1. the process of sampling a signal or data at a rate higher than the minimum required by the Nyquist theorem, used to improve signal quality and reduce noise.
Usage: technical; signal processing; audio and digital engineering
2. in machine learning and statistics, the technique of increasing the number of samples from a minority class in an imbalanced dataset to improve model training.
Usage: technical; machine learning; data science

Examples

• Oversampling at four times the original rate helped reduce aliasing artifacts in the audio recording.
• The engineer used oversampling to improve the signal-to-noise ratio in the digital converter.
• To handle the imbalanced dataset, the team applied oversampling to the minority class before training the classifier.
• Oversampling techniques are commonly used in analog-to-digital conversion to achieve higher fidelity.
• The model's performance improved significantly after applying oversampling to the underrepresented class.