Comprehensive study and Analysis of Extreme Multi-Label Classification Approach
Purvi Prajapati · International Journal of Advanced Trends in Computer Science and Engineering · 2020
Due to transformation of analog to digital world, huge volume of data is generated on daily basis.Millions of users are uploading, downloading and searching data on and from Social Media, Wikipedia, YouTube, Amazon, Websites etc.It is essential to analyze and retrieve significant information from millions of users with millions of categories.Hence, this is a challenging task to build classifier which will classify huge amount of data with its relevant subset of categories.In Recommendation System, the main goal is to recommend users based on the available data.However, this traditional recommendation system fails to deal with millions of items or labels in short time span.Extreme Classification approach is the recently introduced research area to tackle large amount of data with multi-label environment for the classification.Extreme Multi-Label classifier will construct model and predict the relevant categories from the large amount of available categories.This paper discussed different approaches for large scale Recommendation System using Extreme Multi-Label Classification Approach and empirical evaluation carried out on three multi-label datasets which handles large volume of the data.