Purpose Online consumer health forums offer an alternative source of health-related information for internet users seeking specific details that may not be readily available through articles or other one-way communication channels. However, the effectiveness of these forums can be constrained by the limited number of healthcare professionals actively participating, which can impact response times to user inquiries. One potential solution to this issue is the integration of a semi-automatic system. A critical component of such a system is question processing, which often involves sentence recognition (SR), medical entity recognition (MER), and keyphrase extraction (KE) modules. We posit that the development of these three modules would enable the system to identify critical components of the question, thereby facilitating a deeper understanding of the question, and allowing for the re-formulation of more effective questions with extracted key information. Methods This work contributes to two key aspects related to these three tasks. First, we expand and publicly release an Indonesian dataset for each task. Second, we establish a baseline for all three tasks within the Indonesian language domain by employing transformer-based models with nine distinct encoder variations. Our feature studies revealed an interdependence among these three tasks. Consequently, we propose several multi-task learning (MTL) models, both in pairwise and three-way configurations, incorporating parallel and hierarchical architectures. Results Using F1-score at the chunk level, the inter-annotator agreements for SR, MER, and KE tasks were <InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="13326_2025_329_Article_IEq1.gif" Format="GIF" Height="18" Rendition="HTML" Resolution="72" Type="Linedraw" Width="108" /> </InlineMediaObject> <EquationSource Format="TEX">$88{.}61\%, 64{.}83\%$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mn>88.61</mn> <mo>%</mo> <mo>,</mo> <mn>64.83</mn> <mo>%</mo> </mrow> </math></EquationSource> </InlineEquation>, and <InlineEquation ID="IEq2"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="13326_2025_329_Article_IEq2.gif" Format="GIF" Height="16" Rendition="HTML" Resolution="72" Type="Linedraw" Width="51" /> </InlineMediaObject> <EquationSource Format="TEX">$35{.}01\%$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mn>35.01</mn> <mo>%</mo> </mrow> </math></EquationSource> </InlineEquation> respectively. In single-task learning (STL) settings, the best performance for each task was achieved by different model, with <InlineEquation ID="IEq3"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="13326_2025_329_Article_IEq3.gif" Format="GIF" Height="16" Rendition="HTML" Resolution="72" Type="Linedraw" Width="110" /> </InlineMediaObject> <EquationSource Format="TEX">$\textrm{IndoNLU}_{\textrm{LARGE}}$</EquationSource> <EquationSource Format="MATHML"><math> <msub> <mtext>IndoNLU</mtext> <mtext>LARGE</mtext> </msub> </math></EquationSource> </InlineEquation> obtained the highest average score. These results suggested that a larger model did not always perform better. We also found no indication of which ones between Indonesian and multilingual language models that generally performed better for our tasks. In pairwise MTL settings, we found that pairing tasks could outperform the STL baseline for all three tasks. Despite varying loss weights across our three-way MTL models, we did not identify a consistent pattern. While some configurations improved MER and KE performance, none surpassed the best pairwise MTL model for the SR task. Conclusion We extended an Indonesian dataset for SR, MER, and KE tasks, resulted in 1, 173 labeled data points which splitted into 773 training instances, 200 validation instances, and 200 testing instances. We then used transformer-based models to set a baseline for all three tasks. Our MTL experiments suggested that additional information regarding the other two tasks could help the learning process for MER and KE tasks, while had only a small effect for SR task.

Sentences, entities, and keyphrases extraction from consumer health forums using multi-task learning

期刊论文
出版模式：

开放获取
发表日期：

2025年5月6日

查看全文

Tsaqif Naufal,
Rahmad Mahendra,
Alfan Farizki Wicaksono

摘要

Purpose

Online consumer health forums offer an alternative source of health-related information for internet users seeking specific details that may not be readily available through articles or other one-way communication channels. However, the effectiveness of these forums can be constrained by the limited number of healthcare professionals actively participating, which can impact response times to user inquiries. One potential solution to this issue is the integration of a semi-automatic system. A critical component of such a system is question processing, which often involves sentence recognition (SR), medical entity recognition (MER), and keyphrase extraction (KE) modules. We posit that the development of these three modules would enable the system to identify critical components of the question, thereby facilitating a deeper understanding of the question, and allowing for the re-formulation of more effective questions with extracted key information.

Methods

This work contributes to two key aspects related to these three tasks. First, we expand and publicly release an Indonesian dataset for each task. Second, we establish a baseline for all three tasks within the Indonesian language domain by employing transformer-based models with nine distinct encoder variations. Our feature studies revealed an interdependence among these three tasks. Consequently, we propose several multi-task learning (MTL) models, both in pairwise and three-way configurations, incorporating parallel and hierarchical architectures.

Results

Using F1-score at the chunk level, the inter-annotator agreements for SR, MER, and KE tasks were $88{.}61\%, 64{.}83\%$ $88.61 %, 64.83 %$ , and $35{.}01\%$ $35.01 %$ respectively. In single-task learning (STL) settings, the best performance for each task was achieved by different model, with $\textrm{IndoNLU}_{\textrm{LARGE}}$ ${IndoNLU}_{LARGE}$ obtained the highest average score. These results suggested that a larger model did not always perform better. We also found no indication of which ones between Indonesian and multilingual language models that generally performed better for our tasks. In pairwise MTL settings, we found that pairing tasks could outperform the STL baseline for all three tasks. Despite varying loss weights across our three-way MTL models, we did not identify a consistent pattern. While some configurations improved MER and KE performance, none surpassed the best pairwise MTL model for the SR task.

Conclusion

We extended an Indonesian dataset for SR, MER, and KE tasks, resulted in 1, 173 labeled data points which splitted into 773 training instances, 200 validation instances, and 200 testing instances. We then used transformer-based models to set a baseline for all three tasks. Our MTL experiments suggested that additional information regarding the other two tasks could help the learning process for MER and KE tasks, while had only a small effect for SR task.