Bilgisayar Mühendisliği Bölümü Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.11779/1940
Browse
11 results
Search Results
Article Mention Detection in Turkish Coreference Resolution(Tubitak Scientific & Technological Research Council Turkey, 2024-09-23) Demir, Seniz; Akdag, Hanifi IbrahimA crucial step in understanding natural language is detecting mentions that refer to real-world entities in a text and correctly identifying their boundaries. Mention detection is commonly considered a preprocessing step in coreference resolution which is shown to be helpful in several language processing applications such as machine translation and text summarization. Despite recent efforts on Turkish coreference resolution, no standalone neural solution to mention detection has been proposed yet. In this article, we present two models designed for detecting Turkish mentions by using feed-forward neural networks. Both models extract all spans up to a fixed length from input text as candidates and classify them as mentions or not mentions. The models differ in terms of how candidate text spans are represented. The first model represents a span by focusing on its first and last words, whereas the representation also covers the preceding and proceeding words of a span in the second model. Mention span representations are formed by using contextual embeddings, part-of-speech embeddings, and named-entity embeddings of words in interest where contextual embeddings are obtained from pretrained Turkish language models. In our evaluation studies, we not only assess the impact of mention representation strategies on system performance but also demonstrate the usability of different pretrained language models in resolution task. We argue that our work provides useful insights to the existing literature and the first step in understanding the effectiveness of neural architectures in Turkish mention detection.Article Citation - WoS: 2Citation - Scopus: 4Unlocking the Neural Mechanisms of Consumer Loan Evaluations: an Fnirs and Mlbased Consumer Neuroscience Study(Frontiers Media SA, 2024-02-05) Girişken, Yener; Son, Semen; Demircioğlu, Esin Tuna; Filiz, Gözde; Çakar, Tuna; Ertuğrul, Seyit; Sayar, Alperen; Tuna, Esin; Son-Turan, SemenThis study conducted a comprehensive exploration of the neurocognitive processes underlying consumer credit decision-making using cutting-edge techniques from neuroscience and artificial intelligence (AI). Employing functional Near-Infrared Spectroscopy (fNIRS), the research examines the hemodynamic responses of participants while evaluating diverse credit offers. The study integrates fNIRS data with advanced AI algorithms, specifically Extreme Gradient Boosting, CatBoost, and Light Gradient Boosted Machine, to predict participants' credit decisions based on prefrontal cortex (PFC) activation patterns. Findings reveal distinctive PFC regions correlating with credit behaviors, including the dorsolateral prefrontal cortex (dlPFC) associated with strategic decision-making, the orbitofrontal cortex (OFC) linked to emotional valuations, and the ventromedial prefrontal cortex (vmPFC) reflecting brand integration and reward processing. Notably, the right dorsomedial prefrontal cortex (dmPFC) and the right vmPFC contribute to positive credit preferences. This interdisciplinary approach bridges neuroscience and finance, offering unprecedented insights into the neural mechanisms guiding financial choices. The study's predictive model holds promise for refining financial services and illuminating human financial behavior within the burgeoning field of neurofinance. The work exemplifies the potential of interdisciplinary research to enhance our understanding of human financial decision-making.Article Citation - WoS: 4Citation - Scopus: 8Unraveling Neural Pathways of Political Engagement: Bridging Neuromarketing and Political Science for Understanding Voter Behavior and Political Leader Perception(Frontiers Media SA, 2023-12-21) Çakar, Tuna; Filiz, GözdePolitical neuromarketing is an interdisciplinary field that combines marketing, neuroscience, and psychology to understand voter behavior and political leader perception. This interdisciplinary field offers novel techniques to understand complex phenomena such as voter engagement, political leadership, and party branding. This study aims to understand the neural activation patterns of voters when they are exposed to political leaders using functional near-infrared spectroscopy (fNIRS) and machine learning methods. We recruited participants and recorded their brain activity using fNIRS when they were exposed to images of different political leaders. This neuroimaging method (fNIRS) reveals brain regions central to brand perception, including the dorsolateral prefrontal cortex (dlPFC), the dorsomedial prefrontal cortex (dmPFC), and the ventromedial prefrontal cortex (vmPFC). Machine learning methods were used to predict the participants' perceptions of leaders based on their brain activity. The study has identified the brain regions that are involved in processing political stimuli and making judgments about political leaders. Within this study, the best-performing machine learning model, LightGBM, achieved a highest accuracy score of 0.78, underscoring its efficacy in predicting voters' perceptions of political leaders based on the brain activity of the former. The findings from this study provide new insights into the neural basis of political decision-making and the development of effective political marketing campaigns while bridging neuromarketing, political science and machine learning, in turn enabling predictive insights into voter preferences and behaviorArticle Citation - WoS: 3Citation - Scopus: 4A Novel Graph Transformation Strategy for Optimizing Sptrsv on Cpus(Wiley, 2023-05-07) Yılmaz, BuseSparse triangular solve (SpTRSV) is an extensively studied computational kernel. An important obstacle in parallel SpTRSV implementations is that in some parts of a sparse matrix the computation is serial. By transforming the dependency graph, it is possible to increase the parallelism of the parts that lack it. In this work, we present a novel graph transformation strategy to increase the parallelism degree of a sparse matrix and compare it to our previous strategy. It is seen that our transformation strategy can provide a speedup as high as 1.42x$$ 1.42x $$.Article Citation - Scopus: 1A New Benchmark Dataset for P300 Erp-Based Bci Applications(Academic Press Inc Elsevier Science, 2023-04-01) Çakar, Tuna; Özkan, Hüseyin; Musellim, Serkan; Arslan, Suayb S.; Yağan, Mehmet; Alp, NihanBecause of its non-invasive nature, one of the most commonly used event-related potentials in brain -computer interface (BCI) system designs is the P300 electroencephalogram (EEG) signal. The fact that the P300 response can easily be stimulated and measured is particularly important for participants with severe motor disabilities. In order to train and test P300-based BCI speller systems in more realistic high-speed settings, there is a pressing need for a large and challenging benchmark dataset. Various datasets already exist in the literature but most of them are not publicly available, and they either have a limited number of participants or utilize relatively long stimulus duration (SD) and inter-stimulus intervals (ISI). They are also typically based on a 36 target (6 x 6) character matrix. The use of long ISI, in particular, not only reduces the speed and the information transfer rates (ITRs) but also oversimplifies the P300 detection. This leaves a limited challenge to state-of-the-art machine learning and signal processing algorithms. In fact, near-perfect P300 classification accuracies are reported with the existing datasets. Therefore, one certainly needs a large-scale dataset with challenging settings to fully exploit the recent advancements in algorithm design (machine learning and signal processing) and achieve high-performance speller results. To this end, in this article we introduce a new freely-and publicly-accessible P300 dataset obtained using 32-channel EEG, in the hope that it will lead to new research findings and eventually more efficient BCI designs. The introduced dataset comprises 18 participants performing a 40 -target (5 x 8) cued-spelling task, with reduced SD (66.6 ms) and ISI (33.3 ms) for fast spelling. We have also processed, analyzed, and character-classified the introduced dataset and we presented the accuracy and ITR results as a benchmark. The introduced dataset and the codes of our experiments are publicly accessible at https://data .mendeley.com /datasets /vyczny2r4w.(c) 2023 Elsevier Inc. All rights reserved.Article Citation - WoS: 3Citation - Scopus: 3Array Bp-Xor Codes for Hierarchically Distributed Matrix Multiplication(IEEE, 2022-03-01) Arslan, Şuayb ŞefikA novel fault-tolerant computation technique based on array Belief Propagation (BP)-decodable XOR (BP-XOR) codes is proposed for distributed matrix-matrix multiplication. The proposed scheme is shown to be configurable and suited for modern hierarchical compute architectures such as Graphical Processing Units (GPUs) equipped with multiple nodes, whereby each has many small independent processing units with increased core-to-core communications. The proposed scheme is shown to outperform a few of the well–known earlier strategies in terms of total end-to-end execution time while in presence of slow nodes, called stragglers. This performance advantage is due to the careful design of array codes which distributes the encoding operation over the cluster (slave) nodes at the expense of increased master-slave communication. An interesting trade-off between end-to-end latency and total communication cost is precisely described. In addition, to be able to address an identified problem of scaling stragglers, an asymptotic version of array BP-XOR codes based on projection geometry is proposed at the expense of some computation overhead. A thorough latency analysis is conducted for all schemes to demonstrate that the proposed scheme achieves order-optimal computation in both the sublinear as well as the linear regimes in the size of the computed product from an end-to-end delay perspective.Article Citation - WoS: 5Citation - Scopus: 7Founsure 1.0: an Erasure Code Library With Efficient Repair and Update Features(Elsevier, 2021-01-01) Arslan, Şuayb ŞefikFounsure is an open-source software library that implements a multi-dimensional graph-based erasure coding entirely based on fast exclusive OR (XOR) logic. Its implementation utilizes compiler optimizations and multi-threading to generate the right assembly code for the given multi-core CPU architecture with vector processing capabilities. Founsure possesses important features that shall find various applications in modern data storage, communication, and networked computer systems, in which the data needs protection against device, hardware, and node failures. As data size reached unprecedented levels, these systems have become hungry for network bandwidth, computational resources, and average consumed power. To address that, the proposed library provides a three-dimensional design space that trades off the computational complexity, coding overhead, and data/node repair bandwidth to meet different requirements of modern distributed data storage and processing systems. Founsure library enables efficient encoding, decoding, repairs/rebuilds, and updates while all the required data storage and computations are distributed across the network nodes.Article Citation - WoS: 1Citation - Scopus: 2A Novel Genetic Algorithm-Based Improvement Model for Online Communities and Trust Networks(IOS Press, 2021-01-04) Bekmezci, ilker; Cimen, Egemen Berkic; Ermiş, MuratSocial network analysis offers an understanding of our modern world, and it affords the ability to represent, analyze and even simulate complex structures. While an unweighted model can be used for online communities, trust or friendship networks should be analyzed with weighted models. To analyze social networks, it is essential to produce realistic social models. However, there are serious differences between social network models and real-life data in terms of their fundamental statistical parameters. In this paper, a genetic algorithm (GA)-based social network improvement method is proposed to produce social networks more similar to real-life data sets. First, it creates a social model based on existing studies in the literature, and then it improves the model with the proposed GA-based approach based on the similarity of the average degree, the k-nearest neighbor, the clustering coefficient, degree distribution and link overlap. This study can be used to model the structural and statistical properties of large-scale societies more realistically. The performance results show that our approach can reduce the dissimilarity between the created social networks and the real-life data sets in terms of their primary statistical properties. It has been shown that the proposed GA-based approach can be used effectively not only in unweighted networks but also in weighted networks.Article Citation - Scopus: 1On the Distribution of the Threshold Voltage in Multi-Level Cell Flash Memories(Elsevier, 2019-10-01) Pusane, Ali E; Ashrafi, Reza A; Arslan, Şuayb ŞefikIn Multi-Level Cell (MLC) memories, multiple bits of information are packed within the cell to enable higher capacity and lower cost of manufacturing compared to those of the single-level cell flash. However, because of heavy information packing, MLC memories suffer from several error sources including inter-cell interference, retention error, and random telegraph noise which make their lifetime shorter. Having so many error sources that are statistically hard to characterize makes it challenging to properly derive the underlying probability distribution of the sensed threshold voltage, which is vital for finding optimal decision rules to secure better detection performance and hence better lifetime. Although several recent works have already considered this problem, they mostly recourse to few loose assumptions that are far from being realistic. In this study, a more comprehensive/general analysis is conducted to derive the probability density function of the final sensed voltage, and through realistic simplifications, closed form expressions are presented. Extensive computer simulations corroborate the accuracy of the derived analytical expressions, and we think they shall be essential for accurately estimating the reliability and the overall lifetime of modern MLC memories.Article Citation - WoS: 52Citation - Scopus: 66An Investigation of the Neural Correlates of Purchase Behavior Through Fnirs(Emerald Group Publishing Ltd, 2018-02-12) Cakir, Murat Perit; Yurdakul, Dicle; Girisken, Yener; Çakar, TunaPurpose This study aims to explore the plausibility of the functional near-infrared spectroscopy (fNIRS) methodology for neuromarketing applications and develop a neurophysiologically-informed model of purchasing behavior based on fNIRS measurements. Design/methodology/approach The oxygenation signals extracted from the purchase trials of each subject were temporally averaged to obtain average signals for buy and pass decisions. The obtained data were analyzed via both linear mixed models for each of the 16 optodes to explore their separate role in the purchasing decision process and a discriminant analysis to construct a classifier for buy/pass decisions based on oxygenation measures from multiple optodes. Findings Positive purchasing decisions significantly increase the neural activity through fronto-polar regions, which are closely related to OFC and vmPFC that modulate the computation of subjective values. The results showed that neural activations can be used to decode the buy or pass decisions with 85 per cent accuracy provided that sensitivity to the budget constraint is provided as an additional factor. Research limitations/implications The study shows that the fNIRS measures can provide useful biomarkers for improving the classification accuracy of purchasing tendencies and might be used as a main or complementary method together with traditional research methods in marketing. Future studies might focus on real-time purchasing processes in a more ecologically valid setting such as shopping in supermarkets. Originality/value This paper uses an emerging neuroimaging method in consumer neuroscience, namely, fNIRS. The decoding accuracy of the model is 85 per cent which presents an improvement over the accuracy levels reported in previous studies. The research also contributes to existing knowledge by providing insights in understanding individual differences and heterogeneity in consumer behavior through neural activities.