Duygu Bagci Das

Dr. Duygu Bagci Das


Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

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

Representative Publications

View All Publications
3 January 2023 in Journal of the Brazilian Society of Mechanical Sciences and Engineering
Verified
Journal Article
Closed access
PDF file
Oguzhan DasOguzhan Das
Oguzhan Das
DDDuygu Bagci Das
Duygu Bagci Das


7 June 2024 in Journal of Vibration Engineering & Technologies
Verified
Journal Article
Open access
PDF file
DDDuygu Bagci Das
Duygu Bagci Das
Oguzhan DasOguzhan Das
Oguzhan Das
Purpose: As the concept of Industry 4.0 is introduced, the significance of Smart Fault Diagnosis in the industry is increased. Therefore, it is essential to develop accurate, robust, and lightweight intelligent fault diagnosis approach that can be executed in real-time even with embedded systems. Additionally, it is preferable to use a single method for multi-purposes such as the fault detection, identification, and severity assessment. This study proposed a new approach called GaBoT for fault diagnosis of rotating machinery to satisfy those requirements. Method: The proposed approach adopted the concept of the ensemble of ensembles by boosting random forest. The statistical features of discrete wavelet transform were considered since they are easy and fast to obtain. Model optimization was conducted by employing genetic algorithm to alleviate the computational load without decreasing the model performance. The proposed approach has been validated by unseen data from an experimental dataset including shaft, rotor, and bearing faults. Results: The experimental results indicate that the proposed approach can effectively find the fault type with 99.85% accuracy. Besides, it successfully determines the fault severity by accuracy values between 96.45 and 99.72%. GABoT can also determine the imbalance severity in the presence of three bearing faults. Conclusion: Employing GA eliminated most of the redundant features and reduced the model execution time consumption. The results yielded that GABoT is a highly accurate model, and can be utilized in real-time fault diagnosis of rotating machinery.

28 February 2023 in Expert Systems
Verified
Journal Article
Open access
DDDuygu Bagci Das
Duygu Bagci Das
DBDerya Birant
Derya Birant
Human activity recognition (HAR) is the process of classifying a person's actions, and it is an essential task for many human‐centered applications. Multi‐instance learning (MIL) is a special case of machine learning where the training examples are bags containing many instances, and a single class label is assigned for an entire bag of instances. In this study, we integrated these two concepts by introducing a novel approach: “human activity recognition based on multi‐instance learning”, called HAR‐MIL. Unlike previous studies, the proposed HAR‐MIL method represents human activities differently: as a bag of various wearable sensors (gyroscope, magnetometer, accelerometer, and linear acceleration). HAR‐MIL presents an applicable and flexible model by providing multi‐instance representation and eliminating the restrictions of traditional single‐instance representation. Therefore, the adverse effects of missing data, defective sensors, and biased measurement on activity classification performance were minimized. This study is the first to investigate the performance of two MIL algorithms (SimpleMI and MIWrapper) on HAR. In this study, we explored the effect of four main factors (sensor positions, sensor types, base learners, and single or multiple participants) on the multi‐instance representation of relevant human daily activities. The effectiveness of the proposed HAR‐MIL method was demonstrated on 50 participant‐based and sensor‐position‐based activity recognition datasets. The experimental results showed that HAR‐MIL is effective for wearable sensor‐based HAR with high classification accuracy (99.32%). Furthermore, the results showed that the proposed method outperformed the state‐of‐the‐art methods by 10% on average on the same dataset.

23 September 2021 in Turkish Journal of Electrical Engineering and Computer Sciences
Verified
Journal Article
Closed access
DDDuygu Bagci Das
Duygu Bagci Das
Derya BirantDerya Birant
Derya Birant

3 December 2025 in Nondestructive Testing and Evaluation
Verified
Journal Article
Closed access
Oguzhan DasOguzhan Das
Oguzhan Das
DDDuygu Bagci Das
Duygu Bagci Das
HOHasan Ozturk
Hasan Ozturk