Yayınlar

Job recommendation systems face two critical challenges: treating users as a homogeneous group despite diverse job-seeking behaviors, and inadequately modeling the hierarchical nature of job markets. This study proposes Segment-Aware LorentzFM, a novel framework combining behavioral user segmentation with Lorentz-model hyperbolic embeddings. We analyze 1.6 million job applications from 103,896 users on Kariyer.net, revealing three distinct behavioral segments through unsupervised clustering: Ideal Candidates (38.1%) with focused patterns, Career Explorers (51.4%) with diverse behaviors, and Balanced Seekers (10.5%) with moderate exploration. We establish a theoretical foundation connecting segmentation with hyperbolic geometry, proving optimality through information-theoretic and geometric perspectives.

Hyperbolic embedding methods for collaborative filtering constrain all representations to the hyperboloid manifold, imposing a single geometry regardless of data characteristics. We introduce LightConeFM, which removes this constraint and allows embeddings to freely occupy any causal region of Lorentz-Minkowski space — timelike, lightlike, or spacelike — using only standard gradient descent without Riemannian optimization. Experiments on four real-world datasets reveal two consistent findings: unconstrained embeddings outperform their constrained counterparts on every dataset (up to +7.0% AUC), and the learned causal zone distribution predicts where hyperbolic geometry provides benefit over Euclidean alternatives.

Reciprocal recommendation requires satisfying preferences on both sides of a match, which differs from standard one-sided settings and often involves hierarchical structure (e.g., skills, seniority, education). We present biLorentzFM, a multi-objective framework that integrates hyperbolic geometry into factorization machine architectures using Lorentz embeddings with learnable curvature and manifold-aware optimization. On a large-scale recruitment dataset from Kariyer.net (1,150,302 interactions, 229,805 candidates), the model achieves candidate and company AUCs of 0.9964 and 0.9913 respectively, representing 6.6% and 6.0% improvements over the strongest Euclidean baseline while maintaining practical inference latency (2.1 ms per batch).

Digital job platforms vary dramatically in their interaction design, from single-click applications to multi-action marketplaces, yet the behavioral consequences of these design choices remain poorly understood. We present a large-scale comparative analysis of two job platforms with contrasting paradigms: Kariyer.net (single-action, 1.67M interactions, 104K users) and OLX Jobs (8-action marketplace, 65.5M interactions, 3.3M users). Contrary to conventional design wisdom, we find that platforms with richer interaction possibilities are associated with significantly lower user efficiency and goal achievement.

This paper presents a comprehensive analysis of a Wide & Deep Learning approach enhanced with graph features for job recommendation systems. Our methodology combines three key components: a wide linear model for memorization of sparse feature interactions, a deep neural network for generalization, and a graph-based feature extraction mechanism that captures job transition patterns. The model processes five distinct feature groups: categorical cross-features, embeddings of job positions, graph-based transition features, text representations from job descriptions, and numerical attributes.

In the modern era, job searching has evolved into a time-consuming and complex process for both job seekers and recruitment professionals. Recommendation systems technology has high potential to bring job seekers and job opportunities together based on their skills and qualifications. In this research, various recommendation system algorithms are analyzed in-depth using real data gathered from Kariyer.net, Turkey’s leading employment website. The paper proposes a cascade hybrid filtering approach: initially creating a broad recommendation pool through item-based collaborative filtering, and then filtering the pool through TF-IDF text similarity.

Computer-aided diagnosis (CAD) systems are widely used to diagnose breast cancer using mammography screening. In this research, we proposed a new multistage CAD system based on image decomposition with High-Dimensional Model Representation (HDMR), which is a divide-and-conquer algorithm. We used digital mammograms from the Digital Database for Screening Mammography as dataset. We neglected BIRADS classification and used a brand-new clustering based on HDMR constant and breast size. To find the best performance of the HDMR-based CAD system, we compared different pre-processing settings such as contrast enhancement with CLAHE and HDMR, feature extraction with HDMR, feature scaling, and dimension reduction with Linear Discriminant Analysis.

Bu çalışmada köşe yazısı uzunluğundaki yazılarda noktalama ve etkisiz kelime kullanım sıklığı gibi basit özniteliklerin yazar tanımada yeterli olduğu ortaya konmuştur. Cumhuriyet gazetesi yazarlarından sıkça köşe yazan 6 adedi seçilerek her birinin son 120 köşe yazısı alınmış, her bir yazı için bir takım etkisiz kelime ve noktalama işaretlerinin kullanım sıklıklarına dayanan dokuz adet öznitelik elde edilmiştir. Sekiz gözetimli yapay öğrenme algoritması eğitildikten sonra yazının yazarını tanıma başarısı ön işlemsiz ve ön işlemden geçirilmiş veri kümelerinde ayrı ayrı ölçülmüş, asgari %82 ve azami %92 olmak üzere yüksek isabetli sonuçlar elde edilmiştir.