IPSO-FDT-based polymorphic disaster information fusion method

Abstract

The invention relates to a polymorphic disaster information fusion method based on IPSO-FDT, which is an algorithm for fusing polymorphic disaster information, belongs to the field of information driving and machine learning, and is characterized by comprising the following steps: (1) initializing the speed and position of particles; (2) determining the attribute importance of the fuzzy decision tree; (3) determining a category distribution matrix; (4) determining a fuzzy probability; (5) calculating the fitness; (6) determining an update weight; and (7) determining particle speed and displacement parameters. According to the method, the problem of analysis result errors caused by traditional disaster information analysis due to complexity and diversity of disaster information is effectively solved, the disaster information analysis speed is effectively improved by optimizing the information fusion speed, and the disaster information research and judgment efficiency is improved. And a method with relatively high fusion accuracy is provided for the field of polymorphic disaster information fusion.

Description

technical field [0001] The invention relates to the fields of information drive and machine learning, and is a method for fusing polymorphic disaster information. Background technique [0002] The accuracy of disaster information is the basis for scientific disaster relief. At present, the commonly used methods cannot balance accuracy and efficiency well, which is not conducive to post-disaster work. Traditional disaster information fusion includes CPOS and D-S evidence theory. CPOS is an optimization algorithm based on chaotic particle swarm optimization. Compared with the traditional particle swarm optimization algorithm, it has stronger convergence and plays a certain role in improving the results of disaster information. Due to the complexity and diversity of disaster information, different information The weight will directly affect the judgment result. CPOS has limited ability to optimize the weight of disaster information, and cannot accurately identify the key part o...

AI Technical Summary

Problems solved by technology

[0002] The accuracy of disaster information is the basis for scientific disaster relief. At present, the commonly used methods cannot balance accuracy and efficiency well, which is not conducive to post-disaster work. Traditional disaster information fusion includes CPOS and D-S evidence theory

CPOS is an optimization algorithm based on chaotic particle swarm optimization. Compared with the traditional particle swarm optimization algorithm, it has stronger convergence and plays a certain role in improving the results of disaster information. Due to the complexity and diversity of disaster information, different information The weight will directly affect the judgment result. CPOS has limited ability to optimize the weight of disaster information, and cannot accurately identify the key part of the disaster information, which is prone to result deviation and reduces the accuracy of disaster analysis and judgment.

The D-S evidence theory is to obtain the preliminary decision-making results of the disaster information through the states of different information points collected, and then make a comprehensive judgment based on the reliability to obtain the fusion result. However, due to the diversity and uncertainty of the disaster information, the above method cannot Failure to make timely and efficient judgments will delay the timing of disaster relief

Images