Quantum Firefly Algorithm with Random Neighborhood Search for Large-scale Data Analysis

Abstract

With the rapid development of information technology, the continuous expansion of data scale poses higher challenges to data analysis algorithms. This paper proposes a Quantum Computation Optimization Algorithm (QCOA), specifically a quantum firefly algorithm, which leverages quantum superposition, entanglement, and rotation gates to accelerate convergence and avoid local optima in optimization tasks. The proposed QCOA is experimentally tested on a large-scale IMDB dataset with one billion records using a high-performance quantum simulation environment based on Qiskit and Python. Key parameters include 50 fireflies, 32 qubits, 1000 iterations, and a random neighborhood search mechanism. The algorithm's performance is evaluated against classical baselines including SVM, KNN, and GBDT using metrics such as false positive rate, accuracy, and processing time. Results show QCOA achieves the lowest false positive rate (2.7%), highest accuracy (97.3%), and the shortest processing time (90 seconds), outperforming all classical baselines. These findings validate the practical advantages of QCOA and quantum computing in large-scale data processing and optimization, offering a promising approach for future data-intensive applications.