Unlock instant, AI-driven research and patent intelligence for your innovation.

Odor substance pleasure score prediction method, program and system based on chemical molecule descriptors

A technology of molecular descriptors and chemical molecules, which is applied in the field of chemical molecular information processing, can solve the problems of low prediction accuracy, achieve the effects of ensuring prediction accuracy, preventing over-fitting, and improving regression prediction performance

Pending Publication Date: 2021-11-02
HUNAN INST OF TECH
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] One of the purposes of the present invention is to provide a method for predicting pleasure scores based on chemical molecular descriptors to solve the problem of low prediction accuracy faced by existing methods

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Odor substance pleasure score prediction method, program and system based on chemical molecule descriptors
  • Odor substance pleasure score prediction method, program and system based on chemical molecule descriptors
  • Odor substance pleasure score prediction method, program and system based on chemical molecule descriptors

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0041] In order to facilitate those skilled in the art to better understand the improvements of the present invention over the prior art, the present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0042] In this example, figure 1 It shows the specific implementation steps of predicting the pleasantness score of the odor substance for the substance to be tested. Firstly, single-molecule odorant substances with obvious differences in structure and olfactory perception are screened out, and their SMILES numbers are collected, and high-dimensional chemical molecular descriptor features are obtained by using DRAGON7 chemical structure analysis software. Among them, the obtained high-dimensional chemical molecular descriptor features The data dimension is m, expressed as:

[0043] x o =[x 1 ,x 2 ,...,x m ].

[0044] Next, the features that cannot be calculated in the high-dimensional chemical molecule descriptor featu...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an odor substance pleasure score prediction method, program and system based on chemical molecule descriptors, and relates to the technical field of chemical molecule information processing. According to the method, after feature selection based on feature value diversity is carried out on high-dimensional chemical molecule descriptor features of a single-molecule odor substance, dimension reduction processing is carried out on chemical molecule descriptor features by utilizing a principal component analysis method, redundant information among chemical molecule descriptor feature data can be effectively removed while feature dimensions are reduced, and more effective feature data are extracted, so that the prediction performance of the pleasure is improved, and the prediction efficiency of the pleasure is improved. And aiming at the condition that the sample size of the smell perception data set of the smell substances is not large, the pruning operation in the random forest regression prediction model can effectively prevent overfitting, so that the regression prediction performance of the pleasure degree is improved, and the prediction precision is ensured.

Description

technical field [0001] The present invention relates to the technical field of chemical molecular information processing, in particular to a method, program and system for predicting the pleasantness score of odorants based on chemical molecular descriptors. Background technique [0002] Unlike visual and auditory representations, olfactory perception cannot be represented by a single physical quantity. It mainly uses olfactory perception descriptors such as "musky smell" and "fishy smell". In order to represent olfactory perception more accurately, relevant researchers have proposed thousands of descriptors of olfactory perception. Among them, pleasantness is an important dimension in odor perception. The prediction of pleasantness score is of great significance for the development of new products in these industries. The research on the prediction of odorant to pleasantness has always been one of the hot research issues in the field of olfactory perception. [0003] The i...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G16C20/30
CPCG16C20/30
Inventor 李欣贾雅琼俞斌任永梅
Owner HUNAN INST OF TECH