production companies in their strategic and budgetary operations by offering insights into factors that may alter the rating trends of their television shows. Leveraging machine learning methods, our model captures the intricate interplay of diverse variables to predict television ratings within a two-week time window. Using such disparate data sources provides us with a more nuanced insight into the factors contributing to the success of television series than just using traditional viewership metrics. The machine learning model we developed achieves an impressive 90% success rate in forecasting ratings with an average deviation error of 0.25 units within the forecasting window. This study is poised to significantly aid production companies in their strategic and budgetary operations by offering insights into factors that may alter the rating trends of their television series. This study was produced from the "Master's Thesis" of the first author. ORCID NO: 0009-0008-4565-6923

Anahtar Kelimeler: Predictive Analytics in Broadcasting, TV Series Rating Estimation, Support Decision Systems, Machine Learning



 


Keywords: