正則化訓(xùn)練的神經(jīng)網(wǎng)絡(luò)與粗集理論相結(jié)合的股票時間序列數(shù)據(jù)挖掘技術(shù)
Stock Market Time Series Data Mining Based on Regularized Neural Network and Rough Set
-
摘要: 論文提出將正則化神經(jīng)網(wǎng)絡(luò)與粗集理論相結(jié)合應(yīng)用于股票時間序列數(shù)據(jù)庫的數(shù)據(jù)挖掘.首先對時間序列數(shù)據(jù)庫進行預(yù)處理,除去高頻干擾信號,然后將股票時間序列數(shù)據(jù)按照收盤價的變化趨勢分割成一系列靜態(tài)模式,每種模式代表股票價格的一種行為趨勢(上漲或下跌),把決定各種模式的相關(guān)屬性組成一系列信息,形成一個適用于粗集方法的信息表.然后使用正則神經(jīng)網(wǎng)絡(luò)對信息表進行學(xué)習(xí),用粗集理論從正則神經(jīng)網(wǎng)絡(luò)所存儲的知識中抽取規(guī)則,得到的規(guī)則可以用于預(yù)測時間序列在未來的行為。該方法融合了正則神經(jīng)網(wǎng)絡(luò)優(yōu)良的泛化性能和粗集理論的規(guī)則生成能力,實驗表明,該方法預(yù)測效果比較準(zhǔn)確。Abstract: This paper presents a new method of stock market time series data mining. It combines regularized neural network with rough set. The process includes preprocessing of time series and data mining. The preprocessing cleans and filters time series. Then, the time series are partitionel into a series of static patterns, which is based on the trend (i.e., increasing or decreasing) of closing price. The most important predicting attributes identified from every model form an information table. The regularized neural network is used to learn and predict the data. Rough set can extract rule knowledge in the neural network, which can be used to predict the time series behavior in the future. This method combines the generalization faculty of regularized neural network and the rule reduction capability of rough set. The experimental results demonstrate the effectiveness of the algorithm.
-
Das G, Gunopulos D. Finding similar time series. In Proc. of the Conference on Principles of Knowledge Discovery and Data Mining, Trondheim, Norway, 1997: 124-135.[2]Das G, Lin K, Mannila H, Renganathan G, Smyth P. Rule discovery from time series. In Proc. of the 4th Int. Conference on Knowledge Discovery and Data Mining, New York, NY, Aug. 27-31,1998: 179-183.[3]Hansen V J, Nelson R D. Data mining of time series using stacked generalizers[J].Neurocomputing.2002, 43(1):173-184[4]Last M, Klein Y. Knowledge discovery in time series databases[J].IEEE Trans. on System, Man and Cybernetics-part B.2001, 31(1):160-169[5]Qin Zh, Mao Z. A new algorithm for neural network architecture study. In Proc. of the 3rd World Congress on Intelligent Control and Automation, Hefei, China. June 2000: 795-799.[6]Lingras P. Comparison of ne0fuzzy and rough neural networks. Information Sciences, 1998, (110):207-215.[7]Girosi F, Jones M. Regularization theory and neural networks architectures[J].Neural Computation.1995, 7(2):219-269[8]王國胤.Rough集理論與知識獲取.西安:西安交通大學(xué)出版社,2001:99-104.[9]Jiawei H, Micheline K. Data Mining: Concepts and Techniques. San Meteo, CA: Morgan Kaufmann Publishers, Inc., 2001, Chapter 5.[10]Reed R. Pruning algorithms-a survey[J].IEEE Trans. on Neural Networks.1993, 4(5):740-747[11]李冬梅,王正歐.提高前向神經(jīng)網(wǎng)絡(luò)泛化性能和實時性能的新方法.電機與控制學(xué)報,2002,6(3):241-244.[12]常犁云,王國胤等.一種基于Rough Set理論的屬性約簡及規(guī)則提取方法.軟件學(xué)報,1999,10(11):1206-1211.[13]宋擒豹,沈鈞毅.神經(jīng)網(wǎng)絡(luò)數(shù)據(jù)挖掘方法中的數(shù)據(jù)準(zhǔn)備問題.計算機工程與應(yīng)用,2000,36(12):102-104. -
計量
- 文章訪問數(shù): 2952
- HTML全文瀏覽量: 134
- PDF下載量: 695
- 被引次數(shù): 0