[1]程鹏超,杜军平,薛哲.基于多路交叉的用户金融行为预测[J].智能系统学报,2021,16(2):378-384.[doi:10.11992/tis.202006054]
 CHENG Pengchao,DU Junping,XUE Zhe.Prediction of user financial behavior based on multi-way crossing[J].CAAI Transactions on Intelligent Systems,2021,16(2):378-384.[doi:10.11992/tis.202006054]
点击复制

基于多路交叉的用户金融行为预测(/HTML)
分享到:

《智能系统学报》[ISSN:1673-4785/CN:23-1538/TP]

卷:
第16卷
期数:
2021年2期
页码:
378-384
栏目:
吴文俊人工智能科学技术奖论坛
出版日期:
2021-03-05

文章信息/Info

Title:
Prediction of user financial behavior based on multi-way crossing
作者:
程鹏超 杜军平 薛哲
北京邮电大学 智能通信软件与多媒体北京市重点实验室,北京 100876
Author(s):
CHENG Pengchao DU Junping XUE Zhe
Beijing Key Laboratory of Intelligent Telecommunication Software and Multimedia, School of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100876, China
关键词:
行为预测金融多路交叉残差多塔模型预训练挖掘联合训练
Keywords:
behavior predictionfinancialmulti-way crossingresidualmulti-tower modelpre-trainingminingjoint training
分类号:
TP391
DOI:
10.11992/tis.202006054
摘要:
针对通过挖掘用户的金融行为来改善金融领域的服务模式和服务质量的问题,本文提出了一种基于多路交叉特征的用户金融行为预测算法。根据数据包含的属性构建训练的特征,基于因子分解机模型(FM)利用下游行为预测任务对金融数据的特征进行预训练,获取数据特征的隐含向量。引入特征交叉层对金融数据的高阶特征进行提取,解决FM线性模型只能提取低阶特征的缺点。利用残差网络对金融数据的高阶特征进行提取,解决深度神经网络在提取金融数据高阶特征时由于网络层数过深而导致的梯度消失的问题。最后,将FM、特征交叉网络和残差网络整合为统一的多塔模型进行用户金融行为预测,并融合低阶特征与高阶特征进行用户金融行为预测。在多个数据集上对算法的有效性进行了实验验证,实验结果表明,所提出的算法能够取得较好的用户金融行为预测的准确率。
Abstract:
To improve the service mode and service quality in the financial field by mining the financial behaviors of users, a user financial behavior prediction algorithm based on multi-way crossing (MCUP) is proposed in this paper. First, the training features are constructed based on the attributes contained in the data. Based on the FM model, the downstream behavior prediction tasks are used to pre-train the features of the financial data, and the hidden vectors of the features are obtained. Second, the feature cross-layer is introduced to extract high-order features of financial data, overcoming the disadvantage that the FM linear model can only extract low-order features. Then, the residual network structure is used to extract high-order features of financial data, solving the gradient disappearance problem caused by the too deep network layer. Finally, a unified multi-tower model integrated by the FM, feature cross network, and residual network is used to predict user financial behavior, blending low-order and high-order features. Experimental results show that the proposed algorithm can achieve a better accuracy rate in predicting user financial behavior.

参考文献/References:

[1] MAIA M, HANDSCHUH S, FREITAS A,et al. WWW’18 open challenge:financial opinion mining and question answering[C]//Companion Proceedings of the Web Conference 2018. Lyon, France:International World Wide Web Conferences Steering Committee, 2018:1941-1942.
[2] SHI Hongyu, CHEN Ling, XU Zhenxing, et al. Personalized location recommendation using mobile phone usage information[J]. Applied intelligence, 2019, 49(10):3694-3707.
[3] CHEN Jiehao, LI Xueyi, ZHAO Ziqian, et a. A CTR prediction method based on feature engineering and online learning[C]//Proceedings of the 17th International Symposium on Communications and Information Technologies. Cairns, Australia, 2017:1-6.
[4] ZHAO Xibin, WANG Nan, ZHANG Yubo, et al. Beyond pairwise matching:person reidentification via high-order relevance learning[J]. IEEE transactions on neural networks and learning systems, 2018, 29(8):3701-3714.
[5] HAO Tong, WANG Qian, WU Dan, et al. Adaptive recommendation for photo pose via deep learning[J]. Multimedia tools and applications, 2018, 77(17):22173-22184.
[6] RYU G A, NASRIDINOV A, RAH H, et al. Forecasts of the amount purchase pork meat by using structured and unstructured big data[J]. Agriculture, 2020, 10(1):1-14.
[7] 曾宪宇, 刘淇, 赵洪科, 等. 用户在线购买预测:一种基于用户操作序列和选择模型的方法[J]. 计算机研究与发展, 2016, 53(8):1673-1683
ZENG Xianyu, LIU Qi, ZHAO Hongke, et al. Online consumptions prediction via modeling user behaviors and choices[J]. Journal of computer research and development, 2016, 53(8):1673-1683
[8] LIU Xiaoman, LI Jing. Using support vector machine for online purchase predication[C]//Proceedings of 2016 International Conference on Logistics, Informatics and Service Sciences (LISS). Sydney, Australia, 2016:1-6.
[9] 祝歆, 刘潇蔓, 陈树广, 等. 基于机器学习融合算法的网络购买行为预测研究[J]. 统计与信息论坛, 2017, 32(12):94-100
ZHU Xin, LIU Xiaoman, CHEN Shuguang, et al. Research on network purchase behavior prediction based on machine learning fusion algorithm[J]. Statistics and information forum, 2017, 32(12):94-100
[10] 杜刚, 黄震宇. 大数据环境下客户购买行为预测[J]. 管理现代化, 2015, 35(1):40-42
DU Gang, HUANG Zhenyu. Customer purchasing behavior prediction under the environment of big data[J]. Modernization of management, 2015, 35(1):40-42
[11] 李美其, 齐佳音. 基于购买行为及评论行为的用户购买预测研究[J]. 北京邮电大学学报(社会科学版), 2016, 18(4):18-25
LI Meiqi, QI Jiayin. Customer purchase prediction based on buying behavior and comment behavior[J]. Journal of Beijing University of Posts and Telecommunications (social sciences edition), 2016, 18(4):18-25
[12] 张春莲. 客户购买行为的BG/NBD预测模型及其应用研究[D]. 哈尔滨:哈尔滨工业大学, 2006.
ZHANG Chunlian. Research on BG/NBD prediction model and its application of the customer purchase behavior[D]. Harbin:Harbin Institute of Technology, 2006.
[13] 刘柏鳞. 基于客户网络购物行为分析及预测系统研究[D]. 北京:北京工业大学, 2016.
LIU Bailin. The research on the analysis of customer’s online shopping behavior and prediction system[D]. Beijing:Beijing University of Technology, 2016.
[14] 陈洁, 谢文昕, 杨升荣. 在线渠道消费者动态品牌选择购买率预测[J]. 工业工程与管理, 2011, 16(3):52-56
CHEN Jie, XIE Wenxin, YANG Shengrong. Forecasting the pachase rate of online consumer’ s dynamic brand choice[J]. Industrial engineering and management, 2011, 16(3):52-56
[15] 舒方, 马少辉. 客户重复购买的组合预测方法[J]. 计算机与现代化, 2015(5):67-70
SHU Fang, MA Shaohui. A composition forecasting approach of customer repeat purchasing[J]. Computer and modernization, 2015(5):67-70
[16] 胡晓丽, 张会兵, 董俊超, 等. 基于CNN-LSTM的用户购买行为预测模型[J]. 计算机应用与软件, 2020, 37(6):59-64
HU Xiaoli, ZHANG Huibing, DONG Junchao, et al. Prediction model of user buying behavior based on CNN-LSTM[J]. Computer applications and software, 2020, 37(6):59-64
[17] 卢泓宇, 张敏, 刘奕群, 等. 卷积神经网络特征重要性分析及增强特征选择模型[J]. 软件学报, 2017, 28(11):2879-2890
LU Hongyu, ZHANG Min, LIU Yiqun, et a. Convolution neural network feature importance analysis and feature selection enhanced model[J]. Journal of software, 2017, 28(11):2879-2890
[18] ZHANG Weinan, DU Tianming, WANG Jun. Deep learning over multi-field categorical data[C]//Proceedings of 38th European Conference on Information Retrieval. Padua, Italy, 2016:45-57.
[19] GUO Huifeng, TANG Ruiming, YE Yunming, et al. DeepFM:a factorization-machine based neural network for CTR prediction[C]//The International Joint Conference on Artificial Intelligence(IJCAI). Melbourne, Australia, 2017:1725-1731.
[20] WANG Ruoxi, FU Bin, FU Gang, et al. Deep and cross network for Ad click predictions[C]//Proceedings of the ADKDD’17. NS, Canada, 2017:1-7.

相似文献/References:

[1]孔庆超,毛文吉,张育浩.社交网站中用户评论行为预测[J].智能系统学报,2015,10(03):349.[doi:10.3969/j.issn.1673-4785.201403019]
 KONG Qingchao,MAO Wenji,ZHANG Yuhao.User comment behavior prediction in social networking sites[J].CAAI Transactions on Intelligent Systems,2015,10(2):349.[doi:10.3969/j.issn.1673-4785.201403019]
[2]姬晓飞,谢旋,任艳.深度学习的双人交互行为识别与预测算法研究[J].智能系统学报,2020,15(3):484.[doi:10.11992/tis.201812029]
 JI Xiaofei,XIE Xuan,REN Yan.Human interaction recognition and prediction algorithm based on deep learning[J].CAAI Transactions on Intelligent Systems,2020,15(2):484.[doi:10.11992/tis.201812029]

备注/Memo

备注/Memo:
收稿日期:2020-06-30。
基金项目:国家重点研发计划项目(2018YFB1402600);国家自然科学基金项目(61772083,61802028);广西省科技重大专项(桂科AA18118054)
作者简介:程鹏超,硕士研究生,主要研究方向为机器学习、广告推荐、信息检索;杜军平,教授,博士生导师,主要研究方向为人工智能、社交网络分析、数据挖掘、运动图像处理。主持国家重点研发计划、“863”、“973”计划项目、国家自然科学基金重点项目、国家自然科学基金重大国际合作项目、北京市自然科学基金重点项目等多项。发表学术论文400余篇,出版学术专著6部;薛哲,副教授,主要研究方向为机器学习、人工智能、数据挖掘、图像处理。主持国家自然科学基金青年基金项目、参与国家重点研发计划项目等多项。发表学术论文30余篇,出版学术专著1部
通讯作者:杜军平.E-mail:junpingdu@126.com
更新日期/Last Update: 2021-04-25