MNID

Code Repository for the paper 'A Framework to Generate High-quality Datapoints for Multiple Novel Intent Detection' accepted at Findings of NAACL, 2022.

The method consists of two main parts:

1. Out of Domain Intent Detection (OODD)
2. Multiple Novel Intent Detection (MNID)

Requirements

Requires an environment supporting python 3.6 and also need the following:

1. pip install tensorflow
2. pip install sentence-transformers
3. pip install scipy
4. pip install numpy

Citation

Please cite the work if you would like to use it.

@misc{https://doi.org/10.48550/arxiv.2205.02005,
  doi = {10.48550/ARXIV.2205.02005},
  url = {https://arxiv.org/abs/2205.02005},
  author = {Mullick, Ankan and Purkayastha, Sukannya and Goyal, Pawan and Ganguly, Niloy},
  keywords = {Computation and Language (cs.CL), Artificial Intelligence (cs.AI), FOS: Computer and information sciences, FOS: Computer and information sciences},
  title = {A Framework to Generate High-Quality Datapoints for Multiple Novel Intent Detection},
  publisher = {arXiv}, 
  year = {2022}, 
  copyright = {Creative Commons Attribution 4.0 International}
}

For any queries kindly contact:

1. Ankan Mullick (aankanmullick@gmail.com)
2. Sukannya Purkayastha (purkayasthasukannya020@gmail.com)