Aylin Adem

Aylin Adem

Auto-generated profile, based on publicly available information

Aylin Adem

Department of Industrial Engineering, Gazi University

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 April 2020 in SN Applied Sciences
Verified
Journal Article
Open access
PDF file
AAAylin Adem
Aylin Adem
Erman ÇakıtErman Çakıt
Erman Çakıt
MDMetin Dağdeviren
Metin Dağdeviren
The objectives of this study were three-fold: (i) to investigate new occupational health and safety risks that may arise with the integration of Industry 4.0 in production environments; (ii) to identify and group these risks into various categories; and (iii) to prioritize these groups. With this prioritization, a number of recommendations will be made to take into account new risk groups that may arise during the reorganizing of production environments according to Industry 4.0. This study used the Hesitant Fuzzy Analytic Hierarchy Process prioritization approach to determine the ranking of risks, which arose with the use of Industry 4.0. In this paper, possible occupational health and safety risks of the selected method consisted of eye-related disorders, mental fatigue, disorders stemming from the static working position, exposure to unknown dangerous particles as a result of cooperation with robots, and psychological pressure (stemming from the problems of adaptation to duties that require creativity). According to the results, the two most important risk factors were mental fatigue and psychological pressure. Mental fatigue as the principal risk factor was not unexpected owing to the many encounters between humans and machines. However, with respect to psychological pressure, we were unable interpret the results in the same way as the mental fatigue case. Psychological pressure is a complicated problem, and a more in-depth and detailed investigation is needed.

1 July 2018 in Safety Science
Verified
Journal Article
Closed access
AAAylin Adem
Aylin Adem
AÇAli Çolak
Ali Çolak
MDMetin Dağdeviren
Metin Dağdeviren


1 March 2016 in Concussion
Verified
Journal Article
Open access
USUzma Samadani
Uzma Samadani
Meng LiMeng Li
Meng Li
MQMeng Qian
Meng Qian
ELEugene Laska
Eugene Laska
RRRobert Ritlop
Robert Ritlop
CMCharles Marmar
Charles Marmar
Object: The purpose of the current study is to determine the sensitivity and specificity of an eye tracking method as a classifier for identifying concussion. Methods: Brain injured and control subjects prospectively underwent both eye tracking and Sport Concussion Assessment Tool 3. The results of eye tracking biomarker based classifier models were then validated against a dataset of individuals not used in building a model. The area under the curve (AUC) of receiver operating characteristics was examined. Results: An optimal classifier based on best subset had an AUC of 0.878, and a cross-validated AUC of 0.852 in CT- subjects and an AUC of 0.831 in a validation dataset. The optimal misclassification rate in an external dataset (n = 254) was 13%. Conclusion: If one defines concussion based on history, examination, radiographic and Sport Concussion Assessment Tool 3 criteria, it is possible to generate an eye tracking based biomarker that enables detection of concussion with reasonably high sensitivity and specificity.

1 January 2016 in Procedia Computer Science
Verified
Journal Article
Open access
AAAylin Adem
Aylin Adem
MDMetin Dağdeviren
Metin Dağdeviren
The selection of life insurance policy is generally one of the most important and complicate issue in real life. Because there are a lot of alternatives and criteria related to this subject. As in all types of selection problems, to solve of this, any of multi criteria decision making methods (MCDM) can be used. However, since decision maker couldn’t decide superiorities of alternatives and criteria, using classical multi criteria decision making methods to solve of this problem, gained results may not be accurate. Because, expert or decision maker may hesitate between different linguistic term and they need richer expression to express their knowledge. Hesitant fuzzy linguistic model which is novel MCDM methods, come into prominence in this way. Hesitant Fuzzy Linguistic Term Set permit decision maker to express their knowledge more correctly. In this paper we propose a hierarchical hesitant fuzzy linguistic model that contains hesitant linguistic evaluations of multiple experts on multiple criteria for life insurance policy alternatives. In this study, choosing one of the three life insurance policy alternatives has been studied. The main criteria of this problem are company reliability, customer relationship, the scope of insurance, insurance price, easiness of give up insurance. Three alternatives were evaluated based on these main criteria and their sub-criteria used HFLTS by two decision makers. As a result of the study the most suitable alternative has been selected on the basis of preferences of decision-makers

22 September 2025 in IEEE Access
Verified
Journal Article
Open access
MAMeltem Aksoy
Meltem Aksoy
AAAylin Adem
Aylin Adem
MDMetin Dağdeviren
Metin Dağdeviren
As Large language models (LLMs) become increasingly integrated into high-stakes applications, ensuring their trustworthiness has emerged as a critical research concern. This study proposes a novel evaluation framework that applies a multi-criteria decision making (MCDM) methodology, specifically the hesitant fuzzy analytic hierarchy process (AHP), to assess and rank LLMs based on five key trust dimensions: fairness, robustness, integrity, explainability, and safety. Drawing from expert evaluations, the framework systematically determines the relative importance of each criterion and applies a weighted scoring approach to compare seven leading LLMs, including both proprietary models such as GPT-3.5, GPT-4o, Claude 3.5 Sonnet, Gemini 1.5 and open-source models such as Llama 3.1, Mistral Large 2 and DeepSeek V3. Results reveal GPT-4o as the most trustworthy model, significantly outperforming its peers, particularly in robustness and fairness. Open-source models showed lower scores, especially in safety and explainability, highlighting persistent gaps in their alignment with trust expectations. The findings demonstrate the effectiveness of MCDM in capturing expert uncertainty and prioritizing trust criteria, offering a robust and adaptable framework for evaluating LLMs in dynamic and sensitive domains.