税收优惠政策关键要素抽取与可视化分析

doi:10.11959/j.issn.2096-0271.2022035

Abstract

Abstract:

With the rapid increase in the number of preferential tax policies, taxpayers face a large number of preferential tax policies, and it is difficult for taxpayers to quickly locate the preferential content related to them.As a result, many taxpayers do not enjoy the preferential policies they should enjoy.Based on the combination of pre-training language model BERT and rule processing, the visualization was realized of the characterization of preferential tax policies and regulations, the extraction of key elements, and the visual query of tax incentives, so that taxpayers can intuitively and quickly locate tax incentives related to themselves, and visualize the results.The experimental results show that the extraction performance of key elements is superior, and the query of preferential tax policies is quick and intuitive, which can effectively alleviate the problem of massive tax preferential information overload.

Key words: preferential tax policy, pre-trained language model, information extraction, visualization

CLC Number:

TP391.1

Haishan GUAN, Yulong ZHENG, Bifan WEI, Zemin ZHANG, Hao YUE, Bin SHI, Bo DONG. Extraction and visualization analysis of key elements of tax preferential policies[J]. Big Data Research, 2022, 8(5): 106-123.

Figures/Tables 14

References 19

[1]	汪疆平, 肖戎 . 税务大数据分析的技术和典型应用[J]. 大数据, 2017,3(2): 92-103.
	WANG J P , XIAO R . Big data analysis technology and application on taxation[J]. Big Data Research, 2017,3(2): 92-103.
[2]	邵凌云 . 基于纳税人需求优化纳税服务机制[J]. 税务研究, 2013(5): 76-79.
	SHAO L Y . Optimize the tax service mechanism based on the demand of taxpayers[J]. Taxation Research, 2013(5): 76-79.
[3]	谢学刚, 苟仁金 . 提升纳税服务质量的现实选择[J]. 税务研究, 2014(11): 96.
	XIE X G , GOU R J . A realistic choice to improve the quality of tax services[J]. Taxation Research, 2014(11): 96.
[4]	谢波峰 . 基于大数据的税收经济分析和预测探索[J]. 大数据, 2017,3(3): 15-24.
	XIE B F . Exploratory research on big data application of analysis and forecasting in economics of tax[J]. Big Data Research, 2017,3(3): 15-24.
[5]	ZHANG R X , YANG W , LIN L Y ,et al. Rapid adaptation of bert for information extraction on domain-specific business documents[J]. arXiv preprint,2020,arXiv:2002.01861.
[6]	NGUYEN M T , LE D T , LINH L T ,et al. AURORA:an information extraction system of domainspecific business documents with limited data[C]// Proceedings of the 29th ACM International Conference on Information ＆ Knowledge Management. New York:ACM Press, 2020: 3437-3440.
[7]	FRIEDRICH A , ADEL H , TOMAZIC F ,et al. The SOFC-exp corpus and neural approaches to information extraction in the materials science domain[C]// Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. Stroudsburg:Association for Computational Linguistics, 2020: 1255-1268.
[8]	ZEGHDAOUI M W , BOUSSAID O , BENTAYEB F ,et al. Medical-based text classification using FastText features and CNN-LSTM model[C]// Database and Expert Systems Applications. Cham:Springer, 2021: 155-167.
[9]	BRANDES U , KENIS P , WAGNER D . Communicating centrality in policy network drawings[J]. IEEE Transactions on Visualization and Computer Graphics, 2003,9(2): 241-253.
[10]	BRANDES U , PICH C . More flexible radial layout[J]. Journal of Graph Algorithms and Applications, 2011,15(1): 157-173.
[11]	RAJ M , WHITAKER R T . Anisotropic radial layout for visualizing centrality and structure in graphs[C]// Graph Drawing and Network Visualization.[S.l.:s.n.], 2018: 351-364.
[12]	FENU G , SPANO L D . Recommendation Centre:inspecting and controlling recommendations with radial layouts[C]// Workshop on Engineering Computer-Human Interaction in Recommender Systems.[S.l.:s.n.], 2016: 54-61.
[13]	BOSTOCK M , HEER J . Protovis:a graphical toolkit for visualization[J]. IEEE Transactions on Visualization and Computer Graphics, 2009,15(6): 1121-1128.
[14]	BOSTOCK M , OGIEVETSKY V , HEER J . D3 data-driven documents[J]. IEEE Transactions on Visualization and Computer Graphics, 2011,17(12): 2301-2309.
[15]	LI D Q , MEI H H , SHEN Y ,et al. ECharts:a declarative framework for rapid construction of web-based visualization[J]. Visual Informatics, 2018,2(2): 136-146.
[16]	DEVLIN J , CHANG M W , LEE K ,et al. Bert:pre-training of deep bidirectional transformers for language understanding[J]. Proceedings of NAACLHLT, 2019: 4171-4186.
[17]	MCCALLUM A , LI W . Early results for named entity recognition with conditional random fields,feature induction and web-enhanced lexicons[C]// Proceedings of the 7th Conference on Natural Language Learning at HLTNAACL 2003. Morristown:Association for Computational Linguistics, 2003: 188-191.
[18]	HUANG Z H , XU W , YU K . Bidirectional LSTM-CRF models for sequence tagging[J]. arXiv preprint,2015,arXiv:1508.01991.
[19]	QIU X P , SUN T X , XU Y G ,et al. Pretrained models for natural language processing:a survey[J]. Science China Technological Sciences, 2020,63(10): 1872-1897.

Metrics

Recommended 0

No Suggested Reading articles found!

关键要素	描述	任务处理方式
条款内容	即税收优惠主题，指包含税收优惠信息的条款内容	文本分类
享受主体	描述享受优惠的主体对象	序列标注
税种	优惠涉及的税种	文本多标签分类
减免类型	优惠的减免类型	文本分类
减免方式	优惠的方式	文本分类
政策类型	该优惠条款属于何种政策类型	文本分类
有效期限	优惠条款的时间期限	基于规则
标题	优惠条款所属公告标题	基于规则
文号	优惠条款所属公告文号	基于规则

字段名称	举例（正则）
有效期限	[0-9]{1,4}年[0-9]{1,2}月[0-9]{1,2}日(?=起执行)、(?<=执行时间为)[0-9]{1,4}年[0-9]{1,2}月[0-9]{1,2}日至[0-9]{1,4}年[0-9]{1,2}月[0-9]{1,2}日……
标题	(?<=\s)[\u4e00-\u9fa5]+[0-9]*[\u4e00-\u9fa5]+(?<=通知\|公告)、(?<=\s)[\u4e00-\u9fa5]+(?<=通知\|公告)……
文号	(?<=通知\|公告)[\u4e00-\u9fa5]{1,10}〔[0-9]{4,}〕[0-9]+号、[\u4e00-\u9fa5]{1,10}〔[0-9]{4,}〕[09]+号……

关键要素	标题	文号	有效期限
precision	0.939（avg）	0.896（avg）	0.933（avg）

关键要素	BERT			Fasttext			TextCNN
关键要素	F1分数	precision	recall	F1分数	precision	recall	F1分数	precision	recall
条款内容	0.914	0.921	0.898	0.505	0.923	0.347	0.825	0.83	0.825
减免类型/方式	0.97	0.99	0.96	0.857	0.897	0.82	0.663	0.964	0.611
政策类型	0.831 (Micro)	0.83	0.847	0.864 (Micro)	0.592	0.646	0.675 (Micro)	0.658	0.711

关键要素	模型直接处理的Hamming Loss	分步处理后的Hamming Loss
税种	0.023	0.016

Extraction and visualization analysis of key elements of tax preferential policies

RichHTML

PDF下载

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 14

References 19

Related Articles 10

Metrics

Recommended 0

关键要素	BERT+BiLSTM+CRF（本文方法）			BERT+Linear+CRF			BERT+BiGRU+CRF
关键要素	F1分数	precision	recall	F1分数	precision	recall	F1分数	precision	recall
享受主体	0.73	0.71	0.75	0.69	0.67	0.71	0.71	0.68	0.75

[1]	Yadong WU, Jiaming CHEN, Yan LUO, Xuefeng WANG, Dechun HUANG, Chao NI, Jiming LAN, Suiqun LI, Weihan ZHANG, Wei DAI. An overview of Caideng metaverse research [J]. Big Data Research, 2023, 9(3): 97-113.
[2]	Lifan HAN, Zijing JI, Zirui CHEN, Xin WANG. Research on information extraction methods for historical classics under the threshold of digital humanities [J]. Big Data Research, 2022, 8(6): 26-39.
[3]	Yuchu LUO, Hao WU, Yuhan GUO, Shaocong TAN, Can LIU, Ruike JIANG, Xiaoru YUAN. Visualization in digital humanities [J]. Big Data Research, 2022, 8(6): 74-93.
[4]	Liang FAN, Xiaorong ZHANG, Yadong WU, Cheng CHEN, Fang WANG. A sort-last architecture based parallel volume visualization algorithm for unstructured grid [J]. Big Data Research, 2021, 7(4): 117-129.
[5]	Jingxi WEN, Hufei YU, Jiang XIN, Yan TANG. Analysis of gender differences in the brain based on deep learning [J]. Big Data Research, 2021, 7(4): 130-140.
[6]	Chao CHEN, Yadong WU, Chaoshuai FU, Xing TONG, Pan LI, Qikai CHU, Xuenan WANG. Research on visualization of Chinese Baijiu culture [J]. Big Data Research, 2021, 7(2): 78-98.
[7]	Yang WANG, Ye TIAN, Tieshan LI, HENC.L.Philip C, Dongcheng PENG, Yihua ZHOU. Visualization analysis of shipping recruitment information based on Gephi [J]. Big Data Research, 2018, 4(3): 81-91.
[8]	Yilei LU. Practices of hybrid heterogeneous marketing data platform [J]. Big Data Research, 2018, 4(3): 104-110.
[9]	Jianghua ZHAO, Xuezhi WANG, Qinghui LIN, Jianhui LI, Yuanchun ZHOU. Exploration of crowdsourcing in information extraction from remote sensing images [J]. Big Data Research, 2016, 2(6): 53-64.
[10]	Zhiyuan CHENG, Yubin BAO, Fangling LENG. Parallel graph layout algorithm for large-scale graph data [J]. Big Data Research, 2016, 2(5): 12-21.