Computer Engineering

This is a case study of developing and using community knowledge management software. The context of the case study is a school. During a training workshop staff at a school was introduced to a community knowledge management system. The intent was for them to use it for lesson planning and discussions. Instead, they built a series of applications that they could use to make the school administration more efficient. These applications consisted of Admission, Library, Store and Personnel Management. This case study demonstrates that the ICT need for the next billion is to provide them with systems that they can customize and manage for their needs. An anticipated domino effect would be that they help customize applications for their neighbors' needs, thereby narrowing a digital divide caused by the mystification of application development.

This paper presents an overview of the ImageCLEF 2014 evaluation lab. Since its first edition in 2003, ImageCLEF has become one of the key initiatives promoting the benchmark evaluation of algorithms for the annotation and retrieval of images in various domains, such as public and personal images, to data acquired by mobile robot platforms and medical archives. Over the years, by providing new data collections and challenging tasks to the community of interest, the Im-ageCLEF lab has achieved an unique position in the image annotation and retrieval research landscape. The 2014 edition consists of four tasks: domain adaptation, scalable concept image annotation, liver CT image annotation and robot vision. This paper describes the tasks and the 2014 competition, giving a unifying perspective of the present activities of the lab while discussing future challenges and opportunities.

This paper presents several challenges faced when annotating Turkish treebanks in accordance with the Universal Dependencies (UD) guidelines and proposes solutions to address them. Most of these challenges stem from the lack of adequate support in the UD framework to accurately represent null morphemes and complex derivations, which results in significant loss of information for Turkish. This loss negatively impacts the tools that are developed based on these treebanks. We raised and discussed these issues within the community on the official UD portal. This paper presents these issues and our proposals to more accurately represent morphosyntactic information for Turkish while adhering to guidelines of UD. This work aims to contribute to the representation of Turkish and other agglutinative languages in UD-based treebanks, which in turn aids to develop more accurately annotated datasets for such languages.

We present our PPI network visualization system RobinViz (Reliability Oriented Bioinformatic Networks Visualization). Clustering the PPI network based on GO annotations or biclustered gene expression data, providing a clustered visualization model based on a central/peripheral duality, computing layouts with algorithms specialized for interaction reliabilities represented as weights, completely automated data acquisition, processing are notable features of the system.

Motivation: Given protein-protein interaction (PPI) networks of a pair of species, a pairwise global alignment corresponds to a one-to-one mapping between their proteins. Based on the presupposition that such a mapping provides pairs of functionally orthologous proteins accurately, the results of the alignment may then be used in comparative systems biology problems such as function prediction/verification or construction of evolutionary relationships. Results: We show that the problem is NP-hard even for the case where the pair of networks are simply paths. We next provide a polynomial time heuristic algorithm, SPINAL, which consists of two main phases. In the first coarse-grained alignment phase, we construct all pairwise initial similarity scores based on pairwise local neighborhood matchings. Using the produced similarity scores, the fine-grained alignment phase produces the final one-to-one mapping by iteratively growing a locally improved solution subset. Both phases make use of the construction of neighborhood bipartite graphs and the contributors as a common primitive. We assess the performance of our algorithm on the PPI networks of yeast, fly, human and worm. We show that based on the accuracy measures used in relevant work, our method outperforms the state-of-the-art algorithms. Furthermore, our algorithm does not suffer from scalability issues, as such accurate results are achieved in reasonable running times as compared with the benchmark algorithms. Availability: Supplementary Document, open source codes, useful scripts, all the experimental data and the results are freely available at

Trafik sıkışıklığı, günümüz metropollerinin en büyük sorunlarından birisidir. Dünyada her gün milyonlarca saat trafik sebebiyle boşa gitmekte, çalışanların ve öğrencilerin verimi düşmektedir. Trafiği azaltmak için trafiğe katkıda bulunan etkenlerin tespit edilip çözüm yoluna gidilmesi elzemdir. Çözüme yönelmeden önce sorunun asıl kaynağı iyice tespit edilmelidir. Hatalı tasarlanan ulaşım altyapıları yahut plansız kentleşme bu kaynakların en önemlilerindendir. Bu çalışma, İstanbul'daki toplu taşıma ağının analiz edilerek şehirdeki yapısal darboğazların tespit edilmesini amaçlamaktadır. Otobüs güzergahlarının konumsal ağ şeklinde kurgulanıp analiz edilmesiyle trafik yoğunluğunun oluşabileceği noktalar tespit edilmiştir ve tespitler gerçek hayat deneyimleriyle uyumludur.

Traffic congestion is one of the major problems of metropolitan cities. Each day millions of hours are wasted due to traffic congestion in the world, causing reduction in productivity among workers and students. In order to reduce traffic congestion, it is required to figure out major factors contributing to the congestion. The faults in the design of the transportation infrastructures or unplanned urbanization are some among the most important factors. This study aims to detect bottlenecks in the transportation network of Istanbul via network analysis. We detected potential traffic congestion points using bus line route information represented as a spatial transportation network. Our detections are complying with our real life experience.

Background: In 2016, 44,965 people in the United States died by suicide. It is common to see people with suicidal ideation seek help or leave suicide notes on social media before attempting suicide. Many prefer to express their feelings with longer passages on forums such as Reddit and blogs. Because these expressive posts follow regular language patterns, potential suicide attempts can be prevented by detecting suicidal posts as they are written.
Objective: This study aims to build a classifier that differentiates suicidal and nonsuicidal forum posts via text mining methods applied on post titles and bodies.
Methods: A total of 508,398 Reddit posts longer than 100 characters and posted between 2008 and 2016 on SuicideWatch, Depression, Anxiety, and ShowerThoughts subreddits were downloaded from the publicly available Reddit dataset. Of these, 10,785 posts were randomly selected and 785 were manually annotated as suicidal or nonsuicidal. Features were extracted using term frequency-inverse document frequency, linguistic inquiry and word count, and sentiment analysis on post titles and bodies. Logistic regression, random forest, and support vector machine (SVM) classification algorithms were applied on resulting corpus and prediction performance is evaluated.
Results: The logistic regression and SVM classifiers correctly identified suicidality of posts with 80% to 92% accuracy and F1 score, respectively, depending on different data compositions closely followed by random forest, compared to baseline ZeroR algorithm achieving 50% accuracy and 66% F1 score.
Conclusions: This study demonstrated that it is possible to detect people with suicidal ideation on online forums with high accuracy. The logistic regression classifier in this study can potentially be embedded on blogs and forums to make the decision to offer real-time online counseling in case a suicidal post is being written.

We present our protein-protein interaction (PPI) network visualization system RobinViz (reliability-oriented bioinformatic networks visualization). Clustering the PPI network based on gene ontology (GO) annotations or biclustered gene expression data, providing a clustered visualization model based on a central/peripheral duality, computing layouts with algorithms specialized for interaction reliabilities represented as weights, completely automated data acquisition, processing are notable features of the system. Availability: RobinViz is a free, open-source software protected under GPL. It is written in C++ and Python, and consists of almost 30 000 lines of code, excluding the employed libraries. Source code, user manual and other Supplementary Material are available for download at

Motivation: Given protein-protein interaction (PPI) networks of a pair of species, a pairwise global alignment corresponds to a one-to-one mapping between their proteins. Based on the presupposition that such a mapping provides pairs of functionally orthologous proteins accurately, the results of the alignment may then be used in comparative systems biology problems such as function prediction/verification or construction of evolutionary relationships. Results: We show that the problem is NP-hard even for the case where the pair of networks are simply paths. We next provide a polynomial time heuristic algorithm, SPINAL, which consists of two main phases. In the first coarse-grained alignment phase, we construct all pairwise initial similarity scores based on pairwise local neighborhood matchings. Using the produced similarity scores, the fine-grained alignment phase produces the final one-to-one mapping by iteratively growing a locally improved solution subset. Both phases make use of the construction of neighborhood bipartite graphs and the contributors as a common primitive. We assess the performance of our algorithm on the PPI networks of yeast, fly, human and worm. We show that based on the accuracy measures used in relevant work, our method outperforms the state-of-the-art algorithms. Furthermore, our algorithm does not suffer from scalability issues, as such accurate results are achieved in reasonable running times as compared with the benchmark algorithms.

Cognitive radio is a new paradigm in the evolution of wireless communications to improve spectrum efficiency significantly by utilizing white spaces. In this paper, the mathematical model of the cognitive radio simulator that we develop is introduced and resource planning algorithms for cognitive radio networks is considered. A new resource planning method, which is a fixed assignment strategy, is introduced. The proposed method improves probability of blocking, dropping, and forced termination compared to uniform assignment.

The heterogeneous structure of Next Generation Wireless Systems (NGWS) makes admission control very complex. Accessibility of the subsystems at the time of connection or handoff request, availability of resources in the subsystems, user preferences, and connection class need to be considered in admission control. In this paper, we first give a general connection admission control algorithm. We also propose the first analytical model in the literature for admission control in NGWS. We point out the major challenges in modeling for NGWS and propose a neat solution to calculate state probabilities in a reasonable way even as the state space proliferates.

Cooperative spectrum sensing for dynamic spectrum access to unused licensed spectrum has been proposed as a solution to the spectrum scarcity problem. In this paper, a novel sensing algorithm, Cooperative Sensing with Decision Vector (CSDV), that employs uniform quantizer and a weight vector for global decision logic, is proposed. CSDV method is based on energy detection and performs better than conventional hard decision algorithms such as MAJORITY and M/N in terms of probability of detection and false alarm at the cost of a marginal increase in overhead bits. Probability of detection is maximized for a given probability of false alarm by proposed method via convex optimization. For detailed analysis, CSDV is compared with equal gain combiner scheme, which performs far better than hard decision algorithms, via highest bandwidth requirement. Moreover, robustness of CSDV method to sensing error is analyzed when some nodes always report false decisions to the fusion center.

In this paper, an analytical model for Infrastructurebased Cognitive Radio Networks is proposed and its performance evaluated. The architecture is composed of multiple cognitive radio base stations, and the state of the system depends on the activity of primary and secondary users in each cell. An analytical model for Cognitive Radio Network with multiple cells is developed and performance metrics like the probabilities of dropping and blocking for primary and secondary users as well as forced termination and forced frequency handoff for secondary users are investigated. The analytical model is also verified with simulations. The elementary events cause changes in the current state of the whole system. Cell dwell time and frequency holding time are used to determine the handoff behavior of the users. The proposed model is analyzed via simulations to estimate the performance metrics. Numerical results are obtained with different request rates, frequency holding times, cell dwell times, and also under heterogeneous cell load.

Abstract Molecular communication is a new communication paradigm that uses molecules for information transmission between nanomachines. Similar to traditional communication systems, several factors constitute limits over the performance of this communication system. One of these factors is the energy budget of the transmitter. It limits the rate at which the transmitter can emit symbols, i.e., produce the messenger molecules. In this paper, an energy model for the communication via diffusion system is proposed. To evaluate the performance of this communication system, first a channel model is developed, and also the probability of correct decoding of the information is evaluated. Two optimization problems are set up for system analysis that focus on channel capacity and data rate. Evaluations are carried out using the human insulin hormone as the messenger molecule and a transmitter device whose capabilities are similar to a pancreatic β-cell. Results show that distance between the transmitter and receiver has a minor effect on the achievable data rate whereas the energy budget's effect is significant. It is also shown that selecting appropriate threshold and symbol duration parameters are crucial to the performance of the system.