Skip to main content

Question rewriting for conversational search based on unsupervised and supervised methods.

Project description

This is an official source code repository of the paper: Mele et al., Adaptive Utterance Rewriting for Conversational Search, IP&M, 2021.

Authors: Ida Mele, Cristina Ioana Muntean, Franco Maria Nardini, Raffaele Perego, Nicola Tonellotto, Ophir Frieder.

Code: https://github.com/hpclab/adaptive-utterance-rewriting-conversational-search

Data

  • Datasets: CAsT 2019 dataset divided into training and test sets as specified in the paper.

Query resolution

The query resolution module uses both supervised and unsupervised methods for rewriting queries.

For the supervised classification based rewriting run either of the following rewriting strategies "Standard", "Enriched", "LastSE", "First_and_Last_SE", "First_or_Last_SE":

from convrewriting.supervised.pipeline_classification import rewrite_strategy

rewrite_strategy("Standard")

For the unsupervised based rewriting strategy run either of the following:

from convrewriting.unsupervised.pipeline_nlp import rewrite_context, rewrite_first_topic, rewrite_topic_shift

rewrite_context()
rewrite_first_topic()
rewrite_topic_shift()

Citation Licence

This source code is subject to the following citation license:

By downloading and using the source code stored in this GitHub repository, you agree to cite at the undernoted paper in any kind of material you produce where this code has been used to conduct search or experimentation, whether be it a research paper, dissertation, article, poster, presentation, or documentation. By using this software, you have agreed to the citation licence.

Ida Mele, Cristina Ioana Muntean, Franco Maria Nardini, Raffaele Perego, Nicola Tonellotto, Ophir Frieder. 2021. Adaptive Utterance Rewriting in Conversational Search. Information Processing & Management.

@article{mele2021adaptive,
	title = {Adaptive Utterance Rewriting for Conversational Search},
	journal = {Information Processing \& Management},
	volume = {58},
	number = {6},
	pages = {102682},
	year = {2021},
	issn = {0306-4573},
	doi = {https://doi.org/10.1016/j.ipm.2021.102682},
	url = {https://www.sciencedirect.com/science/article/pii/S0306457321001679},
	author = {Ida Mele and Cristina Ioana Muntean and Franco Maria Nardini and Raffaele Perego and Nicola Tonellotto and Ophir Frieder},
}

Contact Us

For any other question/comment, write us an email: {cristina.muntean@isti.cnr.it}

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

convrewriting-0.0.1.tar.gz (5.1 MB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

convrewriting-0.0.1-py3-none-any.whl (5.5 MB view details)

Uploaded Python 3

File details

Details for the file convrewriting-0.0.1.tar.gz.

File metadata

  • Download URL: convrewriting-0.0.1.tar.gz
  • Upload date:
  • Size: 5.1 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.1 CPython/3.9.14

File hashes

Hashes for convrewriting-0.0.1.tar.gz
Algorithm Hash digest
SHA256 448b933cbd1e4ec290f37b70214bc7f5aa0cd3cbb5762def2e600e166722019f
MD5 3f4fb2d08cbd46b88016bc65a3495da5
BLAKE2b-256 2c6ce0710beec8fe80be50f4e6de31c418e494e42b3432647ceccdf81b98b99e

See more details on using hashes here.

File details

Details for the file convrewriting-0.0.1-py3-none-any.whl.

File metadata

  • Download URL: convrewriting-0.0.1-py3-none-any.whl
  • Upload date:
  • Size: 5.5 MB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.1 CPython/3.9.14

File hashes

Hashes for convrewriting-0.0.1-py3-none-any.whl
Algorithm Hash digest
SHA256 32fab77f58fa0d3a332ef5fa7bf2590822502956b6862a5389f0c89fe6fa62d8
MD5 64fdd506581e36e0114467cc2baa3979
BLAKE2b-256 ab4558cb90082ab7b16380ac7a0220f47ff613cb2b9e286ed974121792af43c7

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page