The goal of this thesis is to evaluate the skills mismatch in the curricula of HPC MSc's programs for some of the most relevant positions in the HPC areas in light of the ESCO database criteria and industry requirements. The four separate profiles "Data Science, Computer Architecture, Parallel Programming, and DevOps" that are thought to be crucial in HPC are the focus of this research. The methodology of this contribution explicitly examines the key responsibilities for the aforementioned positions and conducts a gap analysis based on Natural Language Processing (NLP) techniques for the competencies required for each in the MSc degree curriculum. The goal of applying NLP is to determine the degree to which the occupational skills stated in the ESCO and the HPC graduate courses’ syllabuses are semantically similar, as well as the level at which they overlap.

Date: 16.01.2023 / 13:00 Place: B-116

Communicating uncertainty is a challenging task. Personal differences such as culture, cognitive load and even feelings of the user can impact the interpretation of an uncertain situation. Decision support models such as Bayesian networks can aid dealing with uncertainty. However, the outputs of these models are probability distributions, their interpretation can be challenging and visualisations can help with this task. The aim of the study is to investigate how effective visual aids communicating BN predictions are and users preferences regarding these visual aids. Model’s prediction and performance are needed to communicate BN predictions. Hence, in this study both model’s prediction and performance are investigated. BN model that has been developed to make predictions on a medical condition namely, Trauma Induced Coagulopathy (TIC BN) are used as a case study.

Date: 16.01.2023 / 14:00 Place: A212

In this thesis, we proposed new region proposal network to eliminate the traditional region proposal network’s disadvantages. Region Proposal Networks are used for generating object candidate boxes for two stage object detection algorithms to detect objects with higher accuracy rate on Deep Learning area. With this proposed method, the accuracy of the classical methods is passed on MS-COCO dataset.

Date: 19.01.2023 / 13:30 Place: -

The emerging destructive Technologies such as big data, internet of things (IoT), cloud have revealed the Industry 4.0 revolution. Although Industry 4.0 technologies have very important benefits to the manufacturing sector, various obstacles may arise against the application of these technologies. The aim of this thesis is to make detailed studies on the Industry 4.0 revolution and to reveal the obstacles that may arise during the application of Industry 4.0 technologies to the Turkish manufacturing sector. Interpretive Structural Modeling (ISM) technique was used while establishing the relationship between the barriers in front of IE 4.0 adaptation. ISM Framework was obtained with the findings obtained as a result of the study.

Date: 25.01.2023 / 10:30 Place: A212

Recent developments in IT and Analytics have created a multitude of technologies for analytics professionals to use. This technology variety complicates the technology selection process. When improper technologies are selected, analytics teams face problems, inefficiencies, and technical debt. To solve this problem, this thesis proposes a systematic technology selection approach and tests it on a case study.

Date: 29.12.2022 / 14:00 Place: A-108

With the widespread use of cloud computing in software development processes, many industries have started to use cloud computing. Within the scope of this study, a framework has been created to enable the use of cloud computing in the defense industry. While creating the framework, all the constraints required by the defense industry were complied with and the appropriate cloud structure was selected. At the end of the study, the use of the framework was confirmed by the defense industry employees.

Date: 19.01.2023 / 10:00 Place: ODTÜ Enformatik Enstitüsü

This research aims to enhance the perceptual quality of adversarial attacks, regardless of the type of attack used. To achieve this, the study proposes an integration of two seperate attack independent techniques. The first technique, called Normalized Variance Weighting, applies a variance map to intensify the perturbations in high variance areas. The second technique, called the Minimization Method, minimizes the perceptual distance between the benign and adversarial example iteratively. Integrated Method, with the first method applied during the attack and the second method applied after the attack has quantifically the best visual quality.

Date: 23.12.2022 / 10:00 Place: A108

This thesis solves the 3D mesh segmentation problem from a different perspective. This perspective contains a data-centric approach with unsupervised learning. This means that 3D mesh segmentation mapping plays a key role in terms of a data-centric approach. We provide a mapping architecture using the node2vec model that also solves the curse of dimension and transforms 3D mesh data into an embedding vector. Segmentation was obtained using the K-Means clustering algorithm using this embedding vector. Additionally, we provide a new strategy that evaluates the graph embedding vector and a new inertia method calculated on 3D mesh data, geodesic inertia.

Date: 02.12.2022 / 14:00 Place: Computer Engineering - A105

Genomic sequencing technologies opened a new era for genetic disorder diagnostics. ​Currently, in over 80% of the cases, the genetic etiology of the diseases can be ​determined by identifying the causative variations. In this study, we have utilized two next-generation sequencing technologies, Ion-Torrent and Illumina, for a rare disease family with two siblings sharing similar symptoms. In addition to comparing technologies, different assemblies of human reference assemblies are analyzed and the benefits of all are discussed for revealing the variants of unsolved rare disease cases. 

Date: 29.11.2022 / 12:30 Place: A108

The main purpose of this thesis is to investigate the potential effects of the gaze bahavior modeling. To that end, experiments were conducted with human participants in the form of a mock-up job interview setting, where a robot avatar programmed with pre-planned gaze behavior in reference to the model was in charge as the interviewer. The eye-tracking data of the participants were collected and analyzed. The eye-tracking results were compared with traditional Human-Robot Interaction survey data along with the TF-IDF (Term Frequency-Inverse Document Frequency) analysis of the post-experimental oral evaluations.

Date: 29.11.2022 / 16:00 Place: B116