QAmodel for Retrievalchatbot
1.0.0
Many students have sent me emails to inquire about the corpus of QA_demo1. I do not have the right to make this corpus public for the time being. I hope you can forgive me!
The format of the corpus of QA_demo1 is: QQ, QA. In fact, it mainly depends on the QQ data set, because what is done is the similarity between questions. This is done in the industry. QA is only used to show the corresponding effects of the model. , you can go to any website with FAQ instructions to download the QA pair.
Here are 2 Chinese data sets for your reference. I hope it can help you:
1. Question and answer robot based on features such as tf-idf. 2. Question and answer robot based on semantic models, such as CNN, rnn and other deep learning models. 3. Question and answer robot based on ELMO. 4. Question and answer robot based on BERT.
| Project name | data type | technology type | Visualization | completion time |
|---|---|---|---|---|
| Question and answer robot based on tf-idf | Chinese | tf-idf, feature matching | no | 2019/4/4 |
| Question and answer robot based on recall+rerank | Chinese | tf-idf, cnn | no | 2019/7/22 |
| Chatbot Xiaotian 1.0 | Chinese | Route conversion mechanism supports chatting and FAQ task questions and answers | no | 2019/7/25 |
| Question and answer robot based on BERT | Chinese | / | no | Stay tuned |
If you think my work is helpful to you, please don't be stingy with the little star in the upper right corner! Welcome to Fork and Star! You are also welcome to build this project together!
Q&A related items will be updated when there is time. Interested students can follow it.
Please leave a message in Issues or email [email protected]
Cite
If you used QAmodel-for-Retrievalchatbot in your research, please cite it in the following format:
@software{QR-Chatbot,
author = {ZhengWen Xie},
title = {QR-Chatbot: QAmodel for Retrievalchatbot},
year = {2019},
url = {https://github.com/WenRichard/QAmodel-for-Retrievalchatbot},
}
