Mühendislik ve Doğa Bilimleri Fakültesi

Project #

Project Title

Responsible Assistant

Responsible Project Coordinator

Related Topics

1

A java based network intrusion detection system (IDS)

Elif Gül

Dr.Baha ŞEN

Data Mining, Artificial Intelligence, Network Security.

2

SCIP

(Secure Communication Interoperability protocol)

Elif Gül

Dr.Baha ŞEN

Secure Communication

3

Generation of 3D Retinal Model using OCT Images

Melih Üçer

Dr.Baha ŞEN

Medical Imaging, Computer Graphics

4

Three Level Password Authentication System

Elif Yayla

Dr.Baha ŞEN

Identification Management Systems

5

VR to Socialize Autistic Kids

Ömer Mintemur

Dr. Özkan Kılıç

Android App Development, 3D Image Processing, Sound Processing, Computational Geometry

6

VoiceMate

Yusuf Şevki Günaydın

Dr. Özkan Kılıç

Voice Processing, Machine Learning, Natural Language Processing. Signal Processing

7

GazeNet

Yusuf Şevki Günaydın

Dr. Özkan Kılıç

Eye Tracking, NodeJS, Networking, Web Programming

8

StudyBuddy

İbrahim Atlı

Dr. Özkan Kılıç

Mobile Programming, Web Programming, Database Management

9

TrWordToVec

Ömer Mintemur

Dr. Özkan Kılıç

Natural Language Processing, Machine Learning, Linguistics

10

EyeLexia

İbrahim Atlı

Dr. Özkan Kılıç

Eye tracking, Web Programming, Psychology, Machine Learning

11

TV content Restriction

Ayşe Arslan

Dr. Hilal Kaya

Data- Mining, Machine Learning

12

Voice Control of Computer mouse

Ayşe Arslan

Dr. Hilal Kaya

Speech processing, signal processing

13

Body Simulation

Ayşe Arslan

Dr. Hilal Kaya

Graphical User İnterface, text processing

14

Smart Class Attendance

Melih Üçer

Dr. Shafqat Rehman

NFC/FRID, Algorithms, data management

15

Flow Classification and  Traffic Engineering (TE)  in cloud networks using SDN (Software Defined Networking)

Fatma Küçük

Dr. Shafqat Rehman

Computer Networking, Traffic Engineering,, Distributed Systems

16

Intrusion Detection System (IDS) using SDN (Softwar Defined Networking)

Fatma Küçük

Dr. Shafqat Rehman

Network Security, Software Defined Networking, Protocols, Distributed Systems, Operating Systems, Algorithms

17

Simulating Packet Routing in Communication Networks

Fatma Küçük

Dr. Shafqat Rehman

Computer Networking, Distributed Algorithms, Data Visualization

18

Aerial Surveillance using Unmanned Aerial Vehicle (UAV)

Hatice Çataloluk

Dr. Shafqat Rehman

Wireless Networking, IoT, Algorithms, Distributed Systems

19

Smart/Intelligent Home

Yusuf Sevki Günaydın

Dr. Tarek Najjar

Embedded Systems, Mobile Application, Windows

20

Object Searching Robot

Yusuf Sevki Günaydın

Dr. Tarek Najjar

Embedded Systems, Image Processing

21

Product Evaluation

Yusuf Sevki Günaydın

Dr. Rahime Belen Sağlam

Data Mining, Text Mining, Artificial Intelligence, Machine Learning

22

Implementation a Semantic Web Service by Using Open-Source Technologies

Ömer Mintemur

Asst. Prof. Mustafa Yeniad

Service-Oriented Architecture, Semantic Web, Open-Source Technologies.

23

Network Log Management and Analysis with ElasticSearch Technology

Ömer Mintemur

Asst. Prof. Mustafa Yeniad

Log Analysis, Elastic Search, Open-Source Technologies.

24

Implementing a Network Monitoring Service by Using Open Source Technologies

Ömer Mintemur

Asst. Prof. Mustafa Yeniad

Network Monitoring,
Open-Source Technologies.

25

Attendance Management System With Real Time Face Recognition

Hatice Çataloluk

Dr. Osman Serdar Gedik

Embedded Software Development, Pattern Recognition

26

Course and Exam Timetable Software using an Optimization Algorithm

Hatice Çataloluk

Dr. Özkan Kılıç

Artificial Intelligence, Optimization

27

Online Examination System

Hatice Çataloluk

Dr. Hilal Kaya

Web Programming, Database management

28

Automatic classification of hacker/cyber professional accounts on Twitter

Çağrı Burak Aslan

Dr. Rahime Belen Sağlam

Text Mining, Machine Learning, Natural Language Processing

29

3D Reconstruction of a/an room/office/place from multiple 2D images

İbrahim ATLI

Dr. Osman Serdar Gedik

Image Processing, Computer Vision, Robotics, Machine Intelligence

30

Tourist Guide Application

İbrahim ATLI

Dr. Abdullah Bülbül

Web application/Android, Information Retrieval, Text2Voice

31

Face Swapping

İbrahim ATLI

Dr. Abdullah Bülbül

Android, Image Processing, Computer Vision

32

Information Designer

Çağrı Aslan

Dr. Rahime Belen Sağlam

Information Retrieval, Data Mining

33

Cyber attack detection using social media

Çağrı Aslan

Dr. Rahime Belen Sağlam

Text Mining, Machine Learning, Natural Language Processing

34

Automated Hate Speech Recognition

Çağrı Aslan

Dr. Rahime Belen Sağlam

Text Mining, Opinion Mining

35

VR Reader for Islamic  Calligraphies

Elif Yayla

Dr. M. Abdullah Bülbül

Image Processing, Machine Learning, Mobile Programming

36

Jigsaw Puzzle Solver

Elif Gül

Dr. M. Abdullah Bülbül

Image Processing, Algorithms

37

Repository for Historic Buildings

Elif Yayla

Dr. M. Abdullah Bülbül

Computer Graphics, Databases, Image Processing

38

FPGA Implementation of ARM7TDMI Microprocessor

H. Gökçe Bilgiç

Dr. Tarek Najjar

FPGA.SoC, Microprocessor Systems

39

HDL-based Modeling of Memory Hierarchy System

Melih Üçer

Dr. Tarek Najjar

Computer Architecture, Hardware Description Languages, Simulation

40

Interactive Visual Simulation of Supervised Neural Networks

Melih Üçer

Dr. Tarek Najjar

Simulation, Neural Networks. Machine Learning

41

Clustering on Cloud

Elif Gül

Dr. Ahmet Ercan Topçu

42

Weather Data Search Engine using Deep Learning

Fatma Küçük

Dr. Ahmet Ercan Topçu

43

Cloud Services for Internet Documentation for the publications

Elif Yayla

Dr. Ahmet Ercan Topçu

44

Flow –based system implementation for Student Registration Systems.

H. Gökçe Bilgiç

Dr. Ahmet Ercan Topçu

Project #

Description

1

In this project, we designed and built an Intrusion Detection System (IDS) that implements pre-defined algorithms for identifying the attacks over a network. The Java programming language is used to develop the system, JPCap must be used to provide access to the winPcap. The packets in the network are captured online i.e., as they come on the interface of the network. The IDS are designed to provide the basic detection techniques to secure the systems present in the networks that are directly or indirectly connected to the internet.

2

The Secure Communication Interoperability Protocol (SCIP) is a communications standard developed by the National Security Agency (NSA) to enable interoperable secure communications among allies and partners around the globe.

The SCIP-210 Signaling Plan is the specification that defines the application layer signaling used to negotiate a secure end-to-end session between two communication devices, independent of network transport. SCIP negotiates the operational mode (e.g., voice, data, etc.), the cryptographic algorithm suite (e.g., Suite A, Suite B, etc), and the traffic encryption key used for each secure session. It also provides capabilities for cryptographic synchronization and operational mode control between communicating end-point devices. SCIP is designed to operate over any network and is currently utilized in devices operating on a wide variety of networks including PSTN, ISDN, CDMA, GSM, IP, and satellite.

3

Human Eye is the most complex sensory organ comprising of  a multitude of substructures. Retina, a mere set of tissues and one third of a millimeter in width, consists 10 layers. The deformities in these retinal layers account for many diseases related to vision such as retinopathy, age related macular degeneration, eye caner, retinal detachment and inherited retinal degenerations. Two major type of imaging techniques which are being used for analysis of retina are digital fundus images and Optical coherence tomography (OCT). OCT scan is a mean of extracting the information about the inner retinal layers non-invasively. OCT scans provided a video with multiple frames, each of which corresponds to a slice of eye (retina). The purpose of this project is to make a 3D model of retinal layers captured through OCT scans to provide a better visualization of retina to ophthalmologists. There are tow main parts of project which include image processing module and graphics. The frames gathered from OCT scans are processed using image processing techniques to extract the details of the layer sin the form of coordinates. Once the coordinates of all the  points are extracted a point cloud is formed. Using this point cloud, we regenerated a three dimensional model using these points

4

The project is an authentication system that validates user for accessing the system only when they have input correct password. The project involves three levels of user authentication. There are varieties of password systems available, many of which have failed due to bot attacks while few have sustained it but to a limit. In short, almost all the passwords available today can be broken to a limit. Hence this project is aimed to achieve the highest security in authenticating users.

5

It is known that autistic children have very little or no social interaction skills which worsens their cognitive development. It is also shown that they have affinity for computers, artificial environments and electronics. This project aims to create a virtual reality environment to help autistic kids to learn how to interact with artificial faces. The project requires Android app development skills. The 3D glasses will be provided by the professor. 

6

Biometric methods for improved computer security have gained great importance. With introduction of smart phones and household devices, it is important to distinguish identity of a speaker. This project involves using voice processing and machine learning techniques to identify a speaker from a voice signal. It also involves noise reduction and signal processing. MatLab knowledge is suggested.

7

Eye trackers are widely used by tech companies. Smart phones and computer will be enriched with eye trackers. This project aims to construct an online platform which enables multiple computers with eye trackers to be connected in a server-client architecture. The project involves web programming and NodeJS knowledge. The platform will enable psychologists upload and perform their custom experiments with multiple eye-trackers. The trackers will be provided by the processor.

8

College students have hard time organizing their schedule and workload. This mobile app is a smart calendar helping students to organize their workload, manage it and promote group study and get consultation if necessary.

9

Deep Learning architecture makes use of large set of data to learn a distributed semantic representation of a concept. This distributed representation can be used to distinguish inflections from derivations, synonyms etc. This project aims to learn vector representation of Turkish words from a large corpus. The corpus will be provided by the professor.

10

Diagnosis of dyslexia at early age is not easy. Yet, it is confused with low intelligence. It is known that dyslexic children display erratic gaze movements. This project aims develop a testing environment enriched with eye-trackers to detect early dyslexia. It requires Web Programming and Machine Learning knowledge. The eye trackers will be provided by the professor.

11

Children spend much time in watching TV, they may watch all contents of TV that is irrelevant for them such as political programs, discussion programs and some sports programs etc. Our project aim is to restrict irrelevant contents. It is correlated with data mining and machine learning techniques. Program must determine whether content is harmful for children or not. If it is irrelevant, program must shut down the program.

12

The purpose of this project is to understand the preprocessing techniques for voice signal and signal processing techniques. As a social purpose of this project, we are aiming to help arm-disabled people. The students will design a system for computer mouse control through the recognition of simple speech commands from human voice by using a microphone. The system will allow arm disabled people to use their computer mouse through voice commands for various movement directions and specific tasks like “click” and “double-click”. Matlab with signal processing toolbox can help to acquire the requirements.

13

Because the patient story files differ according to each hospital, it is intended to put these reports in a specific form. Body simulation project aims to simulate parts of human body. Students will create a body simulation, user must choose a part of body and program can ask some questions about the complaints to the user. According to the answers, program will create a patient file that is belong to sick person with using text processing. So that, patient story files will be put in a standard form.

14

The goal of this project is to automate the manual class attendance process. Using this system, Instructor will be able to mark the class attendance by scanning the NFC or RFID devices (e.g., smart phones) carried by the students. 

15

Students will implement a system to identify small (mouse) and large (elephant) flows and configure the network to meet the QoS requirements.

16

We will implement following prominent intrusion detection algorithms using software deifned networking.

1.            Threshold Random Walk with Credit Based Rate Limiting

2.            Rate Limiting

3.            Maximum Entropy Detector

4.            NetAD

The project requires background in computer networking, Distributed Systems, Operating systems and algorithms.

17

A key function in computer networks is to route packets from source machine to the destination machine over the network. Routers use routing algorithms to decide which output line should be used to forward the packet. The goal of the project is to demonstrate and animate the working of a few core routing/switching algorithms.

18

The goal of the project is to develop cost affective aerial surveillance system. The system will consist of UAV/quad-copter, smart phone, wireless communication system and camera. The system will enable real time data collection in the form of images, videos, etc. 

19

With using an electronic card like Arduino, Netduino, Raspberry Pi etc. an intelligent home automation will be developed. The devices which can be controlled in a home (light, door, electric, alarm etc.) will be controlled by using a mobile application or desktop application via Internet. Project can be constructed as a small prototype.

20

A robot which tries to find a specific object in a limited area will be developed in this project. Firstly, an object will be shown to the robot then robot will search the object in a restricted area by using its camera, sensors and decide whether it finds the object or not.

21

A program which evaluate a product by analyzing the comments will be developed. The program can use comments for some certain products which can be taken from Internet. Each comment will be analyzed and classified (example: out of 10). After certain number of comment analyzed program will decide the quality of the product (out of 10).

22

A Service-Oriented Architecture (SOA) is essentially a collection of services and these services can communicate with each other. So, service-oriented computing has become one of the predominant factors in current IT research and development efforts over the last few years. Semantic Web Services infrastructure, aimed at eliminating human intervention and thus allowing for seamless integration of information systems. Future’s computer programs must be able to share and process data for automation, integration and reuse even databases used totally independently. This project should be focused on what kind of semantics is needed for semantic web services, where this semantics can come from, how this semantics can be added and what kind of open-source tools could support the semantic enrichment of web services descriptions.

The final product will be a semantic web service implemented by using open-source technologies and should be integrated with a web based application and data stored on a semantic database (triple store).

23

Log data is a definitive record of what’s happening in every business, organization or agency and it’s often an untapped resource when it comes to troubleshooting and supporting broader business objectives. To collect, store, index, search, correlate, visualize, analyze and report on any network / machine-generated data to identify and resolve operational and security issues in a faster, repeatable and more affordable is very important. Conventional database management systems aren’t really designed for full-text searches, and they certainly don’t perform well against loosely structured raw data that resides outside the database. Elasticsearch can scale up to thousands of servers and accommodate petabytes of data. Its enormous capacity results directly from its elaborate, distributed architecture.

The final product will be a platform that provides collecting log data produced by servers or any network device and store on Elastic Stack. Also searching, diagnosing and reporting will be feasible.

24

Network monitoring and analyzing the collected log data in real-time, meaning that if a failure or issue is detected, immediately producing alerts via methods such as email and SMS. This rapid relay of information means that system administrators can be informed about network problems and allowed to instantly take corrective action and minimize potential downtime. This provides security, troubleshooting, saving time and also money.

The final product will be customizable network monitoring platform via using completely open-source tools.

25

The aim of this roject to take the attendance of the students in the classroom automatically. Face recognition will be used to mark the attendance of the students. The Raspberry pi rojec is used for face detection & recognition. The camera will be connected to the Raspberry pi rojec. This raspberry pi rojec will be installed at the front side of class in such a way that we can capture entire class. Face detection and recognition rojec detects faces from the image captured by the camera, and the image of the face is cropped and stored. Using the stored database, absent students will be detected.

26

The aim of this roject work is develop course and exam timetabling software that will be useful to our faculty, using artificial intelligence and optimization algorithms. The main challenge is to schedule courses and exams to timeslots and rooms over a specific period while satisfying a set of constraints.

27

The aim of this roject is to design and implement an online computer based examination system roject courses in the university. This application will deal with question bank maintenance. In addition, the system will provide enhanced security features to avoid exam misprision, and real time processing of results for students.

28

Aim of the study is the classification of hacker/cyber rojectrel accounts on Twitter. Given the seed accounts (like account of RedHack or Anonymous) accounts followed by these accounts and their tweets will be gathered. The accounts that follow these seed groups and their tweets will also be collected. Using Twitter API. Some features will be extracted out of these information and classification methods will be utilized to classify accounts as hacker/not-hacker

29

In this project, a 3D model of a closed-environment like office or room will be reconstructed from multiple 2D images. The final system can be a desktop application or a mobile application which takes a bunch of images and produces a 3D model file such as stl, ply etc.

30

A mobile application roject tourists will be developed for providing information about the place they visit. The place can be a museum, an antique city or a palace. Depending on the GPS data, a place of interest will be automatically detected and information related to this place will presented in a text form. The application should be capable of reading the information too. A final system should have user-friendly interface and use a web-service for retrieving data from a host/website.

31

In this application, user can replace their face with celebrity’s. It can be even a photo or a video. It should have a user-friendly interface and produced photo can be shared via Instagram or a Facebook account.

32

The goal of the study is to help everyone make better clearer more informed decisions about herbal remedies by visualizing the facts about solid scientific evidence for extra health benefits of certain foods. Given the research papers about several health conditions and the herbal remedies; the tool will be able to identify the relationships between the remedies and the health conditions and visualize the power of each remedy on a selected health condition. The roject is inspired from http://www.informationisbeautiful.net/visualizations/snake-oil-superfoods/

33

Security analysts gather essential information about cyber attacks, exploits, vulnerabilities, and victims by manually searching social media sites. This effort can be dramatically reduced using natural language and machine learning techniques. Aim of the study is to detect cyber-attacks using Twitter. The main steps can be summarized as follows;

·         extraction of cyber attack related keywords ranked by TFIDF. 

·         Filtering tweets that contain at least one matching term from the list. 

·         Clustering tweets and extract exemplars that are representatives of clusters. Each exemplar will be annotated to a cyberattack type by computing the cosine similarity and seed queries provided before. 

34

Hateful content published and propagated via the World Wide Web has the potential to cause harm and suffering on an individual basis, and lead to social tension and disorder beyond cyber space. To support the automatic detection of cyber hate online, specifically on Twitter, it is planned to build multiple individual models to classify cyber hate for some  characteristics including race, disability and sex.

35

In this Project, the aim is using a mobile phone for reading Islamic Calligraphies that are usually difficult to read even by the eye.  The advantage is that these writings are not random, usually are verses from Quran. So understanding a portion of it can be sufficient. The end product will roj the writings in latin on top of the camera image. In order to train the system, Machine Learning methods will be used.

36

Solving a jigsaw puzzle is a challenge. Here our aim is to help solving them using computers. This Project was also given last year but this year the aim is a little different. You need to solve real puzzles that are over a surface. First, you will take a photo of this puzzle, recognize the pieces and lead them to their correct positions. You can use reference image to maket he challenge simpler. The difficulties are occlusions, backsides of the pieces etc.

37

In this Project, Computer Graphics methods and already available software will be utilized for generating 3D models of historic buildings from their images. Also, they will be stored in a database and will be shown to users in a nice user interface. Historic buildings need to be categorized and users need to be able to give queries to see the buildings/monuments they are interested in. The system should be robust, there are thousands of historic monuments and the system needs to be responsive to user queries even when there tens of thousands of models in the database.

38

using FPGA programmable chips; the goal of the project is to implement the entire architecture of ARM7TDMI microprocessor using VHDL programming language then to synthesize the code into the chip

39

to model the work and data movement through memory hierarchy system starting from secondary storage until cache memory by using hardware discretion languages and graphical visualization of activity shifts between the mentioned hierarchy

40

real-time graphical visualization of functional simulation of any supervised neural network, where the simulator must enable the user to enter all the required parameters for the desired network then the network is generated and by applying data sets it would be visually possible to monitor the activity of various nodes within the network

41

Clustering is a way of grouping a set of objects that are close to each other and is a well studied problem in literature. Today, these algorithms are used in many fields including machine learning, pattern recognition, bioinformatics, and information retrieval. However, with recent developments in technology, we are living in an information age and classical clustering algorithms don’t work well for large scale data and there is need to implement those algorithms to work in distributed ways. In this project: you will read papers and books about map reduce, spark and other different clustering algorithms. In the end, you will provide an easy to use API that will take input data, run clustering algorithms on cloud and will give clusters as output.

42

With the widespread use of weather data for weather simulation and prediction model, The organizations need fast and efficient model to predict weather pattern using deep learning Using these techniques, scientist can search the weather data to extract meaning and predicting the weather patterns effectively.

43

Lots of data generated rapidly and management is an issue to find the latest version of updated data. Scholary publication data is critical for scientist to find related information for their research. Cloud based systems that  provides services for  publications to manage data is important for finding related metadata.  In this project, using web services data will be collected from different environment and build a search model to extract related publication data.

44

Flow –based system implementation for Student Registration Systems.