A novel hybrid particle swarm optimization and gravitational search algorithm for multi-objective optimization of text mining

Big-data is one of the milestones on the web especially on social media (SM). Due to the widespread popularity of SM on the web, it is a painful task to capture the essence of SM. In this study, mining big social media data is re-formulated into a multi-objective optimization (MOO) task for an extractive summary. A Gravitational Search Algorithm (GSA) is utilized for optimizing several expressive objectives for generating a concise summary of SM. Moreover, particle swarm optimization (PSO) is mixed with GSA in a new shape to strengthen a local search ability and slow convergence speed in standard GSA. Whereas some users may demand the brief at any moment, several groups are constituted for incremental updating process during real-time based on naive Bayes algorithm. From experimental results, the proposed approach outperformed other notable and state-of-art comparative methods. (c) 2020 Elsevier B.V. All rights reserved.