Grand Banquet: A next-generation chemometrics framework for classification analysis on data matrix

The advancement of high-throughput analytical instrumentation has placed analytical science into the "Fourth Paradigm" of data-intensive discovery. In this new era, the limitation of scientific insight is not merely data acquisition but the objective, reproducible interpretation of highdimensional data matrices. Conventional chemometric workflows remain fundamentally reliant on "try-and-trust" methodologies. These expert-assisted paradigms require analysts to manually select preprocessing filters, feature selection methods, and classification models based on intuition or limited resources. Such subjectivity introduces analyst bias, raising concerns about reproducibility when applied to biological or chemical signals. Inspired by the concept "there is no one silver bullet", a next-generation chemometrics framework, named Grand Banquet (GB), is designed to standardise the analytical pipeline in this paper. GB replaces manual selection with a near-exhaustive approach, evaluating 990 pipelines (combinations of 15 preprocessing, 6 feature selection, and 11 classification methods) using various performance indices. Beyond this automatic pipeline, Qwen series Large Language Models (LLMs) have been integrated to generate natural-language narratives of these results, bridging the gap between complex computational outputs and human interpretation. GB offers a near-standard baseline ("Glass-box") for data analysis and a practical step toward method governance and reproducibility. The operation of GB has been tested on four public datasets from the Chemometrics Group at Copenhagen University. GB generates near-exhaustive pipelines for each dataset, with validation accuracies ranging from 40% to 100%, highlighting its ability to systematically explore the data matrix rather than providing only best-performing results. To pay tribute to experts in this field and to promote the standardisation of data analysis practices, related Python files and manuals have been published on GitHub under the Apache License 2.0 licence (GitHub link: https://github.com/Omicsforever867/Grand-Banquet).