Yrd. Doç. Dr. Adil Alpkoçak danışmanlığındaki tezler
12 tez · Dokuz Eylül University
Research of mind shift in testing methodologies and approaches in the world of cloud technologies
This is intended to be a survey of existing test methodologies for cloud-based applications and services with pros and cons of each approach. We will touch base to cloud computing by telling the importance of cloud computing for enterprises, the types of cloud computing platforms used commonly and their attributes. Tools that can be used for cloud-based application testing are reviewed under different categories of testing. A sample Windows Azure application is also being tested with Visual Studio Web Test environment and Selenium Test Suite components to demonstrate the capabilities of these tools.
Design and implementation of Turkish question answering system
In this study, we investigated the design and implementation of named-entity (NE) based question answering system for Turkish text collections. Researches and works on this subject have shown that question answering systems has a complex structure composed of several modules. Thus, we first discussed the structure of a question answering system in three basic phases: question processing, document analysis and answer processing.Firstly, we developed named-entity recognition (NER) tool, which is capable to manage extended named entity hierarchy, annotate data collection, rule-based and dictionary-based named entities extraction and provides a performance evaluation. We also provide a set of rules and dictionaries for NER in Turkish and we present the whole application system in detail. We run a set of experimentation to evaluate the performance of NER system using METU Turkish Corpus. The results we gained from experimentations show that our NER approach produced good results.Then, we propose a new approach, which is named-entity based Questions Answering system for Turkish collections. We designed and implemented our system in structure of boolean information retrieval. We created the structure of indexing for information retrieval and retrieved the relevant documents as results. Then, we found named entities in documents and questions and we matched them using named entity hierarchy. In summary, this work is a starting work in this research area and is thought to produce the best results in terms of performance.
Effects of diacritics in Turkish information retrieval
In this study, we investigated the loss of retrieval performance by writing Turkish text with English Alphabet. As a starting point of this study, we first calculated the loss rate when either documents or queries do not include any Turkish diacritics. After this point we apply document and query expansion techniques to reach the initial performance. Additional to document indexing and query processing techniques, we also investigated the effects of stemming, weighting methods and query lengths on Turkish information retrieval. Our test results show that document expansion technique and equivalence classes produced good results to gain the initial performance.
Dictionary-based effective and efficient Turkish lemmatizer
In this thesis, we present a new Turkish lemmatizer that runs on the GPU and investigate its accuracy and performance. Turkish is an agglutinative language, with a rich morphological structure, contains homographic and inflectional word forms which are lowering the accuracy of stemmers. Thus, in Turkish information retrieval systems, the ability to lemmatize Turkish words efficiently and effectively is important. Our study aims at developing a fast dictionary based lemmatizing approach for indexing and searching documents in Turkish.Recent introduction of CUDA (Compute Unified Device Architecture) libraries for high performance computing on graphic processing units (GPUs) by NVIDIA has increased the trend to use GPUs as general purpose performance environment (GPGPU). Today researchers started to exploit GPU?s high computational capability through CUDA in many applicative contexts requiring intensive use of computational resources such as molecular dynamics, fluid dynamics, cryptology, computer vision, astrophysics and genetics.(e.g. Manavski and Valle, 2008 ) CUDA can be used also in the information retrieval because of its massively workload. Our program, achieves a speedup of as much as 90 times on a recent GPU (NVIDIA GeForce GT240M) over the equivalent CPU-bound version, ultimately with the use of parallelized execution of lemmatization algorithm using a data structure inspired from ?Radix Trie?. Here, we present evaluation results of our string lemmatizing kernels for use in CUDA, which executes parallelized lemmatizing for a test set of query strings. We compared our lemmatization algorithm running on GPU with the serial CPU bound version, and explored issues associated with efficient use of GPU resources with eight different algorithms.
Wireless multicast streaming
This thesis covers performance analysis of Multimedia Broadcast Multicast System (MBMS) streaming delivery method on emulated UMTS environment considering Reed Solomon Forward Error Correction (FEC) algorithm, tune-in delay, rebuffering effect, MPEG-7, Electronic Service Guide (ESG).This thesis introduces MPEG-7 based ESG for mobile TV that provides a multimedia query for MBMS services and sessions, retrieves a tree view of available services and a categorized view according to the genre grouping criteria. The prototype covers OMA BCAST ESG fragments and extends content fragment of ESG by MPEG-7. The proposed ESG prototype has been developed using Visual Studio .Net 2005 Smartphone Emulator.The thesis also covers research on YouTube codecs, interfaces and whether YouTube and MBMS would work together. Another thesis research covers comparison and testing of Xenon Streamer and Darwin Streaming Server (DSS).Keywords: ESG, FEC, MBMS, MPEG-7, mobile TV, multimedia query, performance analysis, rebuffering effect, tune-in delay, UMTS emulation, wireless streaming.
Suffix tree indexing for music information retrieval
This thesis intended for fast and reliable data retrieval from music databases. It introduces new data reduction and indexing approaches for both polyphonic and monophonic music sequences.The study contributes to the literature from three aspects. These are data reduction, suffix tree indexing and tree alignment on external memory. In terms of data reduction, we present a new melody extraction approach for polyphonic music sequences. The new melody extraction approach considers the pitch histogram, and entropy of music sequences. Consequently, accompany channels of the MIDI music sequences are determined for data reduction. In terms of indexing, we present a new suffix tree construction approach for streaming music sequences. Current suffix tree construction algorithms have leaks about indexing music sequences. Hence, we adapted the physical structure of suffix trees for music notes. At last, we consider balance and alignment of suffix trees. In music, alphabet size of music is large. Therefore, we present clustering of music sequence. Therefore each sequence cluster can be indexed by a separate suffix tree to balance the tree.Both our melody extraction and suffix tree construction approaches are tested in detail and discussed. Our evaluation metrics are based on cognition, mathematical proofs and simulations. Experimental results showed that our approaches outperforms.
Multimodal emotion recognition in video
This thesis proposes new methods to recognize emotions in video considering visual, aural, and textual modalities.In visual modality, we proposed a new facial expression recognition algorithm based on curve fitting method for frontal upright faces in still images. Proposed algorithm considers the shape of mouth region to recognize happy, sad and surprise emotions. According to our experiments, our method achieves 89% average accuracy. In addition, we proposed a skip frame based approach for video segmentation.In aural modality, we present an approach to emotion recognition of speech utterances that is based on ensembles of Support Vector Machine classifiers. In addition, we proposed a new approach for Voice Activity Detection in audio signal, and presented a new emotional dataset called Emotional Finding Nemo based on a popular animation film, Finding Nemo.In textual modality, we proposed an emotion classification method based on Vector Space Model (VSM). Experiments showed that VSM based emotion classification on short sentences can be as good as other well-known methods including Naïve Bayes, SVM, and ConceptNet on predicting emotional class of a given sentence.Finally, we use late fusion technique with a web-based interface for emotional browsing of TRECVID dataset, and we developed an emotion-aware video player to demonstrate the system performance.
Archaeological imaging and visualization
The main goal of this thesis is to develop a sophisticated visualization application that can be better used by geophysicists, geologists and archaeologists to explore and understand the spatial visualization of data collected from the historical site. Application has functionalities such as rotating, zooming and cutting the 3D visualization of data from the historical site. Additionally, it is also possible to make some geophysical values transparent for getting a better understanding about the historical site. Application also has capability such as taking snop-shot and recording video abilities.Visualization process can be repeated by using different interpolation and colormap settings. Maximum geophysical data limit value can be defined and changed in application, so users can adjust the insignificant data values and get start the visualization process. System was developed by using Java programming language in Eclipse Platform and VTK (Visualization ToolKit) is used to generate three dimensional graphics.
Comparison of 3D segmentation algorithms for medical imaging
In this thesis we implemented four different 3D segmentation algoritms, and we compared their reults on three different CT Data Sets. These segmentation algoritms are; Seeded Region Growing, Volumetric Segmentation Using Weibull E-SD Fields , Automatic Multilevel Thresholding by using OTSU Method and Unseeded Region Growing. The main results gained from our application as follows; Seeded Region Growing Algorithm produced good result on unnoised datasets with suitable threshold value. Volumetric Segmentation Using Weibull E-SD Fields Algorithm produced good result on our sample dataset which has high amount of contrast difference.However, The results on medical datasets which has low ammont of contrast difference. Automatic Multilevel Thresholding by using OTSU Method Algorithm which takes the segment count as an input by user interaction, produced sufficient results. And lastly, the number of segments produced by the Unseeded Region Growing Algorithm are over the expectations. But they can be considered as sufficent.
Paperless office automation
In this study, a system has been designed and implemented that allows to serve PDF forms over a web site, enable users to fill forms and send form data over internet. The system gives advantage of saving and serving forms in PDF, makes office system paperless, and also allows users to digitally sign PDF documents and signature control. On the server side Windows 2000 server installed with Active Directory, IIS (Internet Information Server), Certificate Server (Root CA) used to create and serve digital signatures. PHP (Hypertext Preprocessor) used to develop a web site which consists of administrator side and user side. Data is stored in a MySQL database. PDF templates stored in the web directory. Adobe Acrobat 7 Professional used to create PDF forms and to convert existing forms in other formats to PDF. php-fdf functions used to manage PDF forms data. Keywords: Paperless Office Automation, Portable Document Format (PDF), Fillable PDF, Forms Data Format (FDF), Digital Signature, internet based paperless office system
Image auto-annotation based on combination of text and visual clustering
The emergence of multimedia technology and the rapidly expanding multimedia collections on the Internet have attracted significant research efforts in providing tools for effective retrieval and management of multimedia data. Traditional image retrieval systems were based on manual annotations of images. This is not powerful enough for proper image retrieval, because of manual annotations. Researchers are focused on extracting image features and annotating images automatically by considering their content that are color, texture and shape. Generally, user needs are high level features and retrieval systems consider/process the low-level features to accomplish the retrieval task. This difference between human interpretation and extracted/processed information is known as the semantic gap of such systems. In this thesis, our aim is to find a linkage between low-level features and high level features to bridge the semantic gap. In this thesis, we propose a novel strategy at an abstract level by combining textual and visual clustering results to retrieve images using semantic keywords and auto-annotate images based on similarity with existing keywords. Our main hypothesis is that images that fall in to the same text-cluster can be described with common visual features of those images. In order to implement this hypothesis, we set out to estimate the common visual features in the textually clustered images. When an un-annotated image is given we find the best image match in the different textual clusters by processing their low-level features. Experiments have demonstrated that good accuracy of proposal and its high potential of use in annotation of images and for improvement of content based image retrieval.
Classification of textile images
ABSTRACT In this study, textile images are classified into striped, plaided and speckled image classes. Striped images are also categorized into the classes of horizontal, vertical, 45 and 135 degree lines. Line and circle features are used for the classification. To detect line and circle features, it is applied to line detection operator and Hough transform method respectively. With the line feature, striped and plaided images are classified, but speckled images are classified with circle feature. For the classification of the textile images, it is used some classification methods as Manual. Thresholding, Nearest Neighbor, ^-Nearest Neighbor, Minimum Distance. After the evaluation of the system, we have found that the best methods to classify striped images are "Nearest Neighbor with Manual Thresholding" and "^-Nearest Neighbor with Manual Thresholding and £=5, k=7, kr=\ 1". The best methods to classify plaided images are "Nearest Neighbor" and "Nearest Neighbor with Manual Thresholding". The best method to classify speckled images is "Manual Thresholding with Line and Circle Threshold > 0". The results of the classification are quite successful.