【摘要】：隨著社會的發(fā)展和人們生活水平的不斷提高,生鮮農(nóng)產(chǎn)品作為餐桌上不可或缺的食物和營養(yǎng)來源,人們對它的需求也與日俱增。生鮮連鎖超市作為一種新的生鮮農(nóng)產(chǎn)品經(jīng)營方式,由于其統(tǒng)一經(jīng)營管理,生鮮農(nóng)產(chǎn)品的安全和質(zhì)量更有保障,因而越來越受到消費者的歡迎與認(rèn)可。然而隨著生鮮超市業(yè)務(wù)的不斷發(fā)展和擴(kuò)張,生鮮超市的業(yè)務(wù)信息系統(tǒng)卻發(fā)展緩慢。企業(yè)現(xiàn)有的超市業(yè)務(wù)系統(tǒng)往往沒有考慮到生鮮農(nóng)產(chǎn)品的獨特性和經(jīng)營中的特殊需求,缺少對整個生鮮業(yè)務(wù)的全過程覆蓋,同時并沒對生鮮超市連鎖企業(yè)在經(jīng)營過程中產(chǎn)生的大量數(shù)據(jù)進(jìn)行數(shù)據(jù)分析和挖掘。本文針對長春市某生鮮連鎖超市企業(yè)的實際情況,設(shè)計和開發(fā)了更加貼合企業(yè)需求的生鮮超市業(yè)務(wù)系統(tǒng),并將數(shù)據(jù)挖掘中的聚類分析和關(guān)聯(lián)規(guī)則運用于企業(yè)的銷售數(shù)據(jù)分析中。將數(shù)據(jù)挖掘中的相關(guān)技術(shù)與業(yè)務(wù)系統(tǒng)相結(jié)合,從而幫助企業(yè)制定更加合理的經(jīng)營管理和營銷策略。主要工作分為兩部分:第一部分為生鮮超市業(yè)務(wù)系統(tǒng)基礎(chǔ)功能的實現(xiàn)工作,系統(tǒng)采用當(dāng)前較為流行的Java語言,開源的SSH框架來完成系統(tǒng)中各個模塊的開發(fā)。系統(tǒng)的基礎(chǔ)業(yè)務(wù)模塊主要包括用戶管理、接受銷售流水、驗收入庫、大庫分貨、門店調(diào)撥、商品報損、采購計劃、門店訂貨、商品管理、在途商品信息管理、退貨處理等,讓生鮮超市經(jīng)營中的每一環(huán)節(jié)都可控可追蹤,完成數(shù)據(jù)共享和整個業(yè)務(wù)的信息化,提高生鮮流轉(zhuǎn)過程中信息化水平。在途商品信息管理中使用開源的Open Layers構(gòu)建了基于JavaScript的輕量級Web GIS模塊,能夠更加直觀的獲取運輸?shù)乩砦恢眯畔�。此外在考慮到采購人員對移動辦公的迫切需求后,完成了采購管理移動App的設(shè)計與開發(fā),幫助采購員更好的開展采購工作;第二部分工作為設(shè)計和實現(xiàn)對生鮮連鎖超市的銷售數(shù)據(jù)分析,由于生鮮超市在經(jīng)營過程中會積累大量數(shù)據(jù)以及每個門店的銷售數(shù)據(jù)不同,因此可以使用各個門店的銷售數(shù)據(jù)來對門店進(jìn)行聚類分析。但是在對聚類分析中的模糊C均值聚類算法進(jìn)行研究后,發(fā)現(xiàn)該算法在聚類前需要指定聚類數(shù)目的缺點后引入了統(tǒng)計學(xué)中的混合F分布,提出了一種能獲取最佳聚類數(shù)目的改進(jìn)方案,能夠解決模糊C均值聚類算法需要事先指定聚類數(shù)目的缺點,并將改進(jìn)后的算法應(yīng)用到生鮮超市銷售特征的分析中。然后針對生鮮超市顧客購買商品之間的潛在關(guān)聯(lián),設(shè)計和實現(xiàn)了顧客購物籃分析。生鮮連鎖超市顧客每次購買商品時的交易數(shù)據(jù)可以稱為購物籃數(shù)據(jù),由于數(shù)據(jù)挖掘技術(shù)中的關(guān)聯(lián)規(guī)則較為適用于生鮮超市顧客購物籃交易數(shù)據(jù)的分析中,因此采用Apriori關(guān)聯(lián)規(guī)則挖掘算法來對顧客的購物籃數(shù)據(jù)進(jìn)行商品間關(guān)聯(lián)規(guī)則的挖掘。通過兩個模塊中的實驗結(jié)果進(jìn)行分析,能夠發(fā)現(xiàn)企業(yè)積累的大量數(shù)據(jù)中潛在的信息,為企業(yè)的實際經(jīng)營管理提供幫助。
[Abstract]:With the development of society and the improvement of people's living standards, fresh agricultural products as an indispensable source of food and nutrition on the table, the demand for it is also increasing. As a new mode of management of fresh agricultural products, fresh supermarket chain is more and more popular and accepted by consumers because of its unified management, and the safety and quality of fresh agricultural products are more and more guaranteed. However, with the development and expansion of fresh supermarket business, the business information system of fresh supermarket is developing slowly. The existing supermarket business systems of enterprises often do not take into account the uniqueness of fresh agricultural products and the special needs in operation, and lack the whole process of covering the whole fresh products business. At the same time, there is no data analysis and mining on a large number of data generated in the operation process of fresh supermarket chain enterprises. In view of the actual situation of a fresh supermarket enterprise in Changchun, this paper designs and develops a fresh supermarket business system that meets the needs of the enterprise, and applies the clustering analysis and association rules in data mining to the sales data analysis of the enterprise. The related technology in data mining is combined with business system to help enterprises to formulate more reasonable management and marketing strategies. The main work is divided into two parts: the first part is the realization of the basic functions of the fresh supermarket business system. The system adopts the popular Java language and the open source SSH framework to complete the development of each module of the system. The basic business modules of the system mainly include user management, acceptance of sales flow, acceptance of warehousing, distribution of stores, allocation of stores, reporting of loss of goods, purchase plan, store ordering, commodity management, information management of goods in transit, return handling, etc. Make every link of fresh supermarket controllable and traceable, complete data sharing and the information of the whole business, improve the level of information in the process of fresh and fresh circulation. In the course of commodity information management, open source Open Layers is used to construct lightweight Web GIS module based on JavaScript, which can obtain geographic location information of transportation more intuitively. In addition, after considering the urgent need of purchasing staff for mobile office, we completed the design and development of the procurement management mobile App, to help buyers to better carry out the procurement work; The second part of the work is the design and implementation of fresh supermarket sales data analysis, because the fresh supermarket in the business process will accumulate a large number of data and the sales data of each store is different. Therefore, we can use the sales data of each store to cluster the stores. However, after studying the fuzzy C-means clustering algorithm in clustering analysis, it is found that the fuzzy C-means clustering algorithm has introduced the mixed F distribution in statistics after it needs to specify the number of clusters before clustering. An improved scheme to obtain the best number of clusters is proposed, which can solve the problem that the fuzzy C-means clustering algorithm needs to specify the number of clusters in advance, and the improved algorithm is applied to the analysis of the sales characteristics of fresh supermarkets. Then the analysis of customer shopping basket is designed and implemented in allusion to the potential relationship between the customers of fresh supermarket and the purchase of goods. The transaction data of fresh supermarket customers every time they buy goods can be called shopping basket data, because the association rules in data mining technology are more suitable for the analysis of shopping basket transaction data of fresh supermarket customers. Therefore, the Apriori association rule mining algorithm is used to mine the association rules between the items of the shopping basket data. Through the analysis of the experimental results in the two modules, we can find the potential information in a large amount of data accumulated by the enterprise, and provide help for the actual management of the enterprise.
【學(xué)位授予單位】：吉林大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP311.13

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 金楓;;Web前端MVC框架的意義與前端發(fā)展方向展望[J];電腦知識與技術(shù);2016年03期

2 王亞蕓;;農(nóng)超對接模式下生鮮超市發(fā)展的問題及對策[J];現(xiàn)代商業(yè);2016年04期

3 楊鵬;鄒時林;;基于OpenLayers的WebGIS客戶端的研發(fā)[J];測繪與空間地理信息;2012年03期

4 李洋;孫永維;許冰;王英雙;;基于Ajax,Struts,Hibernate和Spring的J2EE架構(gòu)[J];吉林大學(xué)學(xué)報(信息科學(xué)版);2011年06期

5 張贊;張亞軍;;我國農(nóng)產(chǎn)品流通渠道終端變革路徑分析[J];現(xiàn)代經(jīng)濟(jì)探討;2011年05期

6 王愛平;王占鳳;陶嗣干;燕飛飛;;數(shù)據(jù)挖掘中常用關(guān)聯(lián)規(guī)則挖掘算法[J];計算機技術(shù)與發(fā)展;2010年04期

7 武寶珠;梁聲灼;牛德雄;;基于Struts2+Spring+Hibernate架構(gòu)構(gòu)建Web應(yīng)用系統(tǒng)[J];計算機與現(xiàn)代化;2009年08期

8 李小平;肖岳峰;宿元;宋瀚濤;姚永標(biāo);;基于J2EE多層架構(gòu)的Web開發(fā)框架研究[J];計算機應(yīng)用研究;2008年05期

9 孫吉貴;劉杰;趙連宇;;聚類算法研究[J];軟件學(xué)報;2008年01期

10 賀玲;吳玲達(dá);蔡益朝;;數(shù)據(jù)挖掘中的聚類算法綜述[J];計算機應(yīng)用研究;2007年01期

，

本文編號：2233069