RETRACTED ARTICLE: A Novel QACS Automatic Extraction Algorithm for Extracting Information in Blockchain-Based Systems

Abstract

Information extraction and data mining are the important aspects in blockchain technology. Big- data mining tools can perform pattern recognition assignments from thousands to billions of blockchain communications to recognize evil users and fraudulent transactions. The current information extraction algorithms have the limitations of large memory occupation, long running time, high dimensionality, low extraction accuracy, poor convergence precision and slow convergence speed. There is a need to devise newer techniques with the potential of fast and accurate information extraction features in blockchain based systems to support high risk international financial transactions. Thus, in this paper, the automatic extraction algorithm of data mining as a case study of blockchain communications based on QACS (quantum adaption cuckoo search) method is proposed. Principal component analysis (PCA) is used to map the original features of blockchain communications into low-dimensional feature space through linear transformation, and then to replace the original features with fewer absolutely needed features. The features of coding information in information management system are reduced, and the optimal feature subset is obtained. K-means is used to extract the key information in the optimal feature subset. An improved and adaptive Cuckoo search is proposed in this paper where quantum operation is introduced into the original K-means algorithm for automatic extraction of coded information in blockchain communications. The results prove that the proposed QACS method has the potential of automatic extraction of the encoded information with lower execution time, higher extraction accuracy, and faster convergence speed, and is significant for blockchain based systems.