همایش بین المللی هوش مصنوعی

صفحه اصلی / دومین همایش بین المللی هوش مصنوعی

Benchmarking Persian Proofreading Tools: A Fine-grained, Error-Typed Evaluation

نویسندگان :

Zahra Saaberi¹ Mehrnoush Shamsfard²

1- Shahid Beheshti University 2- Shahid Beheshti University

کلمات کلیدی :

Proofreading tools،fundamental NLP tools،Persian،benchmarking

چکیده :

We present a dedicated benchmark, ParsProof, for systematic evaluation of Persian proofreading systems, featuring fine-grained annotations across 52 error types tailored to the specific linguistic properties of Persian language. Unlike most of the existing datasets, our benchmark enables detailed error-level analysis. We evaluated seven widely used Persian proofreading tools—Virastyar, Negar, Virastman, Farsiyar, Paknevis, Viravirast, and Neviise—as well as GPT-4.1. Among the specialized proofreading systems, Paknevis achieved the highest F1 score. GPT-4.1, however, generated the largest number of correct edits overall, though it also introduced more unnecessary or erroneous changes. The benchmark highlights the strengths and limitations of current tools at a granular level, providing a robust foundation for developing more accurate and linguistically informed Persian proofreading systems.

لیست مقالات

لیست مقالات بایگانی شده

Mamba Meets Sleep: Do State Space Models Outperform CNNs for EEG Classification?

Mostafa Mehrabi - Hamed Malek

Hybrid ANN and Ant Colony Algorithm for IoT Data Classification

Khadejeh Nemati - Safouro Ashoori - Moohamad hadi Amini

Finite element model updating using computational intelligence - based methods: A Case study of the San Roque Valley bridge, California

Fatemeh Z Dokhanian - Negin G.sabagh - Mahmoud R Shiravand

Fibroglandular Tissue Classification in Breast MRI: A Comparative Study of Automated Decision Strategies

Meysam Khalaj - Arvin Arian - Ala Torabi - Nasrin Ahmadinejad - Masoumeh Gity - Seyedeh Nooshin Miratashi Yazdi - Mohammad Pooya Afshari - Melika Sadeghi Tabrizi - Hamid Soltanian-Zadeh

Intelligent Leaf Disease Diagnosis Using EfficientNetB0 Combined with CBAM in a Multi-Task Framework

Nasrin Nikootadbir - Mohammad Hasan Majidi

Persian Intelligent Assistant in Healthcare Domain

Sarina Chitsaz - Mehrnoush Shamsfard

AdaLayer-CAM: Adaptive Multi-Layer Method for Visualization and Explanation of CNNs

MohammadJavad GhaderiPoor - Latifeh PourMohammadBagher - Zahra Rahimi

NeuroPC-KGNN: A Causality-Driven Graph Neural Network Framework for Early Diagnosis of Parkinson’s Disease

Fatemeh Ahouz - Ahmad Ali Abin - Amin Golabpour

Deep Learning Frailty Model for Heart Failure Survival Prediction

Solmaz Norouzi - Mohammad Asghari Jafarabadi - Ebrahim Hajizadeh - Hossein Khormaei - Nasim Naderi

Hybrid Machine Learning and Deep Learning Approach for Minimum Airfare Prediction

Mohammad Safaiyan - Ali Mahvash Mohammadi

بیشتر