Predicting changes in customer purchasing behavior using CNN on social networks

With the expansion of social networks and the proliferation of image data published by users, data analysis has become an essential tool for understanding customer behavior. Images, as sources of information, can more accurately reflect the emotions, desires, and behaviors of consumers. This study focuses on leveraging advanced image analysis techniques to gain deeper insights into consumer sentiments and preferences. In this study, we utilized Python TensorFlow to train our model on standard datasets, harnessing the power of Convolutional Neural Networks (CNNs) and the VGG۱۶ architecture. CNNs are particularly well-suited for image recognition and analysis tasks due to their ability to automatically detect and learn from hierarchical patterns within images. The VGG۱۶ architecture, known for its depth and effectiveness, further enhances the model's ability to accurately classify and interpret complex visual data. Our models are designed to extract a range of emotional states from images, including happiness, worry, and anger. By analyzing facial expressions, body language, and other visual cues, these models can provide a nuanced understanding of consumer emotions. This capability is particularly valuable in the context of social media, where images often convey more authentic and spontaneous emotions compared to text-based posts.

کلیدواژه ها:

CNN ، VGG۱۶ ، customer behavior ، social networks