Department of Software Engineering
Permanent URI for this collection
Browse
Browsing Department of Software Engineering by Subject "software"
Now showing 1 - 20 of 46
Results Per Page
Sort Options
Item A COMBINED APPROACH OF CLUSTERING AND ASSOCİATİON RULE MİNİNG FOR CUSTOMER PROFILING IN VIDEO ON DEMAND SERVICES(2022-01-24) Güney, Sinem; TURHAN, Çiğdem; PEKER, SerhatToday, IPTV (Internet Protocol Television) service providers offer VoD (Video on Demand) services as part of their business initiative toward generating more revenue. To do this, they need to know about customer behaviors and expectations. Such information related to users is stored in CRM (Customer Relationship Management) systems. Against this backdrop, the present work aims to analyze customers in VoD services with applying clustering and Association Rule Mining techniques. The LRFMP (Length, Recency, Frequency, Monetary, and Periodicity) model is applied to find out the customer behaviors, whereas the k means clustering algorithms allow for determining the number of clusters and customer profiles. As a result, four different customer groups are identified, namely as “consuming and most valuable”, “less consuming and less valuable”, “less consuming but loyal”, and “neither loyal nor valuable”. A major source of information for this study is the content type or genre as regards the content category and rental preferences of subscribers. To this end, the association rule algorithm (Apriori) is employed to predict the customers’ potential rentals. A combined approach as such would be useful for IPTV service providers to further shed light on precise customer behaviors and preferences, thus allowing to create more targeted marketing strategies for each category of subscribers in order to improve customer satisfaction and increase revenues in the long run.Item A COMPARATIVE STUDY OF E-GOVERNMENT EVOLUTION IN AFRICA(2022-03-01) El-Asheibi, Nagat; Mishra, AlokStates are constantly striving to improve their services to their citizens, trying to reach them through the Internet in order to be able to explain their policies and be effective. The delivery of public information and services to citizens through the Internet channel is considered e-government. The e-government uses information technology to serve citizens, businesses and institutions. There are various factors that affect e-government. In general, researchers have examined e-government studies that have been put into practice by various African countries. The purpose of this study is to explore the possibilities offered by e-Government in Africa by documenting few initiatives on the continent that have developed innovative models that contribute to governments through the implementation. To achieve the aim of the thesis, we present former works and e-government evolution in Africa with discussion on the evolution of e-government in different African countries (Libya, Egypt, Botswana, Sudan, South Africa, Nigeria, Gambia, Ghana, Uganda, Kenya, and Mauritius).Item A COMPARATIVE STUDY OF PRIVACY-PRESERVING TECHNIQUES FOR THE CLOUD STORAGE(2022-01-20) Al-Qaisi, Thr; Mishra, AlokInformation and data privacy have become critical concepts in the cloud computing industry, especially as internet users seek to use the cloud environment to store their personal and sensitive data. Many cloud service providers currently offer premium and quality-based services for their users as the first initiative for building a vast cloud community. However, security and privacy violations began to emerge and evolve in the cyber world and threaten most of its infrastructure. Fortunately, much research has been achieved to develop the proper techniques to overcome data privacy's perilous challenges and find better methodologies for protecting cloud storage contents. In this work, our study focuses on comparing several forms of privacy-preserving techniques for cloud storage. The study includes a comprehensive analysis of privacy-preserving techniques and their common attributes for the aim of designing flexible, secure, and efficient solutions for the dilemma that faces data privacy. We also present an attainable solution for privacy-preserving of the cloud storage by proposing a multi layer encryption framework with the use of one-time password authentication technology and a multi-cloud storage structure.Item A COMPARISON OF SQL AND NoSQL DATABASES(2022-02-25) RASHD, SOUAD RASHD ALİ; YAZICI, AliThis study comprises a research about both SQL and NoSQL databases and a comparison in terms of the advantages and disadvantages of each. The comparison is implemented depending on scalability, (ACID and CAP) theory, flexibility, performance, schema, query language, cost, speed and the stored data. We described briefly several examples of SQL databases including My SQL , MS-SQL Server Express Edition , Oracle 11g Express Edition Database. Also, the study included a description for many types of NoSQL database including Key-Value Store Database, Column-Oriented Database, Document Store Database and Graph Databases. The result of the study detected that SQL database is a table based database, supports flexibility, is vertical scalable, and utilizes Structured Language Query. On the other hand, the study discovered that SQL database is not a suitable choice for hierarchical data storage, etc. Whereas in terms of NoSQL, the study discovered that NoSQL databases are less flexible, however, horizontally scalable, have a non-structured querying language and is a good choice for hierarchical databases.Item A COMPONENT BASED MODEL DRIVEN SOFTWARE DEVELOPMENT FRAMEWORK FOR WEB-BASED APPLICATIONS(2017-01-07) Alrubaee, Afrah; Çetinkaya, DenizDeveloping a high quality, cost effective, reliable and on time software systems is a challenging task due to the large size and complexity of these systems. Traditional developing approaches that are concerned with constructing software systems from scratch cannot be suitable for dealing with this challenge. For that reason several approaches have been introduced to increase the productivity of the development. Two of these approaches are component based software engineering and model driven software development. Component based software engineering (CBSE) has been introduced as a solution for software reusability problem by using reusable software components to build new software system. Model driven development (MDD) is another approach in software development that was introduced to increase productivity and decrease the cost and effort. MDD aims to face the challenges of software development process through representing the essential aspects of the required system as models and generating the final source code from these models. The aim of this work is proposing a software development framework that combines MDD and CBSE approaches for merging the advantages and features of these approaches to facilitate software development. The framework is used successfully to develop an e-learning system as a case study. The framework was evaluated by making a comparison between existing component based software development process models and our framework.Item A HYBRID METHOD FOR OBJECT TRACKING IN VIDEO(2022-01-26) Taşan, Hakan; GÖKÇAY, ErhanDetecting the object in the video and tracking it has been emerging as an important research field in computer vision and image processing. Many algorithms have been developed for object tracking and there are some conditions in which each algorithm is successful or unsuccessful. In this thesis, a robust hybrid system that consisting of three object detection and tracking algorithms is proposed for the purpose of tracking object in video. These algorithms are template matching, color-based histogram and SURF based on feature point. OpenCV library have been used to implement these algorithms in hybrid system. While implementing algorithms, different techniques have been applied such as gaussian blur, color space conversions, Otsu thresholding, sliding window approach, feature extraction and description, and distance measurements. Any object from the video can be selected and the selected object can be traced in the rest of the video. To prevent occlusion of the object and to minimize the effects of sudden movement of scene, refreshing selected object approach is used each fifth frame of the video. Aim of the hybrid system is to improve the detection rate of the object to be tracked in sequence of video frames. All performance tests have been performed on NTU-VOI 2018, Visual Tracker Benchmark 2013, NfS 2017 and Davis 2017 datasets. The test results of the proposed hybrid system have been compared with the results of the three individual detecting and tracking algorithms. The results show that hybrid system gives the best performance except for processing time for tracking object in video.Item A MODEL FOR EVALUATION OF USER TRUST TO CLOUD SERVICE PROVIDERS(2017-04-02) İnceler, Deniz Koray; Bostan, Atila; Turhan, ÇiğdemThis thesis reviews the trust on the cloud computing environment from the viewpoint of cloud service users by developing a multiview metric to evaluate diverse cloud service providers. The approaches to user trust concept are studied to show that evaluation of trust necessitates both technical and non-technical approaches that need to be assessed together to obtain an all-in exclusive solution. Trust has been studied in six different domain approaches to fulfill the requirements of cloud users. The motivation in this study is to develop a trust-scale that consists of metrics which are based on observable facts. Such a trust-scale is expected to help cloud users in their trust evaluation of different cloud service providers. Ultimately, the trust-scale is formed with its metrics and subparts and fine-tuned by using the survey data as to map the expert point of view. The proposed evaluation methodology and scale helps cloud users in their trust assessment of diverse cloud service providers using observable and quantifiable metrics. As a result, the accuracy tests of the metric that is developed in this study resulted as %95 accurate with the expert point-of-view. This accuracy test result points out that the trust scale is capable to evaluate trust at a sufficient level of trust.Item A NOVEL SOFTWARE DEVELOPMENT METHODOLOGY FOR RESEARCH-BASED SOFTWARE PROJECTS(2022-02-15) Cereci, İbrahim; Karakaya, ZiyaSoftware development in mid-sized or large-scale projects is usually carried by a group of individuals, whose coordination, choosing suitable development practices for the group and keeping track of the software development process are all challenging tasks. Software development methodologies are heavily utilized for these purposes. Although previously proposed methodologies manage to meet the needs of the industry, they are not tailored to do so for academicians developing research based software projects at universities. Therefore, in this thesis, our first aim is to show the necessity of a new software development methodology for research-based projects carried by the universities. Then, through current literature, interviews and questionnaires, the needs and the best practices of research-based projects are collected using grounded theory qualitative method. Finally, after completing the analysis of the findings, a new software development methodology is proposed tailored to the needs of the researchers that are working on research-based software projects. The proposed methodology is evaluated by the experts and it is found useful for the research-based software projects. Collected issues and best-practices are proposed as a guideline for the project managers, team members, funding agencies, universities, and the end-users of research-based software projects. These guidelines can be utilized to increase the productiveness of such projects.Item A PROGRAMMING LANGUAGE FOR COMPONENT-BASED SOFTWARE DEVELOPMENT(2017-01-07) Delikelli, Ahmet; Çetinkaya, DenizIn this thesis, it is aimed to develop a programming language in which components are defined in the component-based software development process. It is provided that a component is appropriate for reuse and has a specific behavior by taking into account component based software engineering (CBSD). This programming language has been developed as a domain specific language (DSL). During the language development process, Xtext framework which works on Eclipse software development kit (SDK) has been used. The language is Java based and grammar of the programming language, syntax validation and translation into Java language has been implemented with Xtext framework. The proposed programming language has the most common features that a programming language should have. A case study has been implemented in order to show how the features of the language are used. The evaluation of the language is done by the case study and several examples that illustrate the usage of the language.Item A SYSTEMATIC MAPPING STUDY ON THE USE OF EYE-TRACKING IN PROGRAM UNDERSTANDABILITY(2022-01-10) ATIED, SHUKRULLAH; Çağıltay, Nergiz Ercil; Topalli, DamlaEye-tracking technology is used to get insight into a person’s vision and determine where they are looking and gather data on their visual behavior. Researchers have been collecting eye movement data to analyses and comprehend various tasks in Software Engineering (SE) by using eye-tracker technology. However, there are very limited studies analyzing the use of eye-tracking technology in the field of SE and providing a general picture from this perspective. Accordingly, in this thesis study, by undertaking a Systematic Mapping (SM) study, it is aimed to bring all evidence together and give an overview on the use of eye-tracking in SE, specifically for program comprehension. 121 related articles were selected after carefully reviewing and excluding unrelated studies. The result shows that the use of eye-tracking in SE is increased in recent years, and different SE activities were studied. Moreover, four different factors are influential in program comprehension: code elements and entities, tools, developer’s background, and eye movement and reading behaviors. These findings may help the developers to adopt the founded influential factors to help them in program understandability.Item AGE AND GENDER PREDICTION FROM 3D-BODY AND FACE IMAGES(2022-02-21) ÇAMALAN, Seda; ŞENGÜL, GökhanThe biometric data collected from individuals provide an array of information about any population and their environment which can be used in several areas, including transportation (busses, ferries, railways, etc), shopping malls, public areas, sports centers, museums, supermarkets, libraries, etc., not to mention security applications. In detail, this biometric data is related with identity, gender, race, height, weight, and eye and hair color of the person. In this thesis, an image processing-based system to predict the two major aspects, age range and genders of people is developed and integrated as a software tool. A standard RGB camera is used to acquire face images, while a 3D camera is used for body information. To predict the gender and age of each individual, statistical pattern recognition algorithms, deep learning and neural network-based approaches are utilized. For statistical methods, LBP and HOG methods are applied on face images to extract features, then KNN and SVM classification methods are applied as classifiers. Convolutional neural network is used to predict age range of people and the comparison between statistical methods and convolutional neural networks are presented. For age prediction, from face images, statistical methods results yielding a top accuracy of 40.1%; whereas, the best accuracy obtained from CNN deep learning is 59.1%. In addition, 3D body information is used for gender and age prediction by applying statistical and neural network methods. These methods show to improve the gender prediction rate by up to 99.26% and age prediction by 99.41% for the whole-body information. The upper-body and lower body parts are also examined separately to predict the age and gender of the each individual.Item AGILE SOFTWARE DEVELOPMENT METHODOLOGIES: A COMPARATIVE STUDY(2022-02-24) Choukou, Abakar Adoum; Karakaya, Ziya; Yazıcı, AliSoftware systems have been part of the modern society for a couple of decades. Software engineering as one of the fastest-growing disciplines in today's technology, is becoming an integral part of our daily live with tremendous advancement and almost used in all disciplines of life. Agile software development methodologies are largely applied in almost all software companies due to its ability of building software systems faster and with higher quality. Furthermore, Agile methods are always fulfilling the satisfaction of the clients through prompt and continuous delivery of the reliable software systems. The aim of this study is to conduct a systematic literature review to state the different Agile software methodologies’ suitability regarding the usage area and different types of projects. Also, it is within the aim of the research to identify and provide extensive information about different Agile software development methodologies including their popularity, their differences and similarities. Lastly the research summarizes the existing evidence of Agile software methods from the literature review and specify its benefits and limitations. Five questions were designed to achieve the objective of this research, the collection of data and even so the results are also prepared based upon extensive literature review we made for this study.Item AN INTELLIGENT SYSTEM FOR EXERCISE PLANNING AND PHYSICAL ACTIVITY RECOGNITION USING MOBILE TECHNOLOGIES(2022-02-28) Kalem, Güler; Turhan, ÇiğdemIntelligent guidance in the healthcare domain using mobile technologies is an important development since users can benefit from the individual exercise plans that are specifically designed for their purpose, demographic information and health background. Furthermore, with the system continuously tracking their activities, the users are motivated and guided to complete their daily specified exercises. The developed system determines the specific exercise program suitable to the user with case-based reasoning. With the help of the accelerometer and gyroscope facilities of a mobile phone, users' activities are recognized and classified using KNN (K-Nearest Neighbors) algorithm. Based on their individual exercise routine and the performed activities in the current day, the rest of the exercises are calculated and presented to the user as a message to guide and encourage them. For the evaluation, the system is tested by users, and a questionnaire is conducted. The results show that the system is found to be beneficial and effective by all the participants.Item ASPECT-BASED SENTIMENT ANALYSIS IN TURKISH(2022-01-24) Özkan, Deniz; Turhan, ÇiğdemMost customers rely on reviews and comments of other consumers that already purchased and used the products that they intend to purchase. As the sharing opinions and preferences of people on online platforms are widespread, these huge data sources are highly valuable to companies to gather feedback on their products. Therefore, researchers have an essential data mining goal to extract useful information from sentiments. In this thesis, the aim is to perform an aspect-based analysis to determine the sentiment polarity of the reviews for a smart phone using natural language processing techniques in Turkish for the performance, price and camera aspects. The techniques used are data preprocessing, explicit and implicit feature extraction as well as grouping corresponding aspects and lexicon-based sentiment analysis at word-level and word-group level. The evaluations show that the highest values of precision, recall and f1 measure for the aspects examined are found to be 93%, 94% and 93%, respectively. These results reveal that our study has remarkable performance compared to other Turkish aspect-based sentiment analysis studies.Item ASTUDY ON INTEGRATION OF AGILE WITH STRUCTURED SOFTWARE DEVELOPMENT PROCESSES(2017-01-07) Nura, Abdelmagid; Mishra, AlokNowadays, Agile methodologies (AMs) have interred SDLC in all companies. Many organizations have used AMs and traditional methods to enhance öethodologies as an optimal solution in the business area. First, thinks about changes during SDLC; also customer may change his mind. So here is a need for the combination methodologies. Second, in spite of the fact that AMs can be beneficial to organizations, there is a need for traditional methodologies in some phases of the SDLC as a mixed framework.Item AUTOMATED FUNCTIONAL SIZE MEASUREMENT FOR MULTI-TIER OBJECT RELATIONAL MAPPING ARCHITECTURES(2015-10-25) DEMİREL, Hamdi; ÖZKAN, BarışSoftware Functional Size Measurement methods are generic, and additional rules and concept mappings are required for specific software settings in order to improve the measurement objectivity and consistency. Although some operational definitions have been made for specific architectural style measurement procedures, there are still opportunities to improve measurement objectivity and consistency. These opportunities can be achieved through automation by minimizing measurer errors, time/effort costs and dependence to the development environment. In this study, a functional size measurement approach for Three-Tier Object Relational Mapping Business Application is proposed, and a prototype measurement tool and its results for validation are presented in case studies.Item BOILERPLATE DESIGN FOR STRUCTURED SOFTWARE REQUIREMENTS SPECIFICATION IN TURKISH LANGUAGE(2022-06-10) ÖZTEKİN, GONCA CANAN; MENEKŞE DALVEREN, Gonca GökçeMany projects are carried out in e-Government applications, requiring thorough analysis for optimization. In this respect, errors are inevitable in the developed applications as a result of poor analyses, making it very important to correctly understand and transfer the respective sentences. To determine the quality of software requirements, a method or model is necessary to evaluate the statements. Many of these requirements are expressed in the natural language of native speakers. There has been some research on the quality model for software requirement; however, the same is not true for the Turkish language expression, especially in the field of e-government. Furthermore, there is no standard for creating software requirements specification statements in Turkish. Against this backdrop, the main purpose of the present thesis is to define common texts and to ensure that the analysis sentences are developed within a certain structure to minimize and standardize the errors originating from natural language. For this purpose, a structure is created that allows the analyst or requirements engineer to accurately enter the requirement sentences. The proposition here serves as a quality evaluation model for the Turkish Software Requirements Specification (SRS). An interface is developed so that the model may match requirements with common text templates, compute similarity values, and insert the requirements, regulatory documents, and user requirements into the template in a more convenient format. The proposed approach improves the quality of SRS in e-Government applications and, as such, contributes to the literature in this field.Item CLOUD COMPUTING SECURITY ISSUES AND SELECTION OF DEPLOYMENT MODEL AND SERVICE MODEL ACCORDING TO SECURITY REQUIREMENTS(2015-01-31) SEZEN, Arda; YAZICI, Ali; BOSTAN, AtilaThis thesis reviews the necessity of X.800 Recommendation service categories for different cloud service models and cloud deployment models together with some security solution approaches in cloud computing. The thesis evaluates the solution approaches to show that technical and non-technical approaches need to be handled together to produce comprehensive solutions. Six technical solution approach have been included to evaluate the fulfilment of X.800 Recommendation service categories. Eventually, twelve hypotheses have been formulated, tested, and accepted based on the survey data to understand the necessity of X.800 Recommendation service categories for different cloud service models and cloud deployment models.Item COMPARISON OF PREDICTION ALGORITHMS FOR STUDENT PERFORMANCE PREDICTION(2022-02-22) Bah, Amadou; Karakaya, Ziya; Yazıcı, AliThis thesis investigates the application of six machine learning algorithms to student performance prediction, using datasets made up of only students information available at the Atilim University administrative systems. In addition, these learning algorithms were compared using four measures: Accuracy, Precision, Recall and F measure. The study also investigates whether the number of courses predicted together is directly or inversely proportional to the performance of the classifiers used. A measure of the effects of data preprocessing as well as Correlation based Feature Selection (CFS) on the learning algorithms was also conducted, respectively. The algorithms used are: Naive Bayes, Logistic Regression, Multilayer Perceptron, SMO (based on Support Vector Machines), IBk (K-Nearest Neighbor) and J48 (C4.5 Decision Tree). Naïve Bayes and IBk proved to be the best among the compared algorithms. The results also show that as the number of courses being predicted together increases, the prediction performance decreases. Data preprocessing and CFS are also found to generally improve the performance of the machine learning algorithms.Item COMPUTER VISION AND MACHINE LEARNING BASED ADAPTABLE CONVERSION METHOD FOR ANY LIGHT MICROSCOPE TO AUTOMATED CELL COUNTER BY TRYPAN BLUE DYE-EXCLUSION(2017-07-02) Özkan, Akın; İşgör, Sultan Belgin; Şengül, GökhanAlmost all of the cell biology experiments involve counting of cells regularly to monitor cell proliferation and viability. Knowledge of the cell quantity and quality are important parameters for the experimental standardization and toxicity impact estimation. There are two different approaches to count the cells, such as, hemocytometer-based manual counting, and usage of an automated cell counter. Either of the methods have their advantages and disadvantages. High investment and operational cost limit the wide range usage of automated cell counters. On the other hand, manual cell counting based on hemocytometer has various limitations by the fact that reliability of cell counting highly depends on operator’s experience. Moreover, high estimation time requirement and human labor are two more drawbacks of the manual process. This thesis proposes state-of-the-art alternative method (i.e. framework) for the cell counting by defining computer vision and machine learning based conversion methodology. The basis of the proposed method is the adaptation of hemocytomer-based manual counting to automated procedure by adding middleware decision software to reduce its shortcomings. In addition, two novel data sets are collected to test our proposed method in terms of cell counting (i.e non-stained) and cell viability analysis (i.e. stained). The datasets are available for non-profit public usage from “biochem.atilim.edu.tr/datasets/” which will be baseline to future studies on this research domain. Both datasets contain two different types of cancer cell images, namely, caucasian promyelocytic leukemia (HL60), and chronic myelogenous leukemia (K562). From our experimental results, our method reaches up to 92% and 74% in terms of recall scores for HL60 and K562 cancer cells, respectively, with the high precision. The experimental results also validate that the proposed method can be a powerful alternative to the current cell counting approaches.
- «
- 1 (current)
- 2
- 3
- »