Sensitivity Analysis of a BERT-based scholarly recommendation system

With the exponential growth of publicly available datasets, a scholarly recommendation system of datasets would be an essential tool in the field of information filtering. Recommending datasets to users can be formulated as a classification problem where deep learning models can be carefully trained...

Full description

Saved in:
Bibliographic Details
Main Authors: Jie Zhu, Hulin Wu, Ashraf Yaseen
Format: Article
Language:English
Published: LibraryPress@UF 2022-05-01
Series:Proceedings of the International Florida Artificial Intelligence Research Society Conference
Subjects:
Online Access:https://journals.flvc.org/FLAIRS/article/view/130595
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849763277148520448
author Jie Zhu
Hulin Wu
Ashraf Yaseen
author_facet Jie Zhu
Hulin Wu
Ashraf Yaseen
author_sort Jie Zhu
collection DOAJ
description With the exponential growth of publicly available datasets, a scholarly recommendation system of datasets would be an essential tool in the field of information filtering. Recommending datasets to users can be formulated as a classification problem where deep learning models can be carefully trained. In such a case, when preparing training data for the learning models, one needs to consider different ratios of false and true pairs. Therefore, a sensitivity analysis is necessary. In this work, we conduct a sensitivity analysis using different class ratios on a deep learning model (BERT) for recommending datasets. We found out that our BERT-based recommender model is relatively robust using recommender metrics such as Mean Reciprocal Rank (MRR)@k, Recall@k, etc., except for the extreme class imbalance case (1:5000). Therefore, we conclude that a moderate ratio of the random negative sampling scheme, (in our case 1:10) is reasonable, sufficient and time-efficient in the recommendation system training
format Article
id doaj-art-deb6cfff35644db8a06fe757af9df239
institution DOAJ
issn 2334-0754
2334-0762
language English
publishDate 2022-05-01
publisher LibraryPress@UF
record_format Article
series Proceedings of the International Florida Artificial Intelligence Research Society Conference
spelling doaj-art-deb6cfff35644db8a06fe757af9df2392025-08-20T03:05:26ZengLibraryPress@UFProceedings of the International Florida Artificial Intelligence Research Society Conference2334-07542334-07622022-05-013510.32473/flairs.v35i.13059566794Sensitivity Analysis of a BERT-based scholarly recommendation systemJie Zhu0Hulin Wu1Ashraf Yaseen2The University of Texas Health Science Center at HoustonThe University of Texas Health Science Center at HoustonThe University of Texas Health Science Center at HoustonWith the exponential growth of publicly available datasets, a scholarly recommendation system of datasets would be an essential tool in the field of information filtering. Recommending datasets to users can be formulated as a classification problem where deep learning models can be carefully trained. In such a case, when preparing training data for the learning models, one needs to consider different ratios of false and true pairs. Therefore, a sensitivity analysis is necessary. In this work, we conduct a sensitivity analysis using different class ratios on a deep learning model (BERT) for recommending datasets. We found out that our BERT-based recommender model is relatively robust using recommender metrics such as Mean Reciprocal Rank (MRR)@k, Recall@k, etc., except for the extreme class imbalance case (1:5000). Therefore, we conclude that a moderate ratio of the random negative sampling scheme, (in our case 1:10) is reasonable, sufficient and time-efficient in the recommendation system traininghttps://journals.flvc.org/FLAIRS/article/view/130595recommender systembertsensitivity analysis
spellingShingle Jie Zhu
Hulin Wu
Ashraf Yaseen
Sensitivity Analysis of a BERT-based scholarly recommendation system
Proceedings of the International Florida Artificial Intelligence Research Society Conference
recommender system
bert
sensitivity analysis
title Sensitivity Analysis of a BERT-based scholarly recommendation system
title_full Sensitivity Analysis of a BERT-based scholarly recommendation system
title_fullStr Sensitivity Analysis of a BERT-based scholarly recommendation system
title_full_unstemmed Sensitivity Analysis of a BERT-based scholarly recommendation system
title_short Sensitivity Analysis of a BERT-based scholarly recommendation system
title_sort sensitivity analysis of a bert based scholarly recommendation system
topic recommender system
bert
sensitivity analysis
url https://journals.flvc.org/FLAIRS/article/view/130595
work_keys_str_mv AT jiezhu sensitivityanalysisofabertbasedscholarlyrecommendationsystem
AT hulinwu sensitivityanalysisofabertbasedscholarlyrecommendationsystem
AT ashrafyaseen sensitivityanalysisofabertbasedscholarlyrecommendationsystem