Image Process Based Recommender System for Social Media Marketing

Social media has become one of the most powerful marketing tools, particularly comments, emojis, and stickers plays vital role in increasing the company and products branding in the market. The popularity of social media posts with images are getting increased steadily for the past few years, and social media posts with images receive more user attention than normal text posts. The purpose of this research is to devise a recommender system that can evaluate the effectiveness of images, emojis, and stickers in promoting products on social media and use different image processing techniques such as neural networks to process the pixel-level details of an image and filtering images based on image quality assessment. This research demonstrated the impact of emoji and text categorization on famous business Instagram accounts on cosmetics, personal care, and perfumes. CNN model was used in this research and it was found that images based social media postings are getting more comments and sales rather than text based postings.