Author : Chandra Shekhar Gautam and Dr.Prabhat Pandey

An improved query optimization process for big data using a combination of the ACO-GA algorithm and HDFS map-reduce. The methodology consists of two phases: BD arrangement and query optimization. In the first phase, the input data is pre-processed by finding the hash value using the SHA-512 algorithm and removing repeated data using HDFS map-reduce. Then, features such as closed frequent pattern, support, and confidence are extracted and managed using entropy calculation. Based on this calculation, related information is grouped using the Normalized K-Means algorithm. In the second phase, the BD queries are collected, and the same features are extracted. The optimized query is then found using the ACO-GA algorithm, and the similarity assessment process is performed [1].The paper claims that the proposed algorithm out performs other existing algorithms. However, without more details about the experimental setup and the specific metrics used to evaluate the performance of the algorithm, it is difficult to assess the validity of this claim. Additionally, it is unclear how the proposed algorithm compares to other state-of-the-art query optimization techniques. Further research and comparative analysis are needed to fully evaluate the effectiveness of this approach [4][5].

Keywords: Hadoop Distributed File System (HDFS), Normalized K-Means (NKM) algorithm, Ant Colony Optimization-Genetic Algorithm (ACO-GA), Secure Hash Algorithm (SHA-512)