Multi-task Learning for Features Extraction in Financial Annual Reports

被引:0
作者
Montariol, Syrielle [1 ]
Martinc, Matej [1 ]
Pelicon, Andraz [1 ]
Pollak, Senja [1 ]
Koloski, Boshko [1 ]
Loncarski, Igor [2 ]
Valentincic, Aljosa [2 ]
Sustar, Katarina Sitar [2 ]
Ichev, Riste [2 ]
Znidarsic, Martin [1 ]
机构
[1] Jozef Stefan Inst, Jamova Cesta 39, Ljubljana 1000, Slovenia
[2] Univ Ljubljana, Sch Econ & Business, Kardeljeva Pl 17, Ljubljana 1000, Slovenia
来源
MACHINE LEARNING AND PRINCIPLES AND PRACTICE OF KNOWLEDGE DISCOVERY IN DATABASES, ECML PKDD 2022, PT II | 2023年 / 1753卷
关键词
Multi-task learning; Financial reports; Corporate social responsibility; INFORMATION; DISCLOSURE;
D O I
10.1007/978-3-031-23633-4_1
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
For assessing various performance indicators of companies, the focus is shifting from strictly financial (quantitative) publicly disclosed information to qualitative (textual) information. This textual data can provide valuable weak signals, for example through stylistic features, which can complement the quantitative data on financial performance or on Environmental, Social and Governance (ESG) criteria. In this work, we use various multi-task learning methods for financial text classification with the focus on financial sentiment, objectivity, forward-looking sentence prediction and ESG-content detection. We propose different methods to combine the information extracted from training jointly on different tasks; our best-performing method highlights the positive effect of explicitly adding auxiliary task predictions as features for the final target task during the multi-task training. Next, we use these classifiers to extract textual features from annual reports of FTSE350 companies and investigate the link between ESG quantitative scores and these features.
引用
收藏
页码:7 / 24
页数:18
相关论文
共 50 条
[41]   LEARNING DEEP FEATURES FOR MULTIPLE OBJECT TRACKING BY USING A MULTI-TASK LEARNING STRATEGY [J].
Wang, Li ;
Nam Trung Pham ;
Ng, Tian-Tsong ;
Wang, Gang ;
Chan, Kap Luk ;
Leman, Karianto .
2014 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2014, :838-842
[42]   Multi-Task Multi-Sample Learning [J].
Aytar, Yusuf ;
Zisserman, Andrew .
COMPUTER VISION - ECCV 2014 WORKSHOPS, PT III, 2015, 8927 :78-91
[43]   Learning Task Relational Structure for Multi-Task Feature Learning [J].
Wang, De ;
Nie, Feiping ;
Huang, Heng .
2016 IEEE 16TH INTERNATIONAL CONFERENCE ON DATA MINING (ICDM), 2016, :1239-1244
[44]   Learning Tree Structure in Multi-Task Learning [J].
Han, Lei ;
Zhang, Yu .
KDD'15: PROCEEDINGS OF THE 21ST ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, 2015, :397-406
[45]   Learning to Resolve Conflicts in Multi-Task Learning [J].
Tang, Min ;
Jin, Zhe ;
Zou, Lixin ;
Liang Shiuan-Ni .
ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING, ICANN 2023, PT III, 2023, 14256 :477-489
[46]   Multi-task Learning with Modular Reinforcement Learning [J].
Xue, Jianyong ;
Alexandre, Frederic .
FROM ANIMALS TO ANIMATS 16, 2022, 13499 :127-138
[47]   Feature Extraction of Laser Machining Data by Using Deep Multi-Task Learning [J].
Zhang, Quexuan ;
Wang, Zexuan ;
Wang, Bin ;
Ohsawa, Yukio ;
Hayashi, Teruaki .
INFORMATION, 2020, 11 (08)
[48]   Multi-task Learning for Attribute Extraction from Unstructured Electronic Medical Records [J].
Du, Ming ;
Pang, Minmin ;
Xu, Bo .
SEMANTIC TECHNOLOGY, JIST 2019, 2020, 1157 :117-128
[49]   Integration of Relation Filtering and Multi-Task Learning in GlobalPointer for Entity and Relation Extraction [J].
Liu, Bin ;
Tao, Jialin ;
Chen, Wanyuan ;
Zhang, Yijie ;
Chen, Min ;
He, Lei ;
Tang, Dan .
APPLIED SCIENCES-BASEL, 2024, 14 (15)
[50]   Multi-task Learning for Extraction of Adverse Drug Reaction Mentions from Tweets [J].
Gupta, Shashank ;
Gupta, Manish ;
Varma, Vasudeva ;
Pawar, Sachin ;
Ramrakhiyani, Nitin ;
Palshikar, Girish Keshav .
ADVANCES IN INFORMATION RETRIEVAL (ECIR 2018), 2018, 10772 :59-71