【摘要】：單位的運(yùn)營(yíng)狀況會(huì)直接影響股東和廣大人民的利益,針對(duì)運(yùn)營(yíng)狀況可以使用廣義回歸神經(jīng)網(wǎng)絡(luò)進(jìn)行分類。由于廣義回歸神經(jīng)網(wǎng)絡(luò)中徑向基函數(shù)的擴(kuò)展參數(shù)Spread的選取會(huì)導(dǎo)致分類的準(zhǔn)確率,提出了一種果蠅優(yōu)化算法優(yōu)化參數(shù)Spread的分類模型。充分利用了果蠅優(yōu)化算法的尋優(yōu)能力,將優(yōu)化后的參數(shù)代入到廣義回歸神經(jīng)網(wǎng)絡(luò)中對(duì)單位的財(cái)務(wù)數(shù)據(jù)進(jìn)行運(yùn)營(yíng)狀況的分類。結(jié)果表明,與廣義回歸神經(jīng)網(wǎng)絡(luò)做比較,優(yōu)化后的網(wǎng)絡(luò)模型對(duì)數(shù)據(jù)的分類可以達(dá)到很高的準(zhǔn)確率,在相關(guān)領(lǐng)域的分類上有非常大的實(shí)用性。
[Abstract]:The operating condition of the unit will directly affect the interests of shareholders and the broad masses of the people, and the general regression neural network can be used to classify the operating situation. Because the selection of the extended parameter Spread of the radial basis function in the generalized regression neural network will lead to the accuracy of classification, a classification model of the optimization parameter Spread of Drosophila melanogaster optimization algorithm is proposed. Making full use of the optimization ability of Drosophila melanogaster optimization algorithm, the optimized parameters are substituted into the generalized regression neural network to classify the financial data of the unit. The results show that compared with the generalized regression neural network, the optimized network model can achieve high accuracy in the classification of data, and has great practicability in the classification of related fields.
【作者單位】： 中北大學(xué)理學(xué)院;
【基金】：國(guó)家自然科學(xué)基金資助項(xiàng)目(61275120)
【分類號(hào)】：TP183
，

本文編號(hào)：2483708