SSMF: Shifting Seasonal Matrix Factorization

TL;DR

SSMF is an adaptive online matrix factorization method that detects and switches between multiple seasonal patterns in data streams, significantly improving demand forecasting accuracy for applications like taxi-ride counts.

Contribution

The paper introduces SSMF, a novel online matrix factorization approach that automatically detects regime shifts in seasonal patterns without human intervention.

Findings

Outperforms baseline methods in demand forecasting accuracy.

Effectively detects regime shifts in real-time.

Operates in constant time and memory per observation.

Abstract

Given taxi-ride counts information between departure and destination locations, how can we forecast their future demands? In general, given a data stream of events with seasonal patterns that innovate over time, how can we effectively and efficiently forecast future events? In this paper, we propose Shifting Seasonal Matrix Factorization approach, namely SSMF, that can adaptively learn multiple seasonal patterns (called regimes), as well as switching between them. Our proposed method has the following properties: (a) it accurately forecasts future events by detecting regime shifts in seasonal patterns as the data stream evolves; (b) it works in an online setting, i.e., processes each observation in constant time and memory; (c) it effectively realizes regime shifts without human intervention by using a lossless data compression scheme. We demonstrate that our algorithm outperforms state-of-the-art baseline methods by accurately forecasting upcoming events on three real-world data streams.