Feature selection (FS) is essential for enhancing machine learning (ML) performance by identifying relevant features from high-dimensional data. This study introduces a Quantum-inspired Artificial Flora Optimization Algorithm (QAFOA) for feature selection by combining quantum state updates with natural flora propagation principles. Evaluated on multiple benchmark datasets, the QAFOA demonstrated significant improvements in classification accuracy while reducing the number of selected features. Comparative analysis with other state-of-the-art methods revealed that the QAFOA consistently outperforms these methods in terms of classification accuracy and computational efficiency. Future research directions include integrating additional quantum principles such as quantum gates, entanglement, and superposition, as well as exploring the implementation of the QAFOA on actual quantum hardware to harness advancements in quantum computing technology.