Browsing by Author "Sozen, Mert Erkan"

Now showing 1 - 5 of 5

Citation - WoS: 11
Citation - Scopus: 11
Leveraging explainable artificial intelligence in understanding public transportation usage rates for sustainable development
(PERGAMON-ELSEVIER SCIENCE LTD, 2024) Gorkem Sariyer; Sachin Kumar Mangla; Mert Erkan Sozen; Guo Li; Yigit Kazancoglu; Sariyer, Gorkem; Sozen, Mert Erkan; Li, Guo; Mangla, Sachin Kumar; Kazancoglu, Yigit
Public transportation usage prediction is valuable for the sustainable development of transportation systems particularly in crowded megacities. Machine learning technologies are of great interest for predicting public transportation usage. While these technologies outperform many other techniques they suffer from limited interpretability. Explainable artificial intelligence (XAI) tools and techniques that offer post -hoc explanations of the obtained predictions are gaining popularity. This paper proposes an advanced tree-based ensemble algorithm for public transportation usage rate prediction. We aim to explain the predictions both with the most widely used technique of XAI Shapley additive explanation (SHAP) and in the light of the rules presented. To predict the total public transportation usage the proposed model combines all types of public transportation categorized as ferry railway and bus unlike most existing studies focusing on a single kind of public transport. Besides the sort of transportation the day of the week whether the day is special and the daily ratio of passenger types were identified as model features for predicting the daily usage of each type of public transportation. We tested the proposed model using an open data set from Izmir City Turkey. While the model had superior prediction performance the explanations showed that the type of public transportation weekday and the ratio of full-fare passengers have the highest SHAP values and the model features have many interactions. We also validated our results using an online data set showing Google search trends.
Citation - WoS: 2
Citation - Scopus: 2
LLM-based embeddings for clustering and predicting integrated reporting quality levels of companies
(Springer Science and Business Media B.V., 2025) Mert Sarioglu; Gorkem Sariyer; Mert Erkan Sözen; Sarioglu, Mert; Sariyer, Gorkem; Sozen, Mert Erkan
Artificial Intelligence (AI) offers various useful functions and algorithms that provide numerous benefits for firms to enhance their decision-making process. Moreover with the adoption of Integrated Reporting (IR) reporting practices which are critical communication channels for companies have become more practical. Given the importance of subjects it is believed that addressing LLM embeddings based AI methodologies will contribute positively to IR quality (IRQ) to achieve better results. Additionally grouping companies according to their IRQ characteristics will lead time and cost efficiency in decision-making. So that the main purpose of this study is to cluster companies with respect to their IRQ characteristics based on LLM embeddings and to use this grouping in further decision-making. This paper therefore provides significant evidence whether LLM is useful tool of AI techniques in IR practices and LLM-based clustering is an efficient way of generating predictions for decision-making. To do so the sample size of the study consists of 260 published IR in 2019. This study also introduces a novelty to the literature on the applicability of LLM with small data sets considering that the number of integrated reports published in a year is low or when the sample considered will be small. The findings reveal the superiority of LLM while indicating the usefulness of LLM in prediction of IRQ regarding different indicators of firms. Given the empirical evidence shown the techniques and steps should be adapted by firms both in identifying ways to improve IRQ and in different AI applications in the future. © 2025 Elsevier B.V. All rights reserved.
Citation - WoS: 1
Citation - Scopus: 1
Machine Learning Implementations for Multi-class Cardiovascular Risk Prediction in Family Health Units
(Ram Arti Publishers, 2023) Mert Erkan Sözen; Gorkem Sariyer; Mustafa Sözen; Gaurav Kumar Badhotiya; Lokesh Vijavargy; Sariyer, Gorkem; Sozen, Mert Erkan; Vijavargy, Lokesh; Badhotiya, Gaurav Kumar
Cardiovascular disease (CVD) risk prediction plays a significant role in clinical research since it is the key to primary prevention. As family health units follow up on a specific group of patients particularly in the middle-aged and elderly groups CVD risk prediction has additional importance for them. In a retrospectively collected data set from a family health unit in Turkey in 2018 we evaluated the CVD risk levels of patients based on SCORE-Turkey. By identifying additional CVD risk factors for SCORE-Turkey and grouping the study patients into 3-classes "low risk" "moderate risk" and "high risk" patients we proposed a machine learning implemented early warning system for CVD risk prediction in family health units. Body mass index diastolic blood pressures serum glucose creatinine urea uric acid levels and HbA1c were significant additional CVD risk factors to SCORE-Turkey. All of the five implemented algorithms k-nearest neighbour (KNN) random forest (RF) decision tree (DT) logistic regression (LR) and support vector machines (SVM) had high prediction performances for both the K4 and K5 partitioning protocols. With 89.7% and 92.1% accuracies for K4 and K5 protocols KNN outperformed the other algorithms. For the five ML algorithms while for the "low risk" category precision and recall measures varied between 95% to 100% "moderate risk" and "high risk" categories these measures varied between 60% to 92%. Machine learning-based algorithms can be used in CVD risk prediction by enhancing prediction performances and combining various risk factors having complex relationships. © 2023 Elsevier B.V. All rights reserved.
Citation - WoS: 22
Citation - Scopus: 29
Predictive and prescriptive analytics for ESG performance evaluation: A case of Fortune 500 companies
(Elsevier Inc., 2024) Gorkem Sariyer; Sachin Kumar Kumar Mangla; Soumyadeb Chowdhury; Mert Erkan Sozen; Yigit Kazancoglu; Erkan Sozen, Mert; Kumar Mangla, Sachin; Kazancoglu, Yigit; Sariyer, Gorkem; Sozen, Mert Erkan; Chowdhury, Soumyadeb; Mangla, Sachin Kumar
Given the growing importance of organizations’ environmental social and governance (ESG) performance studies employing AI-based techniques to generate insights from ESG data for investors and managers are limited. To bridge this gap this study proposes an AI-based multi-stage ESG performance prediction system consolidating clustering for identifying patterns within ESG data association rule mining for uncovering meaningful relationships deep learning for predictive accuracy and prescriptive analytics for actionable insights. This study is grounded in the big data analytics capability view that has emerged from the dynamic capabilities theory. The model is validated using an ESG dataset of 470 Fortune listed 500 companies obtained from the Refinitiv database. The model offers practical guidance for decision-makers to maintain or enhance their ESG scores crucial in a business landscape where ESG metrics significantly affect investor choices and public image. © 2024 Elsevier B.V. All rights reserved.
Citation - WoS: 2
Citation - Scopus: 2
The Power of Governments in Fight Against COVID-19: High-Performing Health Systems or Government Response Policies?
(WALTER DE GRUYTER GMBH, 2023) Gorkem Sariyer; Mert Erkan Sozen; Mustafa Gokalp Ataman; Ataman, Mustafa Gokalp; Sariyer, Gorkem; Sozen, Mert Erkan
Due to the pandemic situation caused by COVID-19 disease there have been tremendous efforts worldwide to keep the spread of the virus under control and protect the functioning of health systems. Although governments take many actions in fighting this pandemic it is well known that health systems play an undeniable role in this fight. This study aimed to investigate the role of health systems and government responses in fighting COVID-19. By purposively sampling Finland Denmark the UK and Italy and analyzing their health systems' performances governments' stringency indexes and COVID-19 spread variables this study showed that high-performing health systems were the main power of states in managing pandemic environments. This study also measured relations between short and medium-term measures and COVID-19 case and death numbers in all study countries. It showed that medium-term measures had significant effects on death numbers.