Skip to content

AgriOSLab/GANA

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

24 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

A Generative Adversarial Net Assisted Method for User Intention Recognition on Imbalanced Dataset

We will add more descriptions after our paper "A Generative Adversarial Net Assisted Method for User Intention Recognition on Imbalanced Dataset" been published.

this work is published in 2022 IEEE International Conference on Knowledge Graph (ICKG), you may site this by:

[1]. Z. Li, Z. Jin, R. Wang, J. Ji and H. Liu, "A Generative Adversarial Net Assisted Method for User Intention Recognition on Imbalanced Dataset," 2022 IEEE International Conference on Knowledge Graph (ICKG), Orlando, FL, USA, 2022, pp. 157-163, doi: 10.1109/ICKG55886.2022.00027. paperlink

Dataset

  1. the dataset used in our paper is collected by ourself, and we now supply it for everyone.
  2. the dataset consists of questions in agricultural domain.
  3. the dataset is divided into train, dev and test set.
class description Train Dev Test
0 Ask for material 12 20 20
1 Ask for sub region 259 37 72
2 Ask for products something could be made 8 13 12
3 Ask for product place 389 55 110
4 Ask for super classes 2589 369 738
5 Ask for instances of given class 194 27 54
6 Ask for son classes 8 14 12
7 Ask for locations 2231 318 636
8 Others 147 21 40

Code

  1. Config.py: the configuration;
  2. dowmModel.py: dowmload pre-trained language model on local disk;
  3. Model.py: models;
  4. main.py: traditional text classification method based on pre-trained language model, e.g. BERT;
  5. GAN.py: our method
  6. utils.py: functions for loading dataset and calculatinng scores

Details of Model.py

  1. BertClassification: Model using BERT for classification task, using the [CLS] embedding and fully connection layer, output logits;
  2. BertBase: Bert model that outputs the [CLS] embedding;
  3. GANBert: our model contains three part: BertBase, Generator and Discriminator;
  4. Generator: to generate vectors with minority classes vector as input;
  5. Discriminator: a fully connection layer to classify sentence.

Our Model

Generative Adversarial Net Assisted (GANA) Method for User Intention Recognition

model

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages