Abstract:Learning with streaming data has attracted extensive research interest in recent years. Existing online learning approaches have specific assumptions regarding data streams, such as requiring fixed or varying feature spaces with explicit patterns and balanced class distributions. While the data streams generated in many real scenarios commonly have arbitrarily incomplete feature spaces and dynamic imbalanced class distributions, making existing approaches be unsuitable for real applications. To address this issue, this paper proposes a novel Online Learning from Incomplete and Imbalanced Data Streams (OLI $^{2}$ DS) algorithm. OLI $^{2}$ DS has a two-fold main idea: 1) it follows the empirical risk minimization principle to identify the most informative features of incomplete feature spaces, and 2) it develops a dynamic cost strategy to handle imbalanced class distributions in real-time by transforming F-measure optimization into a weighted surrogate loss minimization. To evaluate OLI $^{2}$ DS, we compare it with state-of-the-art related algorithms in three kinds of experiments. First, we adopt 14 real datasets to simulate three scenarios of incomplete feature spaces, i.e., trapezoidal, feature evolvable, and capricious data streams. Second, based on a benchmark online analyzer, we generate 13 datasets to simulate incomplete data streams with different imbalance ratios. Third, we analyze concept drift in two simulated scenes, i.e., online learning and data stream mining, and verify the adaption of OLI $^{2}$ DS on repeated concept drifts and variable imbalance ratios. The results demonstrate that OLI $^{2}$ DS achieves a significantly better performance than its rivals. Besides, a real-world case study on movie review classification is conducted to elaborate on our OLI $^{2}$ DS algorithm's effectiveness. Code is released at https://github.com/youdianlong/OLI2DS.

LIBOL: a Library for Online Learning Algorithms

SOL: A Library for Scalable Online Learning Algorithms

A Local Online Learning Approach for Non-linear Data.

LIBS2ML: A Library for Scalable Second Order Machine Learning Algorithms

MODL: Multilearner Online Deep Learning

CoLLiE: Collaborative Training of Large Language Models in an Efficient Way

LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning

Online learning: A comprehensive survey

A Buffered Online Transfer Learning Algorithm with Multi-Layer Network

Online Learning From Incomplete and Imbalanced Data Streams

LILO: Learning Interpretable Libraries by Compressing and Documenting Code

An Online-Offline Combined Big Data Mining Platform.

TOLA: Topic-oriented Learning Assistance Based on Cyber-Physical System and Big Data.

Deep Optimized Broad Learning System for Applications in Tabular Data Recognition

Fedlearn-Algo: A flexible open-source privacy-preserving machine learning platform

libact: Pool-based Active Learning in Python

LLAMA: Leveraging Learning to Automatically Manage Algorithms

HetEOTL: an Algorithm for Heterogeneous Online Transfer Learning

Dynamic Opposite Learning Enhanced Teaching–learning-Based Optimization

Personalized Online Machine Learning