Figure 1
The diagram starts at the top left, with a database icon labeled “Data Students”, which feeds downward into a rectangular block labeled “Data Preprocessing”. Inside this block, two components are labeled “Feature Encoding” and “Feature Scaling”. A small icon is shown at the bottom left in the Data Preprocessing” box. “Data Preprocessing” branched into two: an arrow leads upward to a database icon labeled “Data Train”, and another arrow leads downward to a database icon labeled “Data Test”. From “Data Train”, an arrow points to a large, rounded rectangle labeled “Modelings”. Inside this modeling block, ten labeled algorithm boxes are arranged in two columns, and five rows, labeled row-wise as “Logistic Regression”, “Support Vector Machines”, “Decision Trees”, “X G Boost Regressor”, “Random Forests”, “Artificial Neural Networks”, “CatBoost”, “Recurrent Neural Network”, “Light G B M”, and “Convolutional Neural Network”. An arrow from the “Modelings” block points downward to a rectangular block labeled “Evaluate, Prediction”. Inside this block, evaluation metrics are labeled on the right as “M S E (Mean Squared Error)”, “R 2 Score”, “R M S E (Root M S E)”, and “M A E (Mean Absolute Error)”. On the left of this block, the word “Cheating?” appears at the bottom alongside a small cartoon-style icon of a person standing next to a checklist and symbols. The “Data Test” database icon leads rightward to the “Evaluate, Prediction” box.The predicted student cheating model was analyzed based on the scores in the learning process