What makes a picture worth a thousand words? Predicting image popularity and evaluating its features, on social media for NGOs.

What makes a picture worth a thousand words? Predicting image popularity and evaluating its features, on social media for NGOs.

Our graduate student Elina Koutromanou, represented the “New Technologies Laboratory – Department of Communication and Media, UoA”, at CCLABS022, the Annual Greek-speaking Conference of Communication Laboratories, presenting her thesis entitled: What makes a picture worth a thousand words? Predicting image popularity and evaluating its characteristics, on social media for NGOs.

This study aims to analyze the role of image in social media posts for NGOs and to predict the popularity of a post based on the characteristics of the image it includes. More specifically, the ultimate goal is to use the results for an effective selection of images by NGOs to raise awareness and inform the greater possible audience. To achieve this goal, this paper analyzes and extracts the characteristics of images posted on NGOs on Facebook and Instagram profiles and then uses them to develop Machine Learning Algorithms, with Python, aiming to predict whether a post will be popular or not, as well as which characteristics are important for prediction.

You can watch the presentation at the link below: ttps://www.youtube.com/watch?v=V52iKEkarUk&t=705s

Leave a Reply

Close Menu