Ferda Nur Alpaslan

An ontology-based retrieval system using semantic indexing

In this paper, we present an ontology-based information extraction and retrieval system and its application to soccer domain. In general, we deal with three issues in semantic search, namely, usability, scalability and retrieval performance. We propose a keyword-based semantic retrieval approach. The performance of the system is improved considerably using domain-specific information extraction, inference and rules. Scalability is achieved by adapting a semantic indexing approach. We implement the system using the state-of-the-art technologies in Semantic Web and evaluate the performance against traditional systems. Further detailed evaluation is provided to observe the performance gain due to domain-specific information extraction and inference.

Text summarization using Latent Semantic Analysis

Text summarization solves the problem of presenting the information needed by a user in a compact form. There are different approaches to creating well-formed summaries. One of the newest methods is the Latent Semantic Analysis (LSA). In this paper, different LSA-based summarization algorithms are explained, two of which are proposed by the authors of this paper. The algorithms are evaluated on Turkish and English documents, and their performances are compared using their ROUGE scores. One of our algorithms produces the best scores and both algorithms perform equally well on Turkish and English document sets.

An intelligent tutoring system for student guidance in Web-based courses

In this study, an intelligent agent to guide students throughout the course material on the Internet is defined. The agent will help students to study and learn the concepts in the course by giving navigational support according to their knowledge level. An overlay model is used to model the knowledge levels of students.

The temporal communication behaviors of global software development student teams

This paper examines the global software development process by using content analysis techniques, as described in an earlier study (Serce et al., 2011), to determine time-variant patterns of communication behaviors among student teams engaged in a global software development project. Data gathered from two software development projects involving students in the US, Panama, and Turkey were used to determine how globally distributed team behavior is temporally patterned in complex ways. A formal, quantitative methodology for time variant analysis of the transcripts of global software student teams based on content analysis is established. Results from the analysis suggest a positive correlation between a teams temporal communication patterns and project outcomes as well as a relationship between variations in communication behaviors and different phases of the software development cycle. The research also found that the temporal variations in communication behaviors between software phases were similar for the two projects. Such findings are intended to strengthen the case for developing new temporal measures for analyzing groups and teams.

Exploring Collaboration Patterns among Global Software Development Teams

This study examines communication behaviors in global software student teams. The authors of the paper characterize the types of communication behaviors that occur when student teams are engaged in a software development project. The authors present findings from a one-semester study that examined factors contributing to successful distributed programming interactions among students enrolled at the University of Atilim (Turkey), Universidad Tecnológica de Panamá, University of North Texas, and Middlesex University (UK). Using content and cluster analyses techniques, we identified distinct patterns of collaboration and examined how these patterns were associated with task, culture, GPA, and performance of collaborative teams. Our results suggest that communication patterns among global software learners may be related to task type, culture and GPA. It is hoped that these findings will lead to the development of new strategies for improving communication among global software teams.

A Comparison of Team Performance Measures for Global Software Development Student Teams

One of the most difficult tasks for global software development researchers is quantifying the performance of groups and students who participate in these distributed projects. There has been much debate about which factors better correlate with team performance and which best describe a successful team. The purpose of this paper is to compare the different approaches that have been used to evaluate the performance of global software learners and show how these techniques can affect research results. Using data from student groups engaged in global software development projects for the past year, the authors apply a number of different assessment methods and show their effects on different performance indicators. Our study suggests that the selection of appropriate measures to evaluate team performance can dramatically affect how one identifies successful teams.

ANFISunfoldedintime for multivariate time series forecasting

This paper proposes a temporal neuro-fuzzy system named ANFIS_unfolded_in_time which is designed to provide an environment that keeps temporal relationships between the variables and to forecast the future behavior of data by using fuzzy rules. It is a modification of ANFIS neuro-fuzzy model. The rule base of ANFIS_unfolded_in_time contains temporal TSK(Takagi-Sugeno-Kang) fuzzy rules. In the training phase, back-propagation learning algorithm is used. The system takes the multivariate data and the number of lags needed to construct the unfolded model in order to describe a variable and predicts the future behavior. Computer simulations are performed by using real multivariate data and a benchmark problem (Gas Furnace Data). Experimental results show that the proposed model achieves online learning and prediction on temporal data. The results are compared with the results of ANFIS.

Global teams: futuristic models of collaborative work for today's software development industry.

This paper emphasises the importance of global teams in the field of software development. The paper presents an approach for setting up pilot studies simulating those key features that make global software development teams particularly attractive to exploit and challenging to manage. The underlying research is supported by a research project funded by the US National Science Foundation with the participation of universities from US, Turkey, Panama and the UK. The paper provides detailed guidelines for setting up simulations resembling globally dispersed software development teams and discusses preliminary data of two pilot studies with involving collaboration between teams residing in the US and the UK. Key concerns of this research are those factors affecting collaborative work when global teams are involved. Such factors include differences caused by distance, culture, time zones and technology.

Teaching Students How to Work in Global Software Development Environments

Given that outsourcing has become a fact of life, it is becoming increasingly obvious that we need to ensure that computer science students are taught the necessary skills to cope with global software development. Unfortunately, the enormous amount of time that it takes to coordinate and support such activities can deter even the most devoted educator. This paper describes a course that used a computer supported collaborative tool help teach distributed teams from Turkey and the US how to work together to solve programming problems. The system contains both collaborative tools that support groups, as well as course management software for helping instructors with administrative tasks. Examples of the usage of the system and data collected from the undergraduate computer science course that used the software are presented. Based on that experience, future plans to refine the system for early detection of problem teams, and the advantages of implementing the software as a Web service are also discussed. This information is designed to provide support for effective multi-institutional learning courses

A Content-Boosted Collaborative Filtering Approach for Movie Recommendation Based on Local and Global Similarity and Missing Data Prediction

Most traditional recommender systems lack accuracy in the case where data used in the recommendation process is sparse. This study addresses the sparsity problem and aims to get rid of it by means of a content-boosted collaborative filtering approach applied to a web-based movie recommendation system. The main motivation is to investigate whether further success can be obtained by combining ‘local and global user similarity’ and ‘effective missing data prediction’ approaches, which were previously introduced and proved to be successful separately. The present work improves these approaches by taking the content information of the movies into account during the item similarity calculations. The comparison of the proposed approach with the original methods was carried out using mean absolute error, and more accurate predictions were achieved.