Skip to content

β3-IRT: A New Item Response Model and its Applications

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Standard

β3-IRT : A New Item Response Model and its Applications. / Chen, Yu; Filho, Telmo M Silva; Prudêncio, Ricardo B. C.; Diethe, Tom; Flach, Peter.

The 22nd International Conference on Artificial Intelligence and Statistics, 16-18 April 2019. ed. / Kamalika Chaudhuri; Masashi Sugiyama . Proceedings of Machine Learning Research, 2019. p. 1013-1021 (Proceedings of Machine Learning Research; Vol. 89).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Harvard

Chen, Y, Filho, TMS, Prudêncio, RBC, Diethe, T & Flach, P 2019, β3-IRT: A New Item Response Model and its Applications. in K Chaudhuri & M Sugiyama (eds), The 22nd International Conference on Artificial Intelligence and Statistics, 16-18 April 2019. Proceedings of Machine Learning Research, vol. 89, Proceedings of Machine Learning Research, pp. 1013-1021.

APA

Chen, Y., Filho, T. M. S., Prudêncio, R. B. C., Diethe, T., & Flach, P. (2019). β3-IRT: A New Item Response Model and its Applications. In K. Chaudhuri, & M. Sugiyama (Eds.), The 22nd International Conference on Artificial Intelligence and Statistics, 16-18 April 2019 (pp. 1013-1021). (Proceedings of Machine Learning Research; Vol. 89). Proceedings of Machine Learning Research.

Vancouver

Chen Y, Filho TMS, Prudêncio RBC, Diethe T, Flach P. β3-IRT: A New Item Response Model and its Applications. In Chaudhuri K, Sugiyama M, editors, The 22nd International Conference on Artificial Intelligence and Statistics, 16-18 April 2019. Proceedings of Machine Learning Research. 2019. p. 1013-1021. (Proceedings of Machine Learning Research).

Author

Chen, Yu ; Filho, Telmo M Silva ; Prudêncio, Ricardo B. C. ; Diethe, Tom ; Flach, Peter. / β3-IRT : A New Item Response Model and its Applications. The 22nd International Conference on Artificial Intelligence and Statistics, 16-18 April 2019. editor / Kamalika Chaudhuri ; Masashi Sugiyama . Proceedings of Machine Learning Research, 2019. pp. 1013-1021 (Proceedings of Machine Learning Research).

Bibtex

@inproceedings{7aa1a29f28e64a689a1d30bbf309d821,
title = "β3-IRT: A New Item Response Model and its Applications",
abstract = "Item Response Theory (IRT) aims to assess latent abilities of respondents based on the correctness of their answers in aptitude test items with different difficulty levels. In this paper, we propose the β3-IRT model, which models continuous responses and can generate a much enriched family of Item Characteristic Curves. In experiments we applied the proposed model to data from an online exam platform, and show our model outperforms a more standard 2PL-ND model on all datasets. Furthermore, we show how to apply β3-IRT to assess the ability of machine learning classifiers.This novel application results in a new metric for evaluating the quality of the classifier’s probability estimates, based on the inferred difficulty and discrimination of data instances.",
author = "Yu Chen and Filho, {Telmo M Silva} and Prud{\^e}ncio, {Ricardo B. C.} and Tom Diethe and Peter Flach",
year = "2019",
month = "3",
day = "10",
language = "English",
series = "Proceedings of Machine Learning Research",
publisher = "Proceedings of Machine Learning Research",
pages = "1013--1021",
editor = "Kamalika Chaudhuri and {Sugiyama }, Masashi",
booktitle = "The 22nd International Conference on Artificial Intelligence and Statistics, 16-18 April 2019",

}

RIS - suitable for import to EndNote

TY - GEN

T1 - β3-IRT

T2 - A New Item Response Model and its Applications

AU - Chen, Yu

AU - Filho, Telmo M Silva

AU - Prudêncio, Ricardo B. C.

AU - Diethe, Tom

AU - Flach, Peter

PY - 2019/3/10

Y1 - 2019/3/10

N2 - Item Response Theory (IRT) aims to assess latent abilities of respondents based on the correctness of their answers in aptitude test items with different difficulty levels. In this paper, we propose the β3-IRT model, which models continuous responses and can generate a much enriched family of Item Characteristic Curves. In experiments we applied the proposed model to data from an online exam platform, and show our model outperforms a more standard 2PL-ND model on all datasets. Furthermore, we show how to apply β3-IRT to assess the ability of machine learning classifiers.This novel application results in a new metric for evaluating the quality of the classifier’s probability estimates, based on the inferred difficulty and discrimination of data instances.

AB - Item Response Theory (IRT) aims to assess latent abilities of respondents based on the correctness of their answers in aptitude test items with different difficulty levels. In this paper, we propose the β3-IRT model, which models continuous responses and can generate a much enriched family of Item Characteristic Curves. In experiments we applied the proposed model to data from an online exam platform, and show our model outperforms a more standard 2PL-ND model on all datasets. Furthermore, we show how to apply β3-IRT to assess the ability of machine learning classifiers.This novel application results in a new metric for evaluating the quality of the classifier’s probability estimates, based on the inferred difficulty and discrimination of data instances.

M3 - Conference contribution

T3 - Proceedings of Machine Learning Research

SP - 1013

EP - 1021

BT - The 22nd International Conference on Artificial Intelligence and Statistics, 16-18 April 2019

A2 - Chaudhuri, Kamalika

A2 - Sugiyama , Masashi

PB - Proceedings of Machine Learning Research

ER -