Exploring Bill Similarity with Attention Mechanism for Enhanced Legislative Prediction

Common legislative prediction methods often emphasize bill content or social relationships. This paper, motivated by the insight that similar policy texts reflect comparable political ideologies and can lead to similar voting outcomes, proposes a deep learning method that exploits attention mechanis...

Full description

Saved in:
Bibliographic Details
Main Authors: Bofeng Wang, Yanyan Li, Cong Xu
Format: Article
Language:English
Published: Tsinghua University Press 2025-06-01
Series:Journal of Social Computing
Subjects:
Online Access:https://www.sciopen.com/article/10.23919/JSC.2025.0005
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849426186582622208
author Bofeng Wang
Yanyan Li
Cong Xu
author_facet Bofeng Wang
Yanyan Li
Cong Xu
author_sort Bofeng Wang
collection DOAJ
description Common legislative prediction methods often emphasize bill content or social relationships. This paper, motivated by the insight that similar policy texts reflect comparable political ideologies and can lead to similar voting outcomes, proposes a deep learning method that exploits attention mechanisms to incorporate semantic similarity between bills into legislative prediction models. Our approach uses attention scores to identify bills that are most similar to the one being predicted, and combines the encoded features of these similar bills as additional auxiliary information. By integrating these related features, the model goes beyond the semantic information of individual bills, leading to a more comprehensive use of roll-call data. Empirical results show that utilizing bill similarity along with traditional social relationships, voter characteristics, and bill content significantly improves performance in terms of accuracy, recall, precision, and F1 score compared to models that ignore bill similarity. The results also confirm that legislators tend to maintain consistent views or voting patterns on bills that are similar in nature. In addition, we demonstrate that the attention mechanism is more effective than conventional similarity measures, such as cosine similarity and Euclidean distance, in capturing the similarities between bills.
format Article
id doaj-art-bb958e5aab7d42f7b61bf64d0f3d0485
institution Kabale University
issn 2688-5255
language English
publishDate 2025-06-01
publisher Tsinghua University Press
record_format Article
series Journal of Social Computing
spelling doaj-art-bb958e5aab7d42f7b61bf64d0f3d04852025-08-20T03:29:31ZengTsinghua University PressJournal of Social Computing2688-52552025-06-016211212510.23919/JSC.2025.0005Exploring Bill Similarity with Attention Mechanism for Enhanced Legislative PredictionBofeng Wang0Yanyan Li1Cong Xu2School of Mathematics and Information Sciences, Yantai University, Yantai 264005, ChinaSchool of Literature and Journalism, Yantai University, Yantai 264005, ChinaSchool of Computer Science and Technology, East China Normal University, Shanghai 200062, ChinaCommon legislative prediction methods often emphasize bill content or social relationships. This paper, motivated by the insight that similar policy texts reflect comparable political ideologies and can lead to similar voting outcomes, proposes a deep learning method that exploits attention mechanisms to incorporate semantic similarity between bills into legislative prediction models. Our approach uses attention scores to identify bills that are most similar to the one being predicted, and combines the encoded features of these similar bills as additional auxiliary information. By integrating these related features, the model goes beyond the semantic information of individual bills, leading to a more comprehensive use of roll-call data. Empirical results show that utilizing bill similarity along with traditional social relationships, voter characteristics, and bill content significantly improves performance in terms of accuracy, recall, precision, and F1 score compared to models that ignore bill similarity. The results also confirm that legislators tend to maintain consistent views or voting patterns on bills that are similar in nature. In addition, we demonstrate that the attention mechanism is more effective than conventional similarity measures, such as cosine similarity and Euclidean distance, in capturing the similarities between bills.https://www.sciopen.com/article/10.23919/JSC.2025.0005legislative predictionbill similarityattention mechanismroll call datadeep learning
spellingShingle Bofeng Wang
Yanyan Li
Cong Xu
Exploring Bill Similarity with Attention Mechanism for Enhanced Legislative Prediction
Journal of Social Computing
legislative prediction
bill similarity
attention mechanism
roll call data
deep learning
title Exploring Bill Similarity with Attention Mechanism for Enhanced Legislative Prediction
title_full Exploring Bill Similarity with Attention Mechanism for Enhanced Legislative Prediction
title_fullStr Exploring Bill Similarity with Attention Mechanism for Enhanced Legislative Prediction
title_full_unstemmed Exploring Bill Similarity with Attention Mechanism for Enhanced Legislative Prediction
title_short Exploring Bill Similarity with Attention Mechanism for Enhanced Legislative Prediction
title_sort exploring bill similarity with attention mechanism for enhanced legislative prediction
topic legislative prediction
bill similarity
attention mechanism
roll call data
deep learning
url https://www.sciopen.com/article/10.23919/JSC.2025.0005
work_keys_str_mv AT bofengwang exploringbillsimilaritywithattentionmechanismforenhancedlegislativeprediction
AT yanyanli exploringbillsimilaritywithattentionmechanismforenhancedlegislativeprediction
AT congxu exploringbillsimilaritywithattentionmechanismforenhancedlegislativeprediction