Seminars

Sıtar Kortik’s “Domain Independent Task Planning: Linear Planning Logic (LPL) and LinGraph” seminar is on December 1st 2016 at Faculty of Engineering Seminar Room at 15:30pm
Dr. Hüseyin Gökhan Akçay’s “Automatic Detection of Compound Structures by Joint Selection of Region Groups from Multiple Hierarchical Segmentations” seminar is on 17 November 2016 at Faculty of Engineering Seminar Room at 15:30pm
Prof.Dr. Hasan Erbay’s “Matrix Decomposition” seminar is on 21 January 2016 at Faculty of Engineering Seminar Room at 10:00am
Dr. Melih Onuş’s “Publish-Subscribe Overlay Network Design” seminar is on 24 December 2015 at Faculty of Engineering Seminar Room at 14:30pm
Dr. Mustafa Berkay Yılmaz’s “Offline Signature Verification with User-based and Global Classifiers” and “Facial Feature Extraction” seminar is on 26th November 2015 at Faculty of Engineering Seminar Room at 14:30pm
Dr. İzzet Şentürk’s “Connectivity in Partitioned Wireless Sensor Networks” seminar is on 04th August 2015 at EEE Seminar room at 14:00 pm
Prof. Moncef Gabbouj’s “Machine Learning and Optimization Tools for Multimedia Big Data Search” and “Object Segmentation by Quantum CUTS” seminar is on 29 June 2015 at Engineering Faculty at 14:00 pm
Dr. Mustafa Yaman’s “Multi-Modal Stereo-Vision & Hysperspectral Image Analysis” seminar is on 26 June 2015 at Engineering Faculty at 14:00 pm
Dr. Joseph Ledet’s “A Model Transformation Process for Simulation Modernization and Reuse” seminar is on 14 May 2015 at Engineering Faculty at 11:00 am
Dr. Hüseyin Kusetoğulları’s “Transmission of Multimedia Information and Meta-heuristic Algorithms” seminar is on 1st Dec 2014 at Engineering Faculty at 14:00 pm
Dr. Taner Danışman’s “Appearance-based gender recognition under unconstrained settings” seminar is on 28 Nov 2014 at Engineering Faculty at 14:00 pm
Dr. Melih Günay seminar “Building a Bioinformatics Solution for TaqMan Array Card (TAC) Data Management” seminar is on 14 May 2014 at Medical School 10:00 am


01.12.2016 – 15:30 – at Faculty of Engineering Seminar Room

Biography – Sıtar Kortik

Sıtar Kortik is a Ph.D. candidate in Computer Engineering Department, Bilkent University. He received his B.S. in Computer Engineering from Dokuz Eylül University (2006) and his M.S. degree in Computer Engineering from Bilkent University (2010). He was a Visiting Researcher at the Computer Science Department, Carnegie Mellon University, USA in 2012, where he introduced a novel framework for task planning problems based on a graph-based theorem prover for a fragment of linear logic. His research interests include artificial intelligence, automated reasoning, robotic task planning, applications of linear logic and theorem proving to domain independent task planning and multiagent systems.

Abstract – Domain Independent Task Planning: Linear Planning Logic (LPL) and LinGraph

Linear Logic is a non-monotonic logic, with semantics that enforce single-use assumptions thereby allowing native and efficient encoding of domains with dynamic state. Robotic task planning is an important example for such domains, wherein both physical and informational components of a robot’s state exhibit non-monotonic properties. We use linear logic to encode deterministic STRIPS planning problems in general, and visual robot navigation in particular. We propose a novel and efficient method for automated construction of proofs for this language, which we call the Linear Planning Logic (LPL). We also introduce LinGraph, an automated planner for deterministic, concurrent domains, formulated as a graph-based theorem prover for a propositional fragment of intuitionistic linear logic. LinGraph substantially improves planning performance by reducing proof permutations that are irrelevant to planning problems particularly in the presence of large numbers of objects and agents with identical properties (e.g. robots within a swarm, or parts in a large factory).


17.11.2016 – 15:30 – at Faculty of Engineering Seminar Room

Biography – Dr. Hüseyin Gökhan Akçay

Dr. Hüseyin Gökhan Akçay received the B.S., M.S. and Ph.D. degrees in Computer Engineering from Bilkent University, Ankara, Turkey, in 2004, 2007 and 2016, respectively. During his M.S. and Ph.D. studies, he was involved in projects aiming at mining of very high spatial resolution remote sensing images sponsored by TÜBİTAK and the European Commission. He was a Visiting Researcher at the Institute for the Protection and Security of the Citizen, European Commission Joint Research Centre, Ispra, Italy in 2010, where he collaborated in a project for extraction of a global built-up structure layer. His research interests include statistical and structural pattern recognition, computer vision and machine learning for the analysis of remote sensing and medical images.

Abstract – Automatic Detection of Compound Structures by Joint Selection of Region Groups from Multiple Hierarchical Segmentations

A challenging problem in remote sensing image interpretation is the detection of heterogeneous compound structures such as different types of residential, industrial, and agricultural areas that are comprised of spatial arrangements of simple primitive objects such as buildings and trees. We describe a generic method for the modeling and detection of compound structures that involve arrangements of unknown number of primitives appearing in different primitive object layers in large scenes. The modeling process starts with example structures, considers the primitive objects as random variables, builds a contextual model of their arrangements using a Markov random field, and learns the parameters of this model via sampling from the corresponding maximum entropy distribution. The detection task is reduced to the selection of multiple subsets of candidate regions from multiple hierarchical segmentations corresponding to different primitive object layers where each set of selected regions constitutes an instance of the example compound structures. The combinatorial selection problem is solved by joint sampling of groups of regions by maximizing the likelihood of their individual appearances and relative spatial arrangements under the model learned from the example structures of interest. Moreover, we incorporate linear equality and inequality constraints on the candidate regions to prevent the co-selection of redundant overlapping regions and to enforce a particular spatial layout that must be respected by the selected regions. The constrained selection problem is formulated as a linearly constrained quadratic program that is solved via a variant of the primal-dual algorithm called the Difference of Convex algorithm by re-writing the nonconvex program as the difference of two convex programs. Extensive experiments using very high spatial resolution images show that the proposed method can provide good localization of unknown number of instances of different compound structures that cannot be detected by using spectral and shape features alone.


21.01.2016 – 10:00 – at Faculty of Engineering Seminar Room

Biography – Prof.Dr. Hasan Erbay

Hasan Erbay was born in Ankara in 1968. He completed his Ph.D. in Computer Science & Engineering and M.S. in Mathematics at Penn State University, USA and his undergraduate studies in Mathematics at Middle East Technical University. Currently, Erbay is a Professor in the Department of Computer Engineering at the Kırıkkale University where he has been a faculty member since 2009.
His research interests lie in the areas of numerical linear algebra and numerical computation, ranging from theory to design to implementation, with a focus on improving software quality. He has collaborated actively with researchers in several other disciplines of computer science, particularly educational data mining to analyze the a ects of information technologies in higher education.
Erbay is married and the father of two sons, born February 1996 and April 2003.

Abstract – Matrix Decomposition

Among the most common tools in computer science is the rectangular array of numbers known as matrices. The numbers in a matrix can represent data, and they can also represent mathematical equations. Matrices arose originally as a way to describe systems of linear equations, a type of problem familiar to anyone who took high school algebra. Linear just means that the variables in the equations don’t have any exponents, so their graphs are straight lines. The first known example of matrix dates back to 300 BC to solve the Chinese problem: Three bundles of a good crop, two bundles of a ordinary crop, and one bundle of a bad crop are sold for 39 dou. Two bundles of good, three ordinary, and one bad are sold for 34 dou; and one good, two ordinary, and three bad are sold for 26 dou. What is the price received for each bundle of a good crop, each bundle of a ordinary crop, and each bundle of a bad crop? The problem can be represented as a system of linear equations [1]. In a range of applications from image processing to latent semantic analysis, computers are often used to solve systems of linear equations – usually with many more than two variables. Even more frequently, they’re used to multiply matrices or to decompose (product of matrices) them.


24.12.2015 – 14:30 – at Faculty of Engineering Seminar Room

Biography – Dr. Melih Onuş

Dr. Melih Onuş received his Ph.D. degree from the Department of Computer Science and Engineering at Arizona State University in 2009. He earned B.S. degree in Computer Engineering at Bilkent University in 2003. His research interests are in the areas of distributed computing, computer networks and algorithms. He worked as an assistant professor at Department of Computer Engineering, Cankaya University for four years.

Abstract – Publish-Subscribe Overlay Network Design

Designing an overlay network for publish/subscribe communication in a system where nodes may subscribe to many different topics of interest is of fundamental importance. For scalability and efficiency, it is important to keep the degree of the nodes in the publish/subscribe system low. It is only natural then to formalize the following problem: Given a collection of nodes and their topic subscriptions connect the nodes into a graph which has least possible maximum degree and in such a way that for each topic t, the graph induced by the nodes interested in t is connected. We present the first polynomial time logarithmic approximation algorithm for this problem and prove an almost tight lower bound on the approximation ratio. Our experimental results show that our algorithm drastically improves the maximum degree of publish/subscribe overlay systems. We also propose a variation of the problem by enforcing that each topic-connected overlay network be of constant diameter, while keeping the average degree low. We present a heuristic for this problem which guarantees that each topic-connected overlay network will be of diameter 2 and which aims at keeping the overall average node degree low. Our experimental results validate our algorithm showing that our algorithm is able to achieve very low diameter without increasing the average degree by much.


26.11.2015 – 14:30 – at Faculty of Engineering Seminar Room

Biography – Dr. Mustafa Berkay Yılmaz

Mustafa Berkay Yılmaz was born in Antalya in 1985. He was the winner of Tübitak National Computer Olympics in 2001 over the Mediterranean region. He completed his B.S. degree in Computer Engineering at Bahçeşehir University between 2003 and 2007. He was the winner of Tübitak Bilim ve Teknik magazine annual programming contest in 2006. He completed his M.S. degree in Mechatronics Engineering at Sabancı University between 2007 and 2009. During his M.S. study, he was a researcher in a project about audio-visual speech recognition, supported by Tübitak. He studied offline signature verification during his Ph.D. in Computer Science and Engineering at Sabancı University between 2009 and 2015. His offline signature verification system won several international signature verification competitions. He was a teaching assistant during his M.S. and Ph.D. studies. He worked as a researcher in a collaborative project between Boğaziçi University and Yapı Kredi Bank on document classification, between 2013 and 2014. He worked as an engineer at Sabancı University, in an online handwriting recognition project in 2015. He continued as a post-doctoral researcher at Telecom SudParis, studying pathology detection from speech, supported by French Embassy in Turkey. He will work as a post-doctoral researcher at École de Technologie Supérieure in Montreal in 2016, granted by Tübitak.

Abstract 1 – Offline Signature Verification with User-based and Global Classifiers

Signature verification deals with the problem of identifying forged signatures of a user from his/her genuine signatures. The difficulty lies in identifying allowed variations in a user’s signatures, in the presence of high intra-class and low interclass variability (the forgeries may be more similar to a user’s genuine signature, compared to his/her other genuine signatures). Signature is considered a behavioral biometric and the problem possesses further difficulties compared to other modalities (e.g. fingerprints) due to the added issue of skilled forgeries. A complete offline (image-based) signature verification system will be introduced in this talk. In order to capture the signature’s stable parts and alleviate the difficulty of global matching, local features based on gradient information and neighboring information are utilized. Scale invariant feature transform (SIFT) matching is also used as a complementary approach. Two different approaches for classifier training are investigated, namely global and user-dependent SVMs. User-dependent SVMs, trained separately for each user, learn to differentiate a user’s (genuine) reference signatures from other signatures. On the other hand, a single global SVM trained with difference vectors of query and reference signatures’ features of all users in the training set, learns how to weight the importance of different types of dissimilarities. The fusion of all classifiers achieves a 6.97% equal error rate in skilled forgery tests using the public GPDS-160 signature database. Former versions of the system have won several signature verification competitions. One of the major benefits of the proposed method is that user enrollment does not require skilled forgeries of the enrolling user, which is essential for real life applications. Pathology Detection from Speech Recently, speech analysis has been used for healthcare tasks such as pathology detection. Several pathologies have been considered so far, including physiological pathologies such as hoarseness, nodule, polyp, cyst, cancer, and neurodegenerative brain disorders such as Parkinson’s disease, Alzheimer disease, mild cognitive impairment. In contrast to speech recognition, content of the speech is often not of interest but characteristics like prosody are. In this talk, general approaches for the most studied pathologies will be presented.

Abstract 2 – Facial Feature Extraction

Facial features such as lip corners, eye corners and nose tip are critical points in a human face. Robust extraction of such facial feature locations is an important problem which is used in a wide range of applications including audio-visual speech recognition, human-computer interaction, emotion recognition, fatigue detection and gesture recognition. In this talk, a probabilistic method for facial feature extraction will be presented. This technique is able to learn location and texture information of facial features from a training set. Facial feature locations are extracted from face regions using joint distributions of locations and textures.


04.08.2015 – 14:00 – at EEE Seminar Room

Biography – Dr. İzzet Şentürk

Izzet Senturk is a former member of the High Performance Computing lab at the Department of Biomedical Informatics in the Ohio State University (OSU). As a post-doctoral researcher at the OSU, he conducted research on the analysis of the emergence, evolution, and spread of infectious diseases and high performance next-generation sequencing analysis on the cloud. His work at OSU was supported by Defense Threat Reduction Agency and Qatar National Research Fund. Before joining OSU, Izzet Senturk received his PhD degree in Computer Science from Southern Illinois University Carbondale (SIUC) in 2013. At SIUC, Izzet Senturk was a member of Advanced Wireless & Sensor Networking Lab where he worked on fault tolerance in wireless sensor networks. His work at SIUC was supported by National Science Foundation. Izzet Senturk received his master’s degree in Computer Science from Cornell University in 2008 where he worked on multi protocol label switching. His work at Cornell was supported by the Ministry of National Education. He is a member of IEEE. His areas of interest include bioinformatics, high performance computing, mobility and fault-tolerance in wireless sensor networks and energy-aware protocol design for mobile sensor networks.

Abstract – Connectivity in Partitioned Wireless Sensor Networks


29.06.2015 – 14:00 – at Faculty of Engineering Seminar Room

Biography – Prof. Moncef Gabbouj


Abstract 1 – Machine Learning and Optimization Tools for Multimedia Big Data Search

The talk deals with a new paradigm for multimedia search based on content. We present an alternative approach to classical search engines for information retrieval, which can be used for Big and generic multimedia repositories. We introduce an incremental evolution scheme within a collective network of (evolutionary) binary classifier (CNBC) framework. The proposed framework addresses the problems of feature/class scalability and achieves high classification and content-based retrieval performances over dynamic image repositories. The secret behind the success of CNBC is a novel design to implement the backbone of CNBC, namely the binary classifier. This is a special neural network, which is optimally designed using the recently developed evolutionary optimization algorithm called multi-dimensional particle swarm optimization.

Particle swarm optimization (PSO) is population based stochastic search and optimization process, which was introduced in 1995 by Kennedy and Eberhart. The goal is to converge to the global optimum of some multi-dimensional fitness function. Two novel techniques, which extend the basic PSO algorithm, are presented. The first algorithm called multi-dimensional PSO (M-D PSO) deals with problems in which the dimension of the solution space is not known a priori. M-D PSO solves such a problem by introducing two interleaved PSO iteration processes, a positional PSO followed by a dimensional PSO in which the dimension of a particle is allowed to vary. In a multidimensional search space where the optimum dimension is unknown, swarm particles can seek both positional and dimensional optima.

Most content-based multimedia search engines available today rely heavily on low-level features. However, such features extracted automatically usually lack discrimination power needed for accurate description of the image content and may lead to poor retrieval performance. To address this problem, we propose an evolutionary feature synthesis technique, which seeks for the optimal linear and non-linear operations over optimally selected features so as to synthesize highly discriminative features. The optimality therein is sought through MD-PSO. The synthesized features are applied over only a minority of the original feature vectors and exhibit a major discrimination power between different classes and extensive CBIR experiments show that a significant performance improvement can be achieved.

Abstract 2 – Object Segmentation by Quantum CUTS

Quantum-cut (QCUT) is an unsupervised, state of the art saliency map generation algorithm for both RGB and RGB-Depth images. QCUT elegantly combines principles from Quantum Mechanics and Spectral Graph Theory. QCUT forms a graph among superpixels extracted from an image and optimizes a criterion related to the image boundary, local-global contrast and area information jointly. QCUT stands out from the competing algorithms by its unique quantum-mechanics interpretation, which also provides the ability to propose several saliency maps with confidence scores. It consistently achieves the state-of-the-art performance over all benchmark saliency detection datasets, containing around 20k images in total as well as on a introduced RGB-Depth saliency dataset. QCUT can be easily applied to videos with a novel graph representation for videos which enables solving QCUT only once for the entire video instead of a frame-wise solution. QCUT is also quite fast and for an average image size of 400×500 it takes 0.85 seconds on average. For the videos, the running time is approximately the same per frame.

In this talk, we will briefly summarize QCUT algorithm and its main characteristics. Then we discuss its performance in comparison to other competing techniques for salient object extraction from images. We will also show some possible extension to image sequences (video signals). The talk will also highlight potential applications of QCUT in several fields.


26.06.2015 – 14:00 – at Faculty of Engineering Seminar Room

Biography – Dr. Mustafa Yaman


Abstract – Multi-Modal Stereo-Vision & Hysperspectral Image Analysis

Using multi-modal cameras for surveillance systems has been very popular for the last decade since using cameras of different modalities, such as a pair of infrared and visible cameras, has advantages over using unimodal cameras in surveillance systems. These advantages include being able to work under low visibility or lighting conditions, better segregation of a target from the background, allowing a richer set of information like thermal signatures in the scene or the different reflectance properties of objects in different bands of the electromagnetic spectrum, etc. In my thesis study, a method for computing disparity maps from a multi-modal stereo-vision system composed of an infrared–visible camera pair was proposed. The method uses mutual information (MI) as the basic similarity measure where a segment-based adaptive windowing mechanism is proposed along with a novel MI computation surface with joint prior probabilities incorporated. On an artificially-modified version of the Middlebury stereo-vision dataset and a Kinect device dataset that we created in this study, we showed that (i) our proposal improves the quality of existing MI formulation, and (ii) our method can provide depth comparable to the quality of Kinect depth data.

On the other hand, Hyperspectral Imaging Technologies are available in the market for a long time that enabled significant research in this field for especially along the last decade. However, the research performed in Turkey was too limited where SISP group in Havelsan Inc. has initiated a couple of leading research projects from target detection to improved agriculture techniques using these technologies.

In this second part of the talk, I will introduce the research projects that I am involved in where significant results are being obtained using hyperspectral imagery for especially target detection. The talk will include the details of what hyperspectral images are, how they are used for enhanced image analysis, remote sensing and target detection and the potential areas of further research opportunities that these technologies provide which are very widely ranging such as; environment & disaster monitoring and analysis, coastal mapping, agriculture, forestry, food industry, pharmacy, medicine, geology, mineral mapping and many more…


14.05.2015 – 11:00 – at Faculty of Engineering Seminar Room

Biography – Dr. Joseph Ledet


Abstract – A Model Transformation Process for Simulation Modernization and Reuse

Due to the increased use of simulation models in computational scientific experimentation, the need for reuse as well as continuous modernization and hence independent replication and reproduction of the models and results of experiment simulation executions has become essential. The desire to refrain from sharing all details of a simulation experiment can, and often does, come from legitimate and necessary reasons, such as a need to maintain security or intellectual property rights. However, the negative impact of not being transparent in these details has resulted in a decrease in credibility of the results obtained from such experiments.
In this presentation, Ledet details the developing of a process utilizing Model-Driven Engineering (MDE) principles, specifically Model Transformations, to produce Platform Independent Models (PIM) for the purpose of enhancing the reliability of the existing Platform Specific Models (PSM). Additionally, a second process could be concurrently developed to transform existing PIMs into PSMs for the purposes of executing these simulations to validate the results obtained in previous deployments. We attempt to address the current state of the development of this process, including Atlas Transformation Language (ATL) rules and Extensible Stylesheet Language Transformations (XSLT) examples. We address many of the lessons learned from developing a hybrid process using multiple transformation tools. Finally, we address the expectations of how this process will be expanded in our future work and how it can be integrated into an experiment management process for experiment simulation replicability and reproducibility.

In this second part of the talk, I will introduce the research projects that I am involved in where significant results are being obtained using hyperspectral imagery for especially target detection. The talk will include the details of what hyperspectral images are, how they are used for enhanced image analysis, remote sensing and target detection and the potential areas of further research opportunities that these technologies provide which are very widely ranging such as; environment & disaster monitoring and analysis, coastal mapping, agriculture, forestry, food industry, pharmacy, medicine, geology, mineral mapping and many more…


01.12.2014 – 14:00 – at Faculty of Engineering Seminar Room

Biography – Dr. Hüseyin Kusetoğulları


Abstract – Transmission of Multimedia Information and Meta-heuristic Algorithms

The transmission of multimedia information over communication channels/paths has become a challenging problem with the increased usage of multimedia services in networks. In multimedia communication, multi-path selection is desirable because even if one packet is lost over a path, the lost packet may be received via another path. In networks, each network link has more than one cost parameter such as packet loss rate, length and bandwidth as it makes the network optimization problem even harder. This has been a strong motivation to find the maximum number of optimal routings with available bandwidth in networks. In networks, the major problem is to obtain the optimal multi-paths efficiently and quickly, hence, two meta-heuristic optimization algorithms, namely Genetic Algorithms (GAs) and Particle Swarm Optimization (PSO) are used. Thereafter, images are transmitted through the optimized networks to analyze and understand the performance of the algorithms implemented. The simulations run over various random network topologies and the results show that the PSO algorithm finds optimal routings effectively and maximizes the received Multiple Descriptions.


28.11.2014 – 14:00 – at Faculty of Engineering Seminar Room

Biography – Dr. Taner Danışman

-Dr. Danışman studied Computer Engineering in Dokuz Eylul University (2000). He has an M.Sc (2002) and Ph.D. (2008) degrees in Computer Engineering from Dokuz Eylul University Graduate School of Natural and Applied Sciences. His Ph.D. thesis focuses on multimodal emotion recognition in video. During his master and Ph.D. he worked as a Research Assistant in Dokuz Eylül University Computer Engineering dept.(2000-2008). After getting his Ph.D., he was started to work for CNRS at the University Lille 1, Lille, France in 2008 and University of Angers ISTIA Lab. in 2011. He worked on several projects (ITEA2 MIDAS, ITEA2 TWIRL, PISE) dealing with emotion and gender recognition. His research interests are in the areas of face analysis, gender recognition, face and facial expression recognition with emphasis on multimedia indexing and multimedia information retrieval.

Abstract – Appearance-based gender recognition under unconstrained settings

This seminar provides an overview of cross-database evaluations of automatic appearance-based gender recognition methodology using normalized raw pixels and SVM classifier under unconstrained settings. Proposed method uses both histogram specification and feature space normalization on automatically aligned faces to achieve reliable recognition rate for real scenarios. Effect of age and It shows that aligned and normalized raw pixel intensities are providing the best performance in case of unconstrained cross-database tests than feature-based studies on unaligned faces. In addition, histogram specification provides better normalization than that of histogram equalization for automatically aligned faces in large databases for gender recognition. Variety of cross-database experiments performed on uncontrolled Image of Groups (88.16%), Genki-4K (91.07%) and LFW databases (91.87%) showed that proposed method provides superior generalization than baseline methods.


14.05.2014 – 10:00 am – at Medical School

Biography – Dr. Melih Günay


Abstract – Building a Bioinformatics Solution for TaqMan Array Card (TAC) Data Management

Recent pandemics increased the need for efficient diagnostic tools that can provide rapid results for a number of pathogens without compromising sensitivity and specificity. To address this, a number of multiple-pathogen detection systems for simultaneous, multiple, rapid pathogen detection have been developed. Among them, systems that are RT-PCR based using TaqMan low-density array cards (TAC) are demonstrated to be an effective surveillance tool for this purpose. However, when multiple labs that may be geographically distributed collect samples and execute TAC runs, it is essential to standardize and store TAC results and configuration parameters in a central database that can be accessed remotely. Such a design not only conveniently enables scientists to upload, share and compare ‘Run Results’ with other labs, but also enables epidemiologists to rapidly access and evaluate the results for in-depth investigation. Therefore, this paper describes the building blocks and development of an integrated data entry and analysis system that; i) stores TaqMan (RT-PCR) Array Card (TAC) experimental setup and run results, ii) enables epidomologist to retrieve and evaluate run results.

Mission

Create and disseminate knowledge by constituting a scholarly community focused on research and education in the area of Computer Science and Engineering. We will engage in inter-disciplinary research and impact all programs at Akdeniz University. Our outreach activities will fuel economic development and contribute to the health and well being of society.

Vision

Our vision is to be within the top 10 of computer science and engineering departments in Turkey. Our graduates will be leaders and innovators in industry, education, and government in Antalya, Turkey and Europe. The faculty will be recognized for the quality of their research and will influence the national and international agenda.

İletişim

Antalya Campus
Dumlupınar Boulevard 07058 Antalya / TURKEY
Tel: +90 242 227 4400 (internal 4361)
Tel: GPS:36°53’51.8″N 30°38’59.3″E
Translate »