基于BP神經(jīng)網(wǎng)絡(luò)的卷煙銷售違規(guī)預(yù)測研究
[Abstract]:The tobacco industry is a high-profit industry, which is one of the main functions of the tobacco monopoly bureau. Because of the development of information technology, it is a problem to be discussed carefully at this stage. In this paper, the BP neural network technology is applied to the tobacco monopoly inspection, and it is of great theoretical and practical significance to explore the prediction model of the cigarette sales violation. Meaning. Through the qualitative analysis of the sales behavior and sales psychology of the cigarette retail, find out the related factors that affect the behavior of the retail account In ord to collect relevant data for a tobacco company in a city, pre-process that data, carry out quantitative analysis on a large amount of historical sales record and relevant data, and establish a cigarette sales violation prediction mode. the method comprises the following steps of: calculating and analyzing a cigarette sales violation prediction model by using a BP neural network algorithm, and finally finding a prediction model with the lowest judgment rate through a plurality of training and changing parameters, The prediction results show that the model forecasting accuracy is high, and it has good universality. It can be used to forecast the violation of each retail account effectively. The method not only can improve the accuracy of the cigarette making, but also can improve the working efficiency of the cigarette department,
【學(xué)位授予單位】:合肥工業(yè)大學(xué)
【學(xué)位級別】:碩士
【學(xué)位授予年份】:2013
【分類號】:TP183;F426.89
【參考文獻】
相關(guān)期刊論文 前10條
1 李響;;基于BP神經(jīng)網(wǎng)絡(luò)的股價預(yù)測[J];大連海事大學(xué)學(xué)報;2008年S1期
2 司繼文;韓瑩瑩;羅希;;Hedonic住宅特征價格模型的BP神經(jīng)網(wǎng)絡(luò)方法[J];管理學(xué)報;2012年07期
3 南鳳蘭;;證券公司風險預(yù)測研究[J];金融研究;2007年11期
4 劉洪,何光軍;基于人工神經(jīng)網(wǎng)絡(luò)方法的上市公司經(jīng)營失敗預(yù)警研究[J];會計研究;2004年02期
5 史玉峰,曹俊如;基于遺傳BP神經(jīng)網(wǎng)絡(luò)的變形數(shù)據(jù)分析處理[J];礦業(yè)研究與開發(fā);2004年04期
6 岑成德,權(quán)凈;服務(wù)屬性對顧客滿意感影響程度研究——人工神經(jīng)網(wǎng)絡(luò)方法[J];南開管理評論;2005年02期
7 劉永建;朱劍英;曾捷;;改進BP神經(jīng)網(wǎng)絡(luò)在發(fā)動機性能趨勢分析和故障診斷中的應(yīng)用[J];南京理工大學(xué)學(xué)報(自然科學(xué)版);2010年01期
8 卜耀華;;基于數(shù)據(jù)挖掘技術(shù)的企業(yè)財務(wù)困境預(yù)測建模[J];計算機仿真;2012年06期
9 石曉軍;;Logistic違約率模型最優(yōu)樣本配比與分界點的模擬分析[J];數(shù)理統(tǒng)計與管理;2006年06期
10 李清;于萍;;財務(wù)危機預(yù)測主要方法比較研究[J];數(shù)理統(tǒng)計與管理;2012年04期
相關(guān)博士學(xué)位論文 前1條
1 姜明輝;商業(yè)銀行個人信用評估組合預(yù)測方法研究[D];哈爾濱工業(yè)大學(xué);2006年
相關(guān)碩士學(xué)位論文 前1條
1 陳鴻;人工神經(jīng)網(wǎng)絡(luò)在宏觀經(jīng)濟預(yù)測中的應(yīng)用研究[D];重慶大學(xué);2005年
本文編號:2509016
本文鏈接:http://www.sikaile.net/guanlilunwen/shengchanguanlilunwen/2509016.html