Domain Adaptation in law:professional fact descriptions to non-professional fact descriptions

Citation Author(s):
Guangyi
Xiao
Xinlong
Liu
Xu
Han
Submitted by:
Han Xu
Last updated:
Thu, 07/16/2020 - 02:52
DOI:
10.21227/3gaa-2s33
License:
76 Views
Categories:
Keywords:
0
0 ratings - Please login to submit your rating.

Abstract 

This dataset concludes source data which is professional fact descriptions from Chinese law office website and target data which is non-professional fact descriptions from daily spoken language.

The dataset is for transfer learning in law domain.
The dataset also concludes processed dictionary and .npy files.
The task of transfer learning is to predict the accusation based on the description.

Instructions: 

The dataset concludes raw law articles,use various word_embedding to process them will get word segmentation results.
The number of classes is the number of labers we want to predict,we set kinds of numbers to test.