Open AccessDescription and Application Research of Multiple Regression Model Optimization Algorithm Based on Data Set Denoising
Author(s) -
Hao Kang,
Hailong Zhao
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1631/1/012063
Subject(s) - data set , computer science , noise (video) , algorithm , set (abstract data type) , regression analysis , reliability (semiconductor) , sample (material) , data mining , regression , process (computing) , mathematical optimization , mathematics , statistics , artificial intelligence , machine learning , power (physics) , chemistry , physics , chromatography , quantum mechanics , image (mathematics) , programming language , operating system
Multiple regression model is based on a large number of data sample set in the prediction process, and the noise data in the data set will have a great impact on the results of the fitting equation, which makes the results unreliable and unreliable. This paper forwards fuzzy least square method based on fuzzy set theory, it can optimize the regression model algorithm and reduce the influence of noise data on the fitting equation. This algorithm is applied to the real estate price forecast, it would obtain the final fitting equation after repeating iterative calculation, which makes the price prediction and eliminates the influence of bad data on the fitting results and improves the reliability and availability of the forecast results.