ToModAPI: A topic modeling API to train, use and compare topic models

Lisena, Pasquale; Harrando, Ismail; Kandakji, Oussama; Troncy, Raphaël
NLP-OSS 2020, 2nd Workshop for Natural Language Processing Open Source Software, 19 November 2020 (Virtual Workshop)

From LDA to neural models, different topic modeling approaches have been proposed in
the literature. However, their suitability and performance is not easy to compare, particularly when the algorithms are being used in the wild on heterogeneous datasets. In this paper, we introduce ToModAPI (TOpic MOdeling API), a wrapper library to easily train, evaluate and infer using different topic modeling algorithms through a unified interface. The library is extensible and can be used in Python environments or through a Web API.

DOI
HAL
Type:
Conference
Date:
2020-11-19
Department:
Data Science
Eurecom Ref:
6371
Copyright:
Copyright ACL. Personal use of this material is permitted. The definitive version of this paper was published in NLP-OSS 2020, 2nd Workshop for Natural Language Processing Open Source Software, 19 November 2020 (Virtual Workshop) and is available at : http://dx.doi.org/10.18653/v1/2020.nlposs-1.19

PERMALINK : https://www.eurecom.fr/publication/6371