Seasonal Averaged One-Dependence Estimators: A Novel Algorithm to Address Seasonal Concept Drift in High-Dimensional Stream Classification

TL;DR

This paper introduces SAODE, a new classifier designed to effectively handle seasonal concept drift in high-dimensional stream data, particularly in news article categorization, outperforming existing methods.

Contribution

The paper proposes SAODE, a novel extension of AODE that incorporates seasonal information as a super parent, effectively addressing seasonal drift in high-dimensional streams.

Findings

SAODE outperforms nine state-of-the-art models across five evaluation techniques.

SAODE significantly improves classification accuracy in seasonal drift scenarios.

The model is validated on large real-world datasets with approximately one million records.

Abstract

Stream classification methods classify a continuous stream of data as new labelled samples arrive. They often also have to deal with concept drift. This paper focuses on seasonal drift in stream classification, which can be found in many real-world application data sources. Traditional approaches of stream classification consider seasonal drift by including seasonal dummy/indicator variables or building separate models for each season. But these approaches have strong limitations in high-dimensional classification problems, or with complex seasonal patterns. This paper explores how to best handle seasonal drift in the specific context of news article categorization (or classification/tagging), where seasonal drift is overwhelmingly the main type of drift present in the data, and for which the data are high-dimensional. We introduce a novel classifier named Seasonal Averaged One-Dependence Estimators (SAODE), which extends the AODE classifier to handle seasonal drift by including time as a super parent. We assess our SAODE model using two large real-world text mining related datasets each comprising approximately a million records, against nine state-of-the-art stream and concept drift classification models, with and without seasonal indicators and with separate models built for each season. Across five different evaluation techniques, we show that our model consistently outperforms other methods by a large margin where the results are statistically significant.