Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

Proposed scheme for thyroid disease prediction and explanation.

More »

Fig 1 Expand

Table 1.

Features of the dataset.

More »

Table 1 Expand

Fig 2.

Proposed data balancing technique.

More »

Fig 2 Expand

Table 2.

Confusion matrix.

More »

Table 2 Expand

Table 3.

Number of instances before and after data balancing.

More »

Table 3 Expand

Table 4.

Hyperparameters tuning of the classifiers using gridsearchCV.

More »

Table 4 Expand

Table 5.

Performance measure of our scheme using K-means+SMOTE+ENN.

More »

Table 5 Expand

Table 6.

Performance measure of our scheme using K-means+SMOTE+KNN.

More »

Table 6 Expand

Fig 3.

ROC curve of K-means+SMOTE+ENN.

More »

Fig 3 Expand

Fig 4.

ROC curve of K-means+SMOTE+KNN.

More »

Fig 4 Expand

Fig 5.

Precision-recall curve of K-means+SMOTE+ENN.

More »

Fig 5 Expand

Fig 6.

Precision-recall curve of K-means+SMOTE+KNN.

More »

Fig 6 Expand

Table 7.

The performance of ML classifiers on the Hungarian heart disease dataset without balancing.

More »

Table 7 Expand

Table 8.

The performance of ML classifiers utilizing K-means+SMOTE+ENN on the Hungarian heart disease dataset.

More »

Table 8 Expand

Table 9.

Performance measure after applying SMOTE.

More »

Table 9 Expand

Table 10.

Performance measure after applying NearMiss.

More »

Table 10 Expand

Table 11.

Performance measure after applying SMOTE+ENN.

More »

Table 11 Expand

Table 12.

Performance measure after applying SMOTE+KNN.

More »

Table 12 Expand

Fig 7.

Performance measure between SMOTE+ENN and K-means +SMOTE+ENN.

More »

Fig 7 Expand

Fig 8.

Performance measure between SMOTE+KNN and K-means+SMOTE+KNN.

More »

Fig 8 Expand

Fig 9.

Feature importance of thyroid disease dataset using Shapash.

More »

Fig 9 Expand

Fig 10.

Impact of the features on model output using SHAP value.

More »

Fig 10 Expand

Fig 11.

Local explanation for not having a thyroid disease for a random instance.

More »

Fig 11 Expand

Fig 12.

Local explanations for having a thyroid disease for a random instance.

More »

Fig 12 Expand

Fig 13.

Local explanations for having a thyroid disease for a random instance.

More »

Fig 13 Expand

Fig 14.

Comparison plot shows the effect of some instances on a model for having thyroid disease.

More »

Fig 14 Expand

Table 13.

Rank and frequency of the domain expert’s opinion.

More »

Table 13 Expand

Fig 15.

Plot of the summarized responses from survey.

More »

Fig 15 Expand

Table 14.

The summarized responses from the survey in three categories.

More »

Table 14 Expand

Fig 16.

Rank and achieve the score of the features from the survey.

More »

Fig 16 Expand

Table 15.

Features ranking according to XAI tools and domain experts.

More »

Table 15 Expand

Table 16.

Comparison table between our proposed scheme and the existing state-of-the-art methods.

More »

Table 16 Expand