1 1 29

Cihat Yıldız PRO

cihatyldz

AI & ML interests

Senior Data Scientist | MSc

Recent Activity

updated a Space 27 minutes ago

cihatyldz/mizan-fact-checker

published a Space 27 minutes ago

cihatyldz/mizan-fact-checker

posted an update about 10 hours ago

CV Matching Engine — a hybrid, explainable CV-to-job matching system — is now live on Spaces. It combines semantic similarity with ontology-based skill extraction to rank candidates against any job description, all within a single interactive Gradio app. The system uses a 4-component weighted scoring pipeline: 🔗 Sentence-BERT (all-MiniLM-L6-v2) for semantic similarity between CV and job description 🧩 Ontology-based skill matching with 400+ skills across 26 categories (from Programming Languages to Mechanical Engineering) using Jaccard similarity 📅 Experience year extraction via regex-based parsing 📍 Soft location scoring with spaCy NER — location informs but never eliminates candidates Key features include multi-entity extraction (universities, companies, languages), a built-in CV Assistant chatbot for natural-language queries about candidates, an XAI detail panel for full scoring transparency, interactive Plotly dashboards, Turkish university database with 200+ entries, dual-column CV support, and keyword/university filtering. Built with: Sentence-Transformers, spaCy, PyMuPDF, scikit-learn, Plotly, Gradio 🤗 Demo: https://huggingface.co/spaces/cihatyldz/cv-matching-engine

View all activity

Organizations

None yet

Posts 2

Post

CV Matching Engine — a hybrid, explainable CV-to-job matching system — is now live on Spaces. It combines semantic similarity with ontology-based skill extraction to rank candidates against any job description, all within a single interactive Gradio app.
The system uses a 4-component weighted scoring pipeline:

🔗 Sentence-BERT (all-MiniLM-L6-v2) for semantic similarity between CV and job description
🧩 Ontology-based skill matching with 400+ skills across 26 categories (from Programming Languages to Mechanical Engineering) using Jaccard similarity
📅 Experience year extraction via regex-based parsing
📍 Soft location scoring with spaCy NER — location informs but never eliminates candidates

Key features include multi-entity extraction (universities, companies, languages), a built-in CV Assistant chatbot for natural-language queries about candidates, an XAI detail panel for full scoring transparency, interactive Plotly dashboards, Turkish university database with 200+ entries, dual-column CV support, and keyword/university filtering.
Built with: Sentence-Transformers, spaCy, PyMuPDF, scikit-learn, Plotly, Gradio
🤗 Demo: cihatyldz/cv-matching-engine

Post

3447

Şifahane, a dual-inference medical classification demo, is now live on Spaces. It features side-by-side Turkish BERT and Qwen2.5 architectures for real-time evaluation of the "Classifier vs. LLM" trade-offs, all within a single space. The system utilizes a fine-tuned Turkish BERT for high-speed, cost-effective inference and the Qwen2.5-7B model for flexible multi-task reasoning, with support for department classification, condition analysis, urgency assessment, and rationale generation across 12 medical departments.

🧠 BERT model: https://lnkd.in/dCUUASqq
📊 Dataset: https://lnkd.in/dGK9y24w
🤗 Demo: https://lnkd.in/dtWjCCPF