计算机科学 ›› 2019, Vol. 46 ›› Issue (10): 84-89.doi: 10.11896/jsjkx.180901771

• 大数据与数据科学* • 上一篇    下一篇

基于深度双向LSTM的股票推荐系统

曾安1, 聂文俊2   

  1. (广东工业大学计算机学院 广州510006)1
    (广东省大数据分析与处理重点实验室 广州510006)2
  • 收稿日期:2018-09-19 修回日期:2019-02-21 出版日期:2019-10-15 发布日期:2019-10-21
  • 通讯作者: 曾安(1978-),女,博士,CCF会员,主要研究方向为数据挖掘, E-mail:437871335@qq.com。
  • 作者简介:聂文俊(1992-),男,硕士,主要研究方向为数据挖掘。
  • 基金资助:
    本文受国家自然科学基金项目(61772143,61300107),广东省自然科学基金项目(S2012010010212),广州市科技计划项目(201601010034,201505031501397),广东省大数据分析与处理重点实验室开放基金项目(201801)资助。

Stock Recommendation System Based on Deep Bidirectional LSTM

ZENG An1, NIE Wen-jun2   

  1. (School of Computer,Guangdong University of Technology,Guangzhou 510006,China)1
    (Guangdong Key Laboratory of Big Data Analysis and Processing,Guangzhou 510006,China)2
  • Received:2018-09-19 Revised:2019-02-21 Online:2019-10-15 Published:2019-10-21

摘要: 面对越来越复杂的数据环境,以经典统计学模型为主的股票预测模型在一定程度上已无法满足人们对预测准确性的要求。深度学习因具有较强的学习能力和抗干扰能力,已逐渐被应用于股票推荐中。但传统的股票推荐模型要么从未考虑时间因素,要么仅考虑时间上的单向关系。因此,文中提出了一种基于深度双向LSTM的神经网络预测模型。该模型充分利用了时间序列上向前、向后两个时间方向的上下文关系,解决了长时间序列上的梯度消失和梯度爆炸问题,能够学习到对时间有长期依赖性的信息。同时,该模型引入了Dropout策略,在一定程度上解决了深层网络模型带来的训练难、收敛速度慢和过拟合等问题。在S&P500数据集上的实验表明,基于深度双向LSTM的神经网络预测模型比现有预测模型在误差上降低了2%~5%,使决定系数(r2)提高了10%。

关键词: 股票预测, 深度RNN, 双向LSTM, 推荐系统

Abstract: With the diversity of applications scenarios and rapid growth of data,the stock prediction models based on classical statistical methods are unable to meet the requirements for high prediction accuracy.But traditional stock reco-mmendation models either never consider the time factor or just consider the unidirectional relationship over time.However,existing stock recommendation models based on deep learning rarely consider the time factor.This paper proposed a deep bidirectional LSTM model for stock prediction,which makes full use of context relationship in the forward direction and backward direction of time series.The problem of vanishing gradient and exploding gradient are solved by introducing LSTM when dealing with long-term sequence.The proposed model can learn information which has long-term dependence on time.At the same time,dropout strategy is introduced to prevent over-fitting caused by deep network model and speed up the training.Experiments on S&P500 dataset show that the neural network prediction model based on the deep bidirectional LSTM outperforms the existing prediction models,the error is about 5% lower,and the coefficient of determination (r2) is increased by 10%.

Key words: Bidirectional long short-term memory, Deep recurrent neural networks, Recommendation system, Stock forecast

中图分类号: 

  • TP181
[1]DEGIANNAKIS S,FILIS G,HASSANI H.Forecasting global stock market implied volatility indices[J].Journal of Empirical Finance,2018,46:111-129.
[2]ZHANG W Y,ZHANG S X,ZHANG S,et al.A multi-factor and high-order stock forecast model based on Type-2 FTS using cuckoo search and self-adaptive harmony search[J].Neurocomputing,2017,240:13-24.
[3]NARENDRA BABU C,REDDY B E.Prediction of selected Indian stock using a partitioning-interpolation based ARIMA-GARCH model[J].Applied Computing and Informatics,2015,11(2):130-143.
[4]ROSAS-ROMERO R,DÍAZ-TORRES A,ETCHEVERRY G. Forecasting of stock return prices with sparse representation of financial time series over redundant dictionaries[J].Expert Systems With Applications,2016,57:37-48.
[5]CHEN Y T,TAO L,LI H B.The Prediction of Dow Jones Cbn China 600 Inde Based on “Most Updated” GM(1,1) Model and Grey Dynamic Neural Network[C]//2010 Third International Conference on Information and Computing.Wuxi:IEEE,2010:56-59.
[6]HUANG K Y,JANE C J.A hybrid model for stock market forecasting and portfolio selection based on ARX,grey system and RS theories[J].Expert Systems With Applications,2008,36(3):5387-5392.
[7]WANG D X,LIU X,WANG M D.A DT-SVM Strategy for Stock Futures Prediction with Big Data[C]//Computational Science and Engineering.Sydney,NSW,Australia:IEEE,2013:1005-1012.
[8]CHEN Y J,HAO Y T.A feature weighted support vector machine and K-nearest neighbor algorithm for stock market indices prediction[J].Expert Systems with Applications,2017,80:340-355.
[9]WANG P G.Research on Stock Price Prediction Based on BP Wavelet Neural Network with Mexico Hat Wavelet Basis[C]//Proceedings of 1st International Conference on Education,Economics and Management Research.Advances in Social Science,Education and Humanities Research (ICEEMR 2017) .Singapore Management University,2017,95:4.
[10]NAYAK S C,MISRA B B,BEHERA H S.Artificial chemical reaction optimization of neural networks for efficient prediction of stock market indices[J].Ain Shams Engineering Journal,2015,8(3):371-390.
[11]CHATZIS S P,SIAKOULIS V,PETROPOULOS A,et al. Forecasting stock market crisis events using deep and statistical machine learning techniques[J].Expert Systems with Applications,2018,112:353-371.
[12]GILES C L,KUHN G M,WILLIAMS R J.Dynamic recurrent neural networks:Theory and applications [J].IEEE Trans.Neural Networks,1994,15:153-156.
[13]SCHUSTER M,PALIWAL K K.Bidirectional Recurrent Neural Networks [J].IEEE Transactions on Signal Processing,1997,45:11.
[14]HOCHREITER S,SCHMIDHUBER J.Long Short-term Me-mory [J].Neural computation,1997,9(8):1735-1780.
[15]GRAVES A,SCHMIDHUBER J.Framewise Phoneme Classification with Bidirectional LSTM and Other Neural Network Architectures [J].Neural Networks,2005,18(5):602-610.
[16]SUN R Q.Research on the price trend forecasting model of us stock index based on LSTM neural network[D].Capital University of Economids and Business,2015.
[17]FISCHER T,KRAUSS C.Deep learning with long?short-term memory networks for financial market predictions[J].European Journal of Operational Research,2018,270(2):654-669.
[18]SRIVASTAVA N,HINTON G,KRIZHEVSKY A.Dropout:A Simple Way to Prevent Neural Networks from Overtting [J].Journal of Machine Learning Research,2014(15):1929-1958.
[19]HOCHREITER S,SCHMIDHUBER J.Long short-term memory[J].Neural computation,1997,9(8).
[20]HIRANSHA M,GOPALAKRISHNAN E A,VIJAY KRISHNA Menon,et al.NSE Stock Market Prediction Using Deep-Learning Models[J].Procedia Computer Science,2018,132:1351-1362.
[21]YETIS Y,KAPLAN H,JAMSHIDI M.Stock market prediction by using artificial neural network[C]//World Automation Congress (WAC) .Waikoloa,HI,USA:IEEE,2014:718-722.
[22]REN J,WANG J H,WANG C M,et al.Stock Index Forecast Based on Regularized LSTM Model[J].Computer Application and Software,2018,35(4):44-48,108.
[1] 程章桃, 钟婷, 张晟铭, 周帆.
基于图学习的推荐系统研究综述
Survey of Recommender Systems Based on Graph Learning
计算机科学, 2022, 49(9): 1-13. https://doi.org/10.11896/jsjkx.210900072
[2] 王冠宇, 钟婷, 冯宇, 周帆.
基于矢量量化编码的协同过滤推荐方法
Collaborative Filtering Recommendation Method Based on Vector Quantization Coding
计算机科学, 2022, 49(9): 48-54. https://doi.org/10.11896/jsjkx.210700109
[3] 秦琪琦, 张月琴, 王润泽, 张泽华.
基于知识图谱的层次粒化推荐方法
Hierarchical Granulation Recommendation Method Based on Knowledge Graph
计算机科学, 2022, 49(8): 64-69. https://doi.org/10.11896/jsjkx.210600111
[4] 方义秋, 张震坤, 葛君伟.
基于自注意力机制和迁移学习的跨领域推荐算法
Cross-domain Recommendation Algorithm Based on Self-attention Mechanism and Transfer Learning
计算机科学, 2022, 49(8): 70-77. https://doi.org/10.11896/jsjkx.210600011
[5] 帅剑波, 王金策, 黄飞虎, 彭舰.
基于神经架构搜索的点击率预测模型
Click-Through Rate Prediction Model Based on Neural Architecture Search
计算机科学, 2022, 49(7): 10-17. https://doi.org/10.11896/jsjkx.210600009
[6] 齐秀秀, 王佳昊, 李文雄, 周帆.
基于概率元学习的矩阵补全预测融合算法
Fusion Algorithm for Matrix Completion Prediction Based on Probabilistic Meta-learning
计算机科学, 2022, 49(7): 18-24. https://doi.org/10.11896/jsjkx.210600126
[7] 蔡晓娟, 谭文安.
一种改进的融合相似度和信任度的协同过滤算法
Improved Collaborative Filtering Algorithm Combining Similarity and Trust
计算机科学, 2022, 49(6A): 238-241. https://doi.org/10.11896/jsjkx.210400088
[8] 何亦琛, 毛宜军, 谢贤芬, 古万荣.
基于点割集图分割的矩阵变换与分解的推荐算法
Matrix Transformation and Factorization Based on Graph Partitioning by Vertex Separator for Recommendation
计算机科学, 2022, 49(6A): 272-279. https://doi.org/10.11896/jsjkx.210600159
[9] 郭亮, 杨兴耀, 于炯, 韩晨, 黄仲浩.
基于注意力机制和门控网络相结合的混合推荐系统
Hybrid Recommender System Based on Attention Mechanisms and Gating Network
计算机科学, 2022, 49(6): 158-164. https://doi.org/10.11896/jsjkx.210500013
[10] 熊中敏, 舒贵文, 郭怀宇.
融合用户偏好的图神经网络推荐模型
Graph Neural Network Recommendation Model Integrating User Preferences
计算机科学, 2022, 49(6): 165-171. https://doi.org/10.11896/jsjkx.210400276
[11] 洪志理, 赖俊, 曹雷, 陈希亮, 徐志雄.
基于遗憾探索的竞争网络强化学习智能推荐方法研究
Study on Intelligent Recommendation Method of Dueling Network Reinforcement Learning Based on Regret Exploration
计算机科学, 2022, 49(6): 149-157. https://doi.org/10.11896/jsjkx.210600226
[12] 余皑欣, 冯秀芳, 孙静宇.
结合物品相似性的社交信任推荐算法
Social Trust Recommendation Algorithm Combining Item Similarity
计算机科学, 2022, 49(5): 144-151. https://doi.org/10.11896/jsjkx.210300217
[13] 陈壮, 邹海涛, 郑尚, 于化龙, 高尚.
基于用户覆盖及评分差异的多样性推荐算法
Diversity Recommendation Algorithm Based on User Coverage and Rating Differences
计算机科学, 2022, 49(5): 159-164. https://doi.org/10.11896/jsjkx.210300263
[14] 陈晋鹏, 胡哈蕾, 张帆, 曹源, 孙鹏飞.
融合时间特性和用户偏好的卷积序列化推荐
Convolutional Sequential Recommendation with Temporal Feature and User Preference
计算机科学, 2022, 49(1): 115-120. https://doi.org/10.11896/jsjkx.201200192
[15] 詹皖江, 洪植林, 方路平, 吴哲夫, 吕跃华.
基于对抗性学习的协同过滤推荐算法
Collaborative Filtering Recommendation Algorithm Based on Adversarial Learning
计算机科学, 2021, 48(7): 172-177. https://doi.org/10.11896/jsjkx.200600077
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!