Conceptual Complexity is to measure an individual’s ability to project their ability to perceive a different perspective of a certain subject. Historically, lt has been seen to be extraneous and time-consuming to manually score large sets of text. Therefore, there is a demand to automate scoring to significantly reduce time and expense. This notebook suggests that using modern machine learning algorithms is a better predictor than traditional methods specifically the efficacy of a presidential speech.
| Algorithms | Configuration | Accuracy |
|---|---|---|
| Support Vector Machine | kernel="rbf",C=13, gamma='scale',97 Component PCA | 92.32% |
| Random Forest | Gini Index | 88.6% |
| XG boosting | gamma=0, max_tree_depth=3, n_estimatror=100 | 94.1% |