Research on tobacco composition analysis and ratio optimization strategy using data mining technology

The study first utilizes data mining techniques for the construction of a tobacco composition detection model, which is used to study the physical and chemical properties of tobacco attributes, sensory quality correlations, and intrinsic quality. In addition, the study uses genetic algorithm-constrained nonlinear optimization to seek the optimal ratio of each single ingredient of tobacco leaves composing the leaf group formulation from the chemical properties of tobacco leaves. The experimental validation results show that after using the leaf group formulation based on the genetic algorithm proposed in this paper for tobacco optimization design, it was found that the content of hydrocyanic acid and crotonaldehyde components were reduced by 79 and 16.55, respectively, compared with that of the traditional tobacco, thus verifying the superiority and feasibility of the ratio optimization strategy designed in this paper.