Implementing Prioritized-Breadth-First-Search for Instagram Hashtag Recommendation

Abstract

Instagram is one of the most used social media platforms where around 500 million users interact with content daily, which creates excellent marketing opportunities. Instagram’s marketing growth is primarily focused on the page’s content, but suitable hashtags are equally essential to traffic. Hashtag research is one of the most complex parts of an Instagram marketing campaign, and usually, online tools are used to get a set of hashtags from one niche-specific hashtag. These tools are suitable for initial days, but their recommendations are often less in number, outdated, and lack connectivity among hashtags. This paper focuses on creating a set of highly related hashtags gathered from real-time data and sorting it in the best way possible using a parent hashtag as an input. This algorithm introduces a prioritization system that takes likes, occurrence, and rank into account and implements a Prioritized-Breadth-First-Search considering each hashtag as a node in a graph instead of an isolated entity.