基于改进DeepFM的心脏病预测应用研究
Applied Research on Heart Disease Prediction Based on Improved DeepFM

DOI: 10.12677/CSA.2021.118217, PP. 2117-2125

Keywords: 心脏病，因子分解机，前馈神经网络，集成树
Heart Disease, Factorization Machine, Feedforward Neural Network, Integrated Tree

Full-Text   Cite this paper   Add to My Lib

Abstract:

近年来，心脏病在全球已严重威胁到人类的身体和生命健康安全，通过利用人工智能等技术手段来辅助医疗诊断的科学技术日益普遍，为提高心脏病诊断的准确性，本文提出了一种在DeepFM模型的基础上改进后的较为新颖的模型——RDF模型。RDF模型由三个组件共同构成，其中因子分解机对低阶特征交互进行建模，BP神经网络对高阶特征交互进行建模，集成树则进一步提高模型的准确性和稳健性。本文在UCI数据集中的303个心脏病样本上进行实验，实验结果显示AUC值为0.8809，准确率为0.8317。
In recent years, heart disease has been a serious threat to human life and health safety, and the technology of medical diagnosis assisted by artificial intelligence is becoming more and more common. In order to improve the accuracy of heart disease diagnosis, based on DeepFM model, this paper proposes a novel model—RDF model. The RDF model is composed of three components: Factor Machine is used to model the low-order feature interaction, the BP neural network is used to model the high-order feature interaction, and the integration tree is used to further enhance the accuracy and robustness of the model. The experiment was performed on 303 heart disease samples from the UCI datasets. Experimental results show that the AUC value is 0.8809 and the accuracy is 0.8317.

References

[1]	马丽媛, 吴亚哲, 陈伟伟. 《中国心血管病报告2018》要点介绍[J]. 中华高血压杂志, 2019, 27(8): 712-716.
[2]	秦文哲, 陈进, 董力. 大数据背景下医学数据挖掘的研究进展及应用[J]. 中国胸心血管外科临床杂志, 2016, 23(1): 55-60.
[3]	？omak, E., Arslan, A. and Turkoglu, ？. (2007) A Decision Support System Based on Support Vec-tor Machines for Diagnosis of the Heart Valve Diseases. Computers in Biology and Medicine, 37, 21-27. https://doi.org/10.1016/j.compbiomed.2005.11.002
[4]	王阶, 李军, 姚魁武, 衷敬柏. 冠心病心绞痛证候要素和冠脉病变的Logistic回归分析[J]. 辽宁中医杂志, 2007(9): 1209-1211.
[5]	陈天华, 郑彧, 韩力群, 唐海滔. 基于神经网络的冠心病无创诊断方法研究[J]. 航天医学与医学工程, 2008, 21(6): 513-517.
[6]	王莉莉, 付忠良, 陶攀, 胡鑫. 基于主动学习不平衡多分类AdaBoost算法的心脏病分类[J]. 计算机应用, 2017, 37(7): 1994-1998.
[7]	Guo, R., Wang, Y.Q., Yan, H.X., et al. (2015) Analysis and Recognition of Traditional Chinese Med-icine Pulse Based on the Hilbert-Huang Transform and Random Forest in Patients with Coronary Heart Disease. Evi-dence-Based Complementary and Alternative Medicine, 2015, Article ID: 895749. https://doi.org/10.1155/2015/895749
[8]	逄凯. 三种机器学习方法在冠心病筛查中的比较研究[D]: [硕士学位论文]. 长春: 吉林大学, 2016.
[9]	Yekkala, I. and Dixit, S. (2018) Prediction of Heart Disease Using Random Forest and Rough Set Based Feature Selection. International Journal of Big Data and Analytics in Healthcare, 3, 1-12. https://doi.org/10.4018/IJBDAH.2018010101
[10]	赵金超, 李仪, 王冬, 张俊虎. 基于优化的随机森林心脏病预测算法[J]. 青岛科技大学学报(自然科学版), 2021, 42(2): 112-118.
[11]	Madhumita, P. and Smita, P. (2021) Pre-diction of Heart Diseases Using Random Forest. Journal of Physics: Conference Series, 1817, Article ID: 012009. https://doi.org/10.1088/1742-6596/1817/1/012009
[12]	Guo, H.F., Tang, R.M., Ye, Y.M., et al. (2017) DeepFM: A Factorization-Machine Based Neural Network for CTR Prediction.
[13]	陈一文. 一种改进的基于DeepFM算法的高效CTR预估方法[D]: [硕士学位论文]. 长春: 吉林大学, 2020.
[14]	Cheng, H.-T., Koc, L., Harmsen, J., et al. (2016) Wide & Deep Learning for Recommender Systems. Proceedings of the 1st Workshop on Deep Learning for Recommender Systems, Boston, 15 September 2016, 7-10. https://doi.org/10.1145/2988450.2988454
[15]	Rendle, S. (2010) Factorization Machines. IEEE International Con-ference on Data Mining, Sydney, 13-17 December 2010, 995-1000. https://doi.org/10.1109/ICDM.2010.127
[16]	周志华. 机器学习[M]. 北京: 清华大学出版社, 2016: 180.

Full-Text