BPBekir Parlak

Bekir Parlak

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Bekir Parlak has a total of 6 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Feature Encoding and Duplication in Gielis Curve Classification: A Superelliptic Frenet Perspective on Robustness

    by Esra Parlak, Bekir Parlak

    Feature encoding and duplication can affect classifier robustness even when representations contain equivalent information. We study these effects using 2500 synthetic three-dimensional Gielis curves obtained by radially scaling a fixed helix. A twelve-component superelliptic Frenet representation can be reconstructed from four radial statistics and one endpoint-sensitive angular moment. Nine representations are compared using support vector machines and random forests, five stratified holdouts, training-only cross-validation, and three paired radial-noise realizations per holdout. All configurations perfectly classify the original clean test data. At δ=0.10, forests using the full descriptor, radial mean, and radial root-mean-square (RMS) attain 99.67±0.11%, 99.97±0.04%, and 99.81±0.10% accuracy, respectively. Repeating the radial RMS eight times also attains 99.81±0.10%, showing sensitivity to redundant features. Dispersions are sample standard deviations across holdout means. A separate clean benchmark with shared exponent ranges and frequency-based labels separates the representations under fixed classifier settings. A sampled Euclidean square-root velocity function (SRVF) control is sensitive to noise but performs well on this harder clean task. The results do not establish a unique Frenet-specific robustness advantage and are limited to the tested synthetic datasets and perturbations.

  • Integrating Analysis, Cyber Security, and Policy: Securing the Security Intelligence for Effective Counterterrorism in a Complex Digital Age

    by Muhammet Mağat, Bekir Parlak

    Intelligence is crucial for shaping both domestic and foreign policy, particularly in volatile regions with pressing security concerns. Objective analysis, free from political bias, is vital for accurate threat assessment and effective counter-terrorism strategies. This is especially important during crises, where understanding the nature of threats– whether from state or non-state actors – is essential for informing decisions on interventions, terrorism escalation, and terrorist profiling. A robust counter-terrorism strategy relies on strong “political security” and necessitates increased information sharing and cooperation on issues like terrorism, smuggling, and threats to national integrity. This strategy must adapt to the “technology factor”, acknowledging how terrorist organizations exploit electronic platforms for everything from propaganda to attacks. This study focuses on “security intelligence” and its role in counterterrorism. Using a qualitative, explanatory-descriptive approach, it examines the intelligence analytical culture, the impact of technology, and the specific function of security intelligence in combating terrorism. It argues that prioritizing intelligence efforts targeting state and non-state actors supporting terrorism is crucial for developing effective counter-terrorism policies in today’s interconnected world.

  • Open Access

    Analysis of National Security Advisory System in True Communication of National Security Issues

    by Muhammet Mağat, Bekir Parlak

    The media has an influence on the stages of national security policy formulation and decision-making. In this context, the national security advisory system is also effective in policy adaptation. National security is inherently filled with case studies as part of a broader whole. The accurate reporting of these case studies is linked to the political communication skills of national security advisory. The study's methodology involves a theoretical foundation review. The scope of the study covers topics such as political communication, news media, and the role of journalists and academics in the formulation of national security policy, as well as media and national security advisory. The study focuses on the concept of "post-truth," particularly how emotions and beliefs influence public opinion, especially when making decisions or interpreting events, within the main debate around national security and the media. The study centers on the question: What is the scene in the media-national security relationship? The context reached in the study emphasizes the ongoing need for rational bureaucratic systems in clarifying and understanding the truth in the field of national security.


  • Open Access

    Küresel ve Bölgesel Terörizme Karşı Türk Devletleri Teşkilatı: Analiz ve Operasyon Düzeyi

    by Bekir Parlak, Muhammet Mağat

    Terörle mücadelede ulusal ve uluslararası anlamda yoğun bir devlet ve devletler üstü sistemlerin varlığı, başarıya ulaşılmasında önemli bir rol oynamaktadır. Uluslararası güvenlik sisteminin yeni ve etkili bir aktörü olduğu değerlendirilen Türk Devletleri Teşkilatının, bu anlamda küresel ve bölgesel terörizme karşı analiz ve operasyon düzeylerinde, uluslararası düzenin geleceğini etkileyecek bir mekanizma olduğu düşünülmektedir. Bu bağlamda Türk Devletleri Teşkilatı, bir güvenlik topluluğu olarak, üye devletlerin ortak paydada olduğu bir politika üretme aygıtıdır. Ortak terör tehditleri çerçevesinde, üye devletlerin, Türk Devletleri Teşkilatını temsilen uluslararası güvenlik doktrininin yeniden şekillenmesinde aktif olmaları, bu çalışmada analiz edilmiştir. Bu çalışmada amaç, Türk Devletleri Teşkilatının “bir uluslararası güvenlik örgütü” olarak tanımlanabilmesinin tartışmasını sunabilmektir. Çalışmanın geneli, Türk Devletleri Teşkilatı nezdinde “terörle mücadele, istihbarat, uluslararası güvenlik doktrini, operasyonel kod/kapasite” gibi konularla çerçevelenmiştir. Çalışmanın birinci bölümünde, Teşkilatın terörle mücadelede istihbarat yönetimi ve görev gücü; ikinci bölümünde, Teşkilatın uluslararası güvenlik doktrini ve üçüncü bölümde de Teşkilatın terörle mücadele kurulu mekanizması düzeyinde olası operasyonlar konuları irdelenmiştir. Sonuç olarak varılan nokta, Türk Devletleri Teşkilatının bir güven ve güvenlik artırıcı fonksiyonu sürdürmesi ve bu doğrultuda Türkiye-Azerbaycan-Özbekistan üçlü kurucu gücün aktif rolünü geliştirmesidir.


  • Open Access

    Comparative Analysis of Globalisation Techniques for Medical Document Classification

    by Bekir Parlak, Salih Berkan Aydemi̇r

    Medical document classification is one of the important topics of text mining. Globalisation techniques play a major role in text classification. It is also known that globalisation techniques play an important role in text classification. Our aim in the study is to conduct a detailed analysis on two data sets with English and Turkish content by using medical text summaries of Turkish articles. These datasets consist of Turkish and English text summaries of the same articles. To observe how successful local feature selection methods in the field of text classification affect the classification performance on these two equivalent data sets by applying different globalisation techniques. The feature selection methods used are CHI2, MI, OR, WLLR. Globalisation techniques are SUM, AVG, MAX. Classifiers are MNB, DT, and SVM.

  • The Effects of Preprocessing on Turkish and English News Data

    by Bekir Parlak

    In a standard text classification (TC) study, preprocessing is one of the key components to improve performance. This study aims to look at how preprocessing effects TC according to news text, text language, and feature selection. All potential combinations of commonly used preprocessing techniques are compared on one domain, namely news data, and in two different news datasets for this aim. Preprocessing technique contributions to classification performance at multiple feature sizes, possible interconnections among these techniques, and technique dependency on corresponding languages are all evaluated in this way. Using best combinations of preprocessing techniques rather than using or not using them all, experimental studies on public datasets reveals that, choosing best combinations of preprocessing techniques can improve classification accuracy significantly.


  • Open Access

    The Role of Feature Weighting Methods on Local Feature Selection Methods for Text Classification

    by Bekir Parlak

    İnternet teknolojilerinin gelişimiyle birlikte metinsel verilerde ciddi bir artış yaşanmıştır. Bu metinsel verilerin anlamlı hale gelebilmesi için otomatik metin sınıflandırma yaklaşımları önemli hale gelmiştir. Otomatik metin sınıflandırma yaklaşımlarında öznitelik seçimi ve öznitelik ağırlıklandırma önemli bir yer tutar. Bu çalışmada, öznitelik ağırlıklandırma metotlarının lokal öznitelik seçim metotları üzerindeki etkisi ayrıntılı bir şekilde incelenmiştir. Çalışmada iki farklı ağırlıklandırma metodu, üç farklı lokal öznitelik seçim metodu, üç farklı kriter veri kümesi ve iki sınıflandırıcı kullanılmıştır. En yüksek Mikro-F1 ve Makro-F1 skoru, Reuters-21578 veri kümesi için 92.88 ve 65.55, 20Newsgroup veri kümesi için 99.02 ve 98.15, Enron1 veri kümesi için 97.19 ve 93.40’tır. Deneysel sonuçlar, OddsRatio (OR) öznitelik seçim metodu, Terim Frekansı (TF) öznitelik ağırlıklandırma ve Destek Vektör Makinesi (DVM) sınıflandırıcı kombinasyonu ile daha iyi sonucun elde edildiğini göstermektedir.