Skip to content
OpenData in insurance area for Machine Learning Tasks
Branch: release
Clone or download
Latest commit b64d681 Jul 13, 2018
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
corpus release pair data Aug 6, 2017
deep_qa_1 Fix typo Aug 15, 2017
tmp Add deep_qa_1 as baseline Aug 12, 2017
visual Add deep_qa_1 as baseline Aug 12, 2017
.gitignore Add deep_qa_1 as baseline Aug 12, 2017
CODE_OF_CONDUCT.md Create CODE_OF_CONDUCT.md Aug 23, 2017
LICENSE Initial commit Jul 26, 2017
README.md Update README.md Jul 13, 2018
Requirements.txt

README.md

chatoper banner

保险行业语料库

详细文档

大家怎么说

看了下您的项目,我觉得这份数据可以用于保险领域的中文问答研究,对于较短的问题翻译很准确,长度较长的答案翻译就有些不连贯的问题,大体上关键词信息和一些上下文信息都有,我觉得是份很好的中文语料。 - fssqawj, East China Normal University

Excellent work! - rgtjf, East China Normal University

基线

Baseline model for insuranceqa-corpus-zh

mini-batch size = 100, hidden_layers = [100, 50], lr = 0.0001.

Epoch 25, total step 36400, accuracy 0.9031, cost 1.056221.

Deps

Python3+

pip install -r Requirements.txt

Run

A very simple network as baseline model.

python3 deep_qa_1/network.py
python3 visual/accuracy.py
python3 visual/loss.py

下载文档了解更多关于DeepQA-1的实现和Baseline。

声明

声明1 : insuranceqa-corpus-zh

本数据集使用翻译 insuranceQA而生成,代码发布证书 GPL 3.0。数据仅限于研究用途,如果在发布的任何媒体、期刊、杂志或博客等内容时,必须注明引用和地址。

InsuranceQA Corpus, Hai Liang Wang, https://github.com/Samurais/insuranceqa-corpus-zh, 07 27, 2017

任何基于insuranceqa-corpus衍生的数据也需要开放并需要声明和“声明1”和“声明2”一致的内容。

声明2 : insuranceQA

此数据集仅作为研究目的提供。如果您使用这些数据发表任何内容,请引用我们的论文:Applying Deep Learning to Answer Selection: A Study and An Open Task。Minwei Feng, Bing Xiang, Michael R. Glass, Lidan Wang, Bowen Zhou @ 2015

You can’t perform that action at this time.