Data Science

The process of building accurate predictive models for spectroscopy instruments is often a journey of discovery, with many twists and turns. Once we have an initial model, every step of the process deserves scrutiny. Are we using the best method of sample preparation? Is our reference data accurate and reliable?

The success of generative AI applications such as ChatGPT and DALL-E has increased public awareness of the power of artificial intelligence software. Sagitto's Benchmarking Service allows users of infrared spectroscopy instruments to benchmark their current models against models generated by machine learning.

Outlier detection is an important step in preparing spectroscopy data for machine learning models. Hotelling's T2 and Q-Residuals are two outlier detection methods commonly used in chemometrics. However, Sagitto has found that they need to be used with caution to avoid discarding unusual but valid data.

When building a machine learning model, our customers often ask "How much training data will we need?" It's rather like kids in the back seat of the car on a long journey, asking how much further until we get there?