How does it work?
The authors of the paper focused on Instagram influencers specifically. The researchers built a content analysis tool for user and target brand profiles. The researchers then sampled 20 unique user accounts with five content themes: dogs, cats, mountains, cars, and pizza, and downloaded the brand profiles, which each serve markets that fall into one of these categories.They then applied an image classification algorithm to analyse the content to extract common themes, which output a list of the five most likely tags for each image. Then, the three most likely tags were compiled into a ‘string’ for each user. Finally, the influencer and target brand data were fed into a model that suggested matches.The results
The results indicated that the algorithm, when presented with a variety of potential influencer profiles, is able to identify those profiles that are most closely aligned with a particular brand. There is more work to be done, as the researchers said the AI hasn’t yet been tested on less distinct categories and predictions rely on accurate tagging but the image classification algorithm. By incorporating additional user data, using a better image analysis engine and allowing for manual input of what a brand is looking for in an influencer, things could be improved.This AI could encourage the creation of high-quality content in the future and save brands time and money. It is an exciting step towards an automated influencer finding system and it will be exciting to see what the future holds.
How do you think AI will affect influencer marketing in the future? Comment below or reach out on Twitter or LinkedIn.