Long-Tail Data-Driven Recommendations - Innovative Solutions for Financial Recommender Systems

Abstract

Social media financial recommender systems are demonstrating their ability to better address the interests of netizens and offer greater promise for better meeting their financial service needs. This article discusses the role of long-tail data from network users in financial recommender systems. Long-tail data is supposed to improve the accuracy of financial recommendations, expand the customer base, and increase the availability of financial services. To do this, consider the model of operation of data of long tails (LTD) through verification and correction by discriminator systems (DM) prepared by generative filters (GM) of financial recommendations (R): LTD → DM → GM → R.