İnsan aktivitesi tanımada derin öğrenme yöntemlerinin uygulanması
Is this your thesis?
This record came from a bulk archive import. If it’s yours, link it to your profile.
Abstract (EN)
Traditional sensor-based human activity recognition (HAR) has been defined as a time-series data classification problem and requires feature extraction. The current HAR systems still lack transparent, interpretable, and explainable approaches that can generate human-understandable information. This thesis proposes an approach, called Human Activity Recognition on Signal Images (HARSI), which defines the HAR problem as an image classification problem to improve both explainability and recognition accuracy. The proposed HARSI approach transforms the smart sensor data into some visual images to take advantage of the strengths of convolutional neural networks (CNN) in handling image data. The experimental results carried out on a real-world dataset showed that a significant improvement was achieved by the proposed HARSI model compared to the traditional machine learning models. The results also showed that our method outperformed the state-of-the-art methods in terms of classification accuracy.
Author
Kemal Baysarı
How to Cite
Kemal Baysarı (Master Thesis). İnsan aktivitesi tanımada derin öğrenme yöntemlerinin uygulanması, 2022, Dokuz Eylül University.
License
Tüm Hakları Saklıdır
This work is shared under the specified license terms.
More theses from Dokuz Eylül University
- Synthesis, characterization and applications of polyphenol-Fe(III) complexes and tannic acid resin(2013)
- Effect of employee perceived power bases on organizational commitment: A study(2009)
- Lazer pattern creation techniques and applications on ceramic surface(2018)
- The effect of pumice and zeolite on alkali silica reaction(2006)
- Foça kule evleri(1997)
- The effect of affective comorbidity on functioning and neurocognition in the schizophrenia spectrum(2023)
