Check patentability & draft patents in minutes with Patsnap Eureka AI!

Deep learning model prediction factor interpretation method and application thereof in soil water content prediction

A technology of soil water content and deep learning, which is applied in the interpretation of deep learning model prediction factors and the application field of soil water content prediction, which can solve problems that cannot meet the needs of internal analysis, and achieve the effect of scientific and effective management

Pending Publication Date: 2021-12-31
TIANJIN NORMAL UNIVERSITY
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the deep learning model directly maps the output results through the input variables. This operation method cannot meet the needs of different fields for internal analysis.

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
  • Deep learning model prediction factor interpretation method and application thereof in soil water content prediction
  • Deep learning model prediction factor interpretation method and application thereof in soil water content prediction
  • Deep learning model prediction factor interpretation method and application thereof in soil water content prediction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0041] A deep learning model predictive factor interpretation method, comprising the following steps:

[0042] Step 1: Standardize the time series data for different predictors and target outcomes.

[0043] Different factors and the unit of the target result data are not the same, directly put it into the deep learning model for learning and training, it is very easy to cause the model to be inaccurate due to the large numerical gap. Therefore, all factors and the resulting data were first standardized, and then the transformed data was input into the model to ensure the fairness of the contribution of different factors to the model.

[0044] Suppose the time series data of the original predictor is P, and the processed data series is Q, then: Q=[P-mean(P)] / std(P). Among them, mean(P) and std(P) are the mean and standard deviation of the time series data P, respectively.

[0045] Step 2: Based on the standardized prediction factors in step 1 and the target result data training...

Embodiment 2

[0076] On the basis of Embodiment 1, this embodiment applies the above-mentioned deep learning model prediction factor interpretation method to soil moisture prediction, including the following steps:

[0077] Step 1: Taking the weather factors related to soil water content as predictors, and processing the time series data of the predictors of weather factors related to soil water content and the target results of soil water content.

[0078] The units of different predictive factors and target result data are not the same. If they are directly put into the deep learning model for learning and training, it is very easy to cause the model to be inaccurate due to the large numerical gap. In order to ensure the fairness of the contribution of different predictors to the model, the data of all predictors were standardized, and then the transformed data were input into the model.

[0079] Suppose the time series data of the original predictor is P, and the processed data series is...

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 a deep learning model prediction factor interpretation method and application thereof in soil water content prediction. The method comprises the following steps of 1, standardizing time sequence data of different prediction factors and target results, 2, training based on the prediction factors after standardization processing in the step 1 and target result data to obtain a deep learning model for predicting the dynamic change of an output result, 3, calculating contribution degrees of different prediction factors to an output result of the constructed deep learning model by adopting a Shapley value method, and 4, according to the contribution degree of each prediction factor calculated in the step 3, determining important prediction factors in the deep learning model, removing unimportant influence factors, and simplifying different prediction factor data.

Description

technical field [0001] The invention belongs to the technical field of artificial intelligence, and in particular relates to a deep learning model prediction factor interpretation method and its application in soil water content prediction. Background technique [0002] Deep learning (Deep Learning) is a new research direction in the field of machine learning (Machine Learning). It learns the internal laws and representation levels of sample data, and the information obtained internally interprets the data, so that computers can be as capable as humans. Ability to learn, analyze and identify. Among them, the multi-layer perceptron (MLP) and convolutional neural network (CNN) models are the most classic. Around 2010, computer hardware and computing performance, as well as technologies such as big data and Internet + became increasingly mature, making deep learning develop rapidly. At present, deep learning models and frameworks have been extended to many fields such as imag...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G01N33/24G06N3/04G06N3/08
CPCG01N33/246G06N3/04G06N3/08
Inventor 黄新张楠
Owner TIANJIN NORMAL UNIVERSITY
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com