This study investigates the connection between design patterns, object oriented (OO) quality metrics and software maintainability. The literature on OO metrics, design patterns and software maintainability are reviewed, the relation between OO metrics and software maintainability is investigated, and then, in terms of obtained maintainability indicator metrics, the maintainability change of an application due to usage of design patterns is observed.

AMR (Automatic Measurement Reading) applications usually require measurement data to be collected from separate locations. In order to combine the data retrieved from separate sources into a meaningful result, all sources should share a common time sense. Therefore, it is necessary to implement a synchronization scheme in measurement networks. In this thesis, a synchronization scheme which combines GPS (Global Positioning System) and two high accuracy WSN (Wireless Sensor Network) time synchronization algorithms will be proposed and evaluated. The synchronization accuracy of the proposed method is compared to the accuracy of NTP (Network Time Protocol) by simulation. This research work is fully supported by the Public Research Grant Committee (KAMAG) of TUBiTAK within the scope of National Power Quality Project of Turkey with the project No: 105G129.

Promotion is one of the four major marketing elements of the marketing mix (others are product, price and place) in implementing marketing strategy. Promotion is dealing with the ways a company communicates with its customers to persuade them to buy the product. Promotion mix covers all the different ways a company choose to communicate with its customers such as advertising, personnel selling, PR, sales promotion and others. Selecting the optimal blend of the promotion mix is a tough and critical issue for marketers and does not have a fix operative formula. The fast pace of improvements in digitization in this era led companies produce digital products. Due to their inherent characteristic of digital products, such as intangibility, promotion mix selection is a more challenging issue. In my thesis study, I proposed a framework in classifying the digital products and then apply a fuzzy linguistic decision model approach with appropriate genetic algorithms to reach an optimum promotion mix set for digital products. Optimization is targeting to justify the objectives of the company, provide a satisfying marketing performance for the companies of digital product producers and utilize their budget effectively. The proposed model is implemented on an empirical case and produced satisfactory results.

Video annotation has become an important issue due to the rapidly increasing amount of video available. For efficient video content searches, annotation has to be done beforehand, which is a time-consuming process if done manually. Automatic annotation of faces for person identification is a major challenge in the context of content-based video retrieval. This thesis work focuses on the development of a semi-automatic face annotation system which benefits from online learning methods. The system creates a face database by using face detection and tracking algorithms to collect samples of the encountered faces in the video and by receiving labels from the user. Using this database a learner model is trained. While the training session continues, the system starts offering labels for the newly encountered faces and lets the user acknowledge or correct the suggested labels hence a learner is updated online throughout the video. The user is free to train the learner until satisfactory results are obtained. In order to create a face database, a shot boundary algorithm is implemented to partition the video into semantically meaningful segments and the user browses through the video from one shot boundary to the next. A face detector followed by a face tracker is implemented to collect face samples within two shot boundary frames. For online learning, feature extraction and classification methods which are computationally efficient are investigated and evaluated. Sequential variants of some robust batch classification algorithms are implemented. Combinations of feature extraction and classification methods have been tested and compared according to their face recognition accuracy and computational performances.

This thesis describes an information extraction system that is designed to process free text Turkish radiology reports in order to extract and convert the available information into a structured information model. The system uses natural language processing techniques together with domain ontology in order to transform the verbal descriptions into a target information model, so that they can be used for computational purposes. The developed domain ontology is effectively used in entity recognition and relation extraction phases of the information extraction task. The ontology provides the flexibility in the design of extraction rules, and the structure of the ontology also determines the information model that describes the structure of the extracted semantic information. In addition, some of the missing terms in the sentences are identified with the help of the ontology. One of the main contributions of this thesis is the usage of ontology in information extraction that increases the expressive power of extraction rules and helps to determine missing items in the sentences. The system is the first information extraction system for Turkish texts. Since Turkish is a morphologically rich language, the system uses a morphological analyzer and the extraction rules are also based on the morphological features. TRIES achieved 93% recall and 98% precision results in the performance evaluations.

This thesis focuses on the technological trends in smart products, and analyzes their conformity to the user expectations. The advances in computation technologies have totally revolutionized the product concept, and with the integration of microchips, software and sensors into the classical everyday objects, smart products, able to sense the context, reason about the sensed data and act according to the situation, have emerged. This new way of computing basing on the ubiquitous and calm computing visions, has distributed the digital information into the surrounding environment, and once freed from the limited resources provided by the classical desktop based computing, attempted to enhance user product communication and collaboration in everyday environments. Via their sensing - decision making - acting process and advanced interaction capabilities, smart products have gained the ability to better interpret user needs and intuitively communicate with users through simplified interfaces involving the majority of the senses without even disturbing or overburdening their users. The study first, throughout a literature review, examines these improvements in computation technologies and determines the trends related to smart products. An empirical research is then conducted to find out to what extend user expectations from smart products overlap with the ongoing researches in this area. The findings including users&rsquo / conception about smartness and expectations from different types of smart products are analyzed regarding to the technological trends to deduce the coherence between literature&rsquo / s orientation and user preferences. The study considered the technological trends as a database and takes the user expectations as the design motivation.

Performance measurement makes up the core of all health care systems in transition. Many countries and institutions monitor different aspects of health care delivery systems for differing purposes. Health care deliverers are compared, rated, and given incentives with respect to their measured performance. However, a global health care domain is currently striving for attaining commonly accepted performance measurement models and base-standards that can be used in information systems. The objective of this thesis is to develop an ontological framework to represent performance measurement and apply this framework to interpret performance measurement studies semantically. More specifically, this study made use of a formal ontology development methodology by utilizing web ontology and semantic web rule languages with description logic in order to develop a commonly accepted health care performance measurement ontology and knowledge base system. In the ontology developed, dimensions, classes, attributes, rules and relationships used in health care delivery and performance measurement domain are defined while forming an initial knowledge base for performance measurement studies and indicators. Furthermore, we applied the developed performance measurement ontology to the knowledge base while driving those related performance indicators for predefined categories. The ontology is evaluated within the features of the Turkish health care system. Health care deliverer categories are identified and by executing inference rules on the knowledge base, related indicators are retrieved. Results are evaluated by domain experts coming from regulatory and care provider institutions. The major benefit of the developed ontology is that it presents a sharable and extensible knowledge base that can be used in the newly emerging performance measurement domain. Moreover, this conceptualization and knowledge base system serve as a semantic indicator search tool that can be used in different health care settings.

As dynamic web applications are becoming widespread nearly in every area, ASP.NET is one of the popular development languages in this domain. The errors in these web applications can reduce the credibility of the site and cause possible loss of a number of clients. Therefore, testing these applications becomes significant. We present an automated tool to test ASP.NET web applications against execution errors and HTML errors that cause displaying inaccurate and incomplete information. Our tool, called Mamoste, adapts concolic testing technique which interleaves concrete and symbolic executions to generate test inputs dynamically. Mamoste also considers page events as inputs which cannot be handled with concolic testing. We have performed experiments on a subset of an heavily used ASP.NET application of a government office. We have found 366 HTML errors and a faulty component which is used almost every ASP.NET page in this application. In addition, Mamoste discovered that a common user control is misused in several generated pages.

Classification of migraineur and healthy subjects using statistical pattern classifiers on functional Near Infrared Spectroscopy (NIRS) data is the main purpose of this study. Also a statistical comparison between trials that have different type of classifiers, classifier settings and feature sets is done. Features are extracted from raw light measurement data acquired with NIRS device, namely Niroxcope, during two separate previous studies, using Modified Beer-Lambert Law. After feature extraction, Na&iuml / ve Bayes classifier and k Nearest Neighbor classifier are utilized with and with-out Principal Component Analysis in separate trials. Results obtained are compared within each other using statistical hypothesis tests namely Mc Nemar and Cochran Q.

Public services have been provided in an electronic format to reduce time and cost and to improve service quality. In this context, significant investments have been made since 2003 in Turkey and many government agencies started to provide services via the Internet. Government agencies need to learn the effectiveness of the offered services and the benefits of these services to citizens. The purposes of this thesis are to evaluate e-government websites in a comprehensive manner and to identify website design related problems. Especially, the study focuses on user-centered design of the e-government application. For this purpose, 33 e-government websites are evaluated, and an interview was conducted with eight government agencies websites&rsquo / designers. As a result of these studies, the problems encountered while using government electronic services are identified. Also, causes of these problems are found out. The study aims to provide a guideline in order to design user-centered e-government applications.