This study aims to identify the best deep learning and machine learning models to identify the unethical behavior patterns of learners using distance education exam data of an educational institution. One hundred twenty-nine online exam data were analyzed by the researcher with three different scenarios to reveal the best model performance in regression and classification. For regression and classification, deep neural network (DNN) from deep learning algorithms and support vector machine (SVM), decision trees (DTs), k-nearest neighbor (KNN), random forest (RF), logistic regression (LR), and extreme gradient boosting (XGBoost) algorithms from machine learning algorithms were used. In the regression analysis conducted within the scope of Scenario-1, the model we proposed to detect “cheating” behavior, which is one of the unethical learner behaviors, was found to be a 5-layer DNN model with a test performance success of 80.9%. In the binary classification analysis for Scenario-2, students who “copied” from unethical behaviors were obtained with an accuracy rate of 96.9% by the model established by the 10-layer DNN algorithm we proposed. In the triple classification analysis for Scenario-3 defined in the study, the XGBoost model was found to have the highest accuracy rate of 97.7% for students who “cheated” due to unethical behaviors and the highest performance in all other metric values. In addition, SHAP and LIME methods, which are explanatory methods for the XGBoost model, which is one of the best-performing models, were applied, and the attributes and percentages affecting the model were shared. As a result of this study, it has been shown that the application of the most appropriate layer functions and parameter selection that will increase performance can be effective in estimating complex problems and target values that cannot be solved using classical mathematical models. The proposed models can provide educational institutions with a roadmap and insight in evaluating online examination practices and ensuring academic integrity. Future researchers may need more data sets and different analyses for better performance of the established models.

This study aims to reveal the best deep learning models that are improved and optimized by predicting undesirable behavior patterns using a dataset consisting of artificial and real exam data of students taking online distance education courses in an online environment through the distance education system. Using online exam data of 129 students, the researchers conducted analysis with two different scenarios to determine the best prediction performance through regression and classification models. The model we proposed was determined as a four-layer DNN with 80.4% test performance in detecting students who “cheated” from undesirable behavior patterns, which was performed with K-10, K-5 and K-3 cross-validation. The results prove that students' online distance education exam data can be easily applied to the DNN model. The models presented in the study provide a roadmap for educational institutions to evaluate their online examination practices and develop more effective strategies for academic honesty.

The global Covid-19 pandemic has accelerated the implementation of distance education and training, resulting in a shift of course activities and processes to an online environment. The introduction of online exams in this new educational landscape has raised concerns among learners regarding academic honesty. This situation has prompted worries about safeguarding the ongoing institutional identity and reputation of academic institutins, thereby raising questions about academic honesty. This study presents a systematic review of the literature on protecting academic honesty in online examinations, focusing on recent publications within the last five years. This period is marked by intensified online exam evaluations and the impact of Covid-19. Based on the literature review, a total of 50 studies were examined by three experts in the field. Descriptive and content analysis techniques were utilized to identify research trends. The findings of these studies indicated that learners tend to accept unethical behavior violations, which are also increasing. In conclusion, it is recommended that future studies consider factors such as a wider range of publication years, employed research methods, increased publication rates, success rates, dataset size, and the utilization of assistive technologies.