cjymz886/text-cnn
A CNN model for Chinese text classification using Word2vec word embeddings trained with TensorFlow.
Not currently ranked — collecting fresh signals.
star history
This project implements a convolutional neural network for Chinese text classification, using pre-trained Word2vec embeddings as input features. The model classifies text into 10 categories (sports, finance, real estate, etc.) using the THUCNews dataset. It achieves 97.2% test accuracy, improving upon the baseline by incorporating word-level embeddings, multi-size convolutional kernels, and regularization techniques.
Frequently asked
- What is cjymz886/text-cnn?
- A CNN model for Chinese text classification using Word2vec word embeddings trained with TensorFlow.
- Is text-cnn open source?
- Yes — cjymz886/text-cnn is open source, released under the MIT license.
- What language is text-cnn written in?
- cjymz886/text-cnn is primarily written in Python.
- How popular is text-cnn?
- cjymz886/text-cnn has 448 stars on GitHub.
- Where can I find text-cnn?
- cjymz886/text-cnn is on GitHub at https://github.com/cjymz886/text-cnn.