共查询到19条相似文献,搜索用时 203 毫秒
1.
2.
流量建模与预测对于大规模网络的规划设计和网络资源管理等方面都具有积极的意义,是网络流量工程重要组成部分。该文结合网络流量的时间序列特性,提出一种基于支持向量机的网络流量预测算法,并针对由于支持向量机采用先验知识选择参数会导致不同数据对先验知识适应程度不同,给出了一个动态调整优化参数策略。实验仿真结果表明,该算法具有很好的预测精度和适用性。 相似文献
3.
4.
基于支持向量回归机的股票价格预测 总被引:1,自引:0,他引:1
研究股票价格预测问题,股票价格变化具有非线性、时变性,传统线性预测模型难以准确刻画股价变化规律,且非线性神经网络存在过拟合、局部最小等缺陷,预测精度比较低。为提高股票价格预测精度,提出一种基于粒子群优化支持向量机的股票价格预测模型。利用粒子群算法良好的寻优能力,对支持向量机参数进行优化,加快支持向量机学习速度,再采用非线性预测能力优异的支持向量机对股票价格进行预测。以南天信息股票价格对模型性能进行仿真,实验结果证明,支持向量机预测模型能全面反映股票价格变化的非线性的时变规律,获得更高预测精度,预测结果对股民实际操作具有较大的指导价值。 相似文献
5.
针对希尔伯特-黄变换中的边界效应,提出了基于支持向量回归机的时间序列预测方法.在支持向量回归机的应用当中,参数的选取对它的泛化性能有很大影响.在讨论了参数对支持向量回归机的泛化性能的影响基础上,提出了通过微粒群优化算法来优化支持向量回归机参数的方法,使得支持向量回归机在应用中能够自适应的选择最优参数,从而获得了更好的泛化性能,提高了在端点处的延拓精度,很好地抑制了端点效应.试验表明,该优化算法能够很好解决支持向量回归机的参数选取问题.通过与神经网络的延拓方法和黄等人的HHTDPS结果对比,基于支持向量回归机的时间序列预测方法可以更好地解决在希尔伯特-黄变换中存在的边界效应,得到的固有模态函数具有较小的失真. 相似文献
6.
基于粗集支持向量机的区域经济走势预测方法 总被引:4,自引:4,他引:0
针对区域经济预测中多属性而支持向量机方法无法有效选择的问题,提出了一个基于粗集理论和支持向量机的区域经济走势预测方法.方法利用粗集理论在处理多属性数据方面的优势对区域经济走势预测的条件属性进行约简,约简后的数据进入支持向量机的预测系统,从而减少了支持向量机的训练数据,在一定程度上克服了支持向量机方法处理速度慢的缺点.将方法应用于某区域经济走势预测中,获得了较好的预测结果.实证结果表明,方法具有较好的预测能力,与标准支持向量机方法相比,方法具有明显的优势. 相似文献
7.
提出基于小波变换和支持向量机的水质预测模型。该模型运用小波变换得到水质时间序列在不同尺度下的变化特性,并用改进后的粒子群算法优化回归支持向量机的三个参数,提高了模型预测精度。运用该模型对王江泾自动监测站测得的溶解氧浓度进行了1步预测及2步预测,10组测试样本最高MAPE为4.54%,并用基于BP神经网络的预测结果进行了比较。结果表明,该模型性能良好、预测精度高、简便易行,比基于BP神经网络的模型具有更好的预测效果,为水质预测提供了一种有效的方法。 相似文献
8.
支持向量机(SVM)是近年来发展起来的一种通用的机器学习方法,在小样本数据的拟合中已获得了很好的效果。采用新型的支持向量机——最小二乘支持向量机(LS-SVM)对孔隙度、渗透率和饱和度进行了预测,获得了满意的结果。该方法易于使用,很少受不确定性因素的影响,并具有较强的信息整合能力以及更高的预测准确性。 相似文献
9.
10.
基于小波变换的动态关联规则元规则GM(1,1)挖掘 总被引:1,自引:1,他引:0
提出了一种把小波变换应用到动态关联规则元规则挖掘中并提高规则预测精度的方法。该方法首先利用小波变换技术对挖掘出的动态关联规则元规则支持度计数进行变换,然后通过小波变换的多分辨率特点提取出近似部分和细节部分,并利用这两部分别进行单支重构,随后利用GM(1,1)对重构的两部分进行预测,从而得到最后的预测结果,最后通过实验证明了该方法具有较高的预测精度。 相似文献
11.
12.
针对短期风电功率预测,提出一种基于稀疏表示特征提取的建模方法.为了构建预测模型,将历史风电功率数据构成具有时延的输入-输出数据对,将时延输入数据向量作为初始字典,由K-均值奇异值分解(K-SVD)算法将其进行稀疏分解与变换至稀疏域以得到学习后的字典,由正交匹配追踪(OMP)算法获取相应的稀疏编码向量,再将该向量作为极限学习机(ELM)或支持向量机(SVM)的输入来构建全局回归模型.为了验证所提出的方法的有效性,将所提出的方法用于短期风电功率预测中,在同等条件下与单一SVM、ELM方法和非字典学习的其他稀疏表示建模方法进行了比较.实验结果表明,不同的稀疏表示建模方法均能取得很好的预测效果,其中所提出的方法具有更好的预测效果,显示出其有效性. 相似文献
13.
This paper proposes a modified version of support vector machines (SVMs), called ε-descending support vector machines (ε-DSVMs),
to model non-stationary financial time series. The ε-DSVMs are obtained by incorporating the problem domain knowledge – non-stationarity
of financial time series into SVMs. Unlike the standard SVMs which use a constant tube in all the training data points, the
ε-DSVMs use an adaptive tube to deal with the structure changes in the data. The experiment shows that the ε-DSVMs generalize
better than the standard SVMs in forecasting non-stationary financial time series. Another advantage of this modification
is that the ε-DSVMs converge to fewer support vectors, resulting in a sparser representation of the solution.
This revised version was published online in June 2006 with corrections to the Cover Date. 相似文献
14.
R. Vinu Chandran A. Jeyaram V. Jayaraman S. Manoj K. Rajitha C. K. Mukherjee 《International journal of remote sensing》2013,34(17):4479-4491
Forecast of potential fishing grounds can save considerable amounts of fuel and time involved in marine fishing operations, making the operation more profitable. This study investigates the prospects of using a decision support system for forecasting potential fishery zones and prioritizing them according to profitability using multi-criteria analysis. Ocean Colour Monitor-derived chlorophyll concentration and National Oceanic and Atmospheric Administration Advanced Very High Resolution Radiometer-derived sea surface temperature maps were integrated to demarcate the potential fishery zones (PFZs). QuickSCAT (NASA) scatterometer wind data obtained from the global 25 km × 25 km gridded dataset were used for updating the locations of PFZ features. An analytical hierarchy process-based prioritization model, which considers different parameters favouring a targeted species, distance to the zone and historical catch per unit effort as the major decision-making parameters, was used to rank the zones based on feasibility. A software package was developed incorporating the above-mentioned functionalities for automatic data processing and PFZ map generation. 相似文献
15.
《Expert systems with applications》2005,28(1):119-126
Aquaculture plays a very important role in adjusting Agricultural structure and increasing farmers' income. But the technological advances in the production and storage of fishery products have exceeded the development of effective market demand over the past one decade. As a result, participants within the fishery industry have frequently found themselves facing increased variable and declining prices negatively affected the fishery industry. So it is desirable to provide decision aids to price forecasting to avoid such market imbalance. An aquatic products price forecasting support system (APPFSS) was developed by China agricultural University. Based on questionnaire and interviews, we analyze the decision problems and user needs. Then the architecture and development of APPFSS were described. At last we discussed on problems we had encountered during development and promotion. 相似文献
16.
Minyoung Kim 《Applied Intelligence》2018,48(11):3889-3901
Sparse coding, often called dictionary learning, has received significant attention in the fields of statistical machine learning and signal processing. However, most approaches assume iid data setup, which can be easily violated when the data retains certain statistical structures such as sequences where data samples are temporally correlated. In this paper we formulate a novel dynamic sparse coding problem, and propose an efficient algorithm that enforces smooth dynamics for the latent state vectors (codes) within a linear dynamic model while imposing sparseness of the state vectors. We overcome the added computational overhead originating from smooth dynamic constraints by adopting the recent first-order smooth optimization technique, adjusted for our problem instance. We demonstrate the improved prediction performance of our approach over the conventional sparse coding on several interesting real-world problems including financial asset return data forecasting and human motion estimation from silhouette videos. 相似文献
17.
基于支持向量机的股市预测 总被引:2,自引:1,他引:2
针对股票市场高燥声、强非线性和不确定性等特点和以往传统神经网络预测方法存在的不足,提出了一种基于支持向量机的股市预测方法。该方法主要运用了支持向量机回归的方法结合滚动时间窗来学习建摸。首先通过把低维输入空间的输入向量映射到高维特征空间,将非线性问题转化为线性,然后在结构风险最小化原则下进行二次规划,并求得最优解,从而建立模型。从仿真实验中可以看到,该方法建立的模型较为准确地预测了600009、000815两只股票的日均价,表现出了较强的泛化能力。 相似文献
18.