Journal: PLOS One

Article Title: Advanced computational approaches for predicting sunflower yield: Insights from ANN, ANFIS, and GEP in normal and salinity stress environments

doi: 10.1371/journal.pone.0319331

Figure Lengend Snippet: Sunflower seed yield prediction steps using ANN, ANFIS, and GEP models.

Article Snippet: Training of the ANFIS model continued until the MSE fell below 0.001 or a maximum of 1,000 epochs was reached, similar to the ANN model. MATLAB’s Fuzzy Logic Toolbox was used to implement the ANFIS model. Every fuzzy system includes three main parts: fuzzifying the data by defining the membership function, creating a connection between the input and output by means of a series of rules (if-then), and gathering the results of the system and non-fuzzification.

Techniques:

Journal: PLOS One

Article Title: Advanced computational approaches for predicting sunflower yield: Insights from ANN, ANFIS, and GEP in normal and salinity stress environments

doi: 10.1371/journal.pone.0319331

Figure Lengend Snippet: Evaluating the efficacy of three models (ANN, ANFIS, and GEP) to predict sunflower grain yield under normal and salt stress.

Article Snippet: Training of the ANFIS model continued until the MSE fell below 0.001 or a maximum of 1,000 epochs was reached, similar to the ANN model. MATLAB’s Fuzzy Logic Toolbox was used to implement the ANFIS model. Every fuzzy system includes three main parts: fuzzifying the data by defining the membership function, creating a connection between the input and output by means of a series of rules (if-then), and gathering the results of the system and non-fuzzification.

Techniques:

Journal: PLOS One

Article Title: Advanced computational approaches for predicting sunflower yield: Insights from ANN, ANFIS, and GEP in normal and salinity stress environments

doi: 10.1371/journal.pone.0319331

Figure Lengend Snippet: Comparison of the accuracy evaluation statistics of models (ANN, ANFIS, and GEP) to predict the yield of sunflower grains in the test stage: (a) Normal conditions and (b) Salt stress conditions.

Article Snippet: Training of the ANFIS model continued until the MSE fell below 0.001 or a maximum of 1,000 epochs was reached, similar to the ANN model. MATLAB’s Fuzzy Logic Toolbox was used to implement the ANFIS model. Every fuzzy system includes three main parts: fuzzifying the data by defining the membership function, creating a connection between the input and output by means of a series of rules (if-then), and gathering the results of the system and non-fuzzification.

Techniques: Comparison

Journal: PLOS One

Article Title: Advanced computational approaches for predicting sunflower yield: Insights from ANN, ANFIS, and GEP in normal and salinity stress environments

doi: 10.1371/journal.pone.0319331

Figure Lengend Snippet: Violin diagrams of models (ANN, ANFIS, and GEP) to predict the yield of sunflower grains in the test stage: (a) Normal conditions and (b) Salt stress conditions.

Article Snippet: Training of the ANFIS model continued until the MSE fell below 0.001 or a maximum of 1,000 epochs was reached, similar to the ANN model. MATLAB’s Fuzzy Logic Toolbox was used to implement the ANFIS model. Every fuzzy system includes three main parts: fuzzifying the data by defining the membership function, creating a connection between the input and output by means of a series of rules (if-then), and gathering the results of the system and non-fuzzification.

Techniques:

Journal: PLOS One

Article Title: Advanced computational approaches for predicting sunflower yield: Insights from ANN, ANFIS, and GEP in normal and salinity stress environments

doi: 10.1371/journal.pone.0319331

Figure Lengend Snippet: Taylor diagrams to compare the performance of models (ANN, ANFIS, and GEP) to predict the yield of sunflower grains in the test stage: (a) Normal conditions and (b) Salt stress conditions.

Article Snippet: Training of the ANFIS model continued until the MSE fell below 0.001 or a maximum of 1,000 epochs was reached, similar to the ANN model. MATLAB’s Fuzzy Logic Toolbox was used to implement the ANFIS model. Every fuzzy system includes three main parts: fuzzifying the data by defining the membership function, creating a connection between the input and output by means of a series of rules (if-then), and gathering the results of the system and non-fuzzification.

Techniques: