ZeroMat: Solving Cold-start Problem of Recommender System with No Input Data

TL;DR

ZeroMat is a novel recommender system technique that effectively addresses the cold-start problem by predicting user-item ratings without requiring any initial input data, outperforming random recommendations and matching traditional methods in accuracy.

Contribution

This paper introduces ZeroMat, a new approach that eliminates the need for input data in cold-start scenarios, providing a competitive and fair recommendation performance.

Findings

ZeroMat achieves comparable Mean Absolute Error to classic matrix factorization.

ZeroMat outperforms random placement in recommendation quality.

ZeroMat maintains fairness metrics similar to data-driven methods.

Abstract

Recommender system is an applicable technique in most E-commerce commercial product technical designs. However, nearly all recommender system faces a challenge called the cold-start problem. The problem is so notorious that almost every industrial practitioner needs to resolve this issue when building recommender systems. Most cold-start problem solvers need some kind of data input as the starter of the system. On the other hand, many real-world applications place popular items or random items as recommendation results. In this paper, we propose a new technique called ZeroMat that requries no input data at all and predicts the user item rating data that is competitive in Mean Absolute Error and fairness metric compared with the classic matrix factorization with affluent data, and much better performance than random placement.