H.3. Artificial Intelligence
Morteza Mohammadi Zanjireh; Nasrin Nazari
Abstract
Predicting customer purchase intention is an important yet challenging task in e-commerce due to the sequential nature of user interactions and class imbalance in real-world shopping data. This study presents a purchase prediction framework that combines conventional session-based behavioral features ...
Read More
Predicting customer purchase intention is an important yet challenging task in e-commerce due to the sequential nature of user interactions and class imbalance in real-world shopping data. This study presents a purchase prediction framework that combines conventional session-based behavioral features with semantic similarity measures derived from learned product embeddings. Experiments were conducted on a large-scale online cosmetics store dataset containing more than 12 million user interaction events. After preprocessing and session construction, product sequences were represented using four embedding techniques: Word2Vec, Doc2Vec, GloVe, and FastText. Rather than directly using embedding vectors as predictive inputs, the proposed approach derives semantic similarity measures from product representations and integrates them with conventional behavioral attributes. To address class imbalance, SMOTE and RandomUnderSampler were evaluated. The resulting feature sets were assessed using six machine learning classifiers: Decision Tree, Random Forest, Bagging, AdaBoost, Logistic Regression, and XGBoost. The results show that semantic similarity features generally improve prediction performance across the evaluated configurations. Doc2Vec achieved the strongest overall performance among the investigated embedding techniques, while Random Forest and Bagging performed particularly well. Additional ROC-AUC and PR-AUC analyses further support the usefulness of the proposed similarity-based representation beyond the primary F1-score evaluation. Overall, the findings indicate that semantic relationships among products provide complementary information to conventional behavioral features and can contribute to improved purchase prediction in e-commerce environments.
A.1. General
Morteza Mohammadi Zanjireh; Farzad Morady
Abstract
This paper predicts the severity of crashes based on the analysis of multiple variables and using machine learning methods. For this purpose, data related to the years 2012 to 2024 of Tempe city in the state of Arizona USA was used. Features were selected using the metaheuristic method. Then, by using ...
Read More
This paper predicts the severity of crashes based on the analysis of multiple variables and using machine learning methods. For this purpose, data related to the years 2012 to 2024 of Tempe city in the state of Arizona USA was used. Features were selected using the metaheuristic method. Then, by using decision tree and artificial neural network, the classification of the severity of crashes was carried out. Based on the metrics, decision tree with an overall accuracy of 54% was the optimal. Finally, using the permutation feature importance method, the optimal model was interpreted. The results show that the characteristics of the year with 0.22 and the spatial characteristics with 0.11 and the collision manner with 0.1 have a higher importance in predicting the severity of crashes on urban roads.
