1 1 32

Cihat Yıldız PRO

cihatyldz

AI & ML interests

Senior Data Scientist | MSc

Recent Activity

posted an update about 11 hours ago

Çarşı, a dual-model Turkish e-commerce NER demo, is now live on Spaces. It performs side-by-side Named Entity Recognition using a fine-tuned Turkish BERT and Qwen2.5-7B zero-shot extraction for real-time evaluation of the "Specialized NER vs. Zero-shot LLM" trade-offs. The system extracts 7 entity types from Turkish e-commerce text — PRODUCT, BRAND, PRICE, COLOR, SIZE, MATERIAL, and GENDER — with color-coded entity highlighting. The fine-tuned BERT delivers ~10-50ms inference on CPU while the LLM handles flexible extraction via HF Inference API, trained on 8,000 synthetic Turkish e-commerce sentences with BIO-tagged token classification. 🧠 BERT model: https://huggingface.co/cihatyldz/carsi-bert-turkish-ecommerce-ner 📊 Dataset: https://huggingface.co/datasets/cihatyldz/carsi-turkish-ecommerce-ner 🤗 Demo: https://huggingface.co/spaces/cihatyldz/carsi-ecommerce-ner

updated a Space about 12 hours ago

cihatyldz/carsi-ecommerce-ner

published a Space about 12 hours ago

cihatyldz/carsi-ecommerce-ner

View all activity

Organizations

None yet

Posts 2

Post

Çarşı, a dual-model Turkish e-commerce NER demo, is now live on Spaces. It performs side-by-side Named Entity Recognition using a fine-tuned Turkish BERT and Qwen2.5-7B zero-shot extraction for real-time evaluation of the "Specialized NER vs. Zero-shot LLM" trade-offs. The system extracts 7 entity types from Turkish e-commerce text — PRODUCT, BRAND, PRICE, COLOR, SIZE, MATERIAL, and GENDER — with color-coded entity highlighting. The fine-tuned BERT delivers ~10-50ms inference on CPU while the LLM handles flexible extraction via HF Inference API, trained on 8,000 synthetic Turkish e-commerce sentences with BIO-tagged token classification.

🧠 BERT model: cihatyldz/carsi-bert-turkish-ecommerce-ner
📊 Dataset: cihatyldz/carsi-turkish-ecommerce-ner
🤗 Demo: cihatyldz/carsi-ecommerce-ner

Post

3495

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