丁穎+王愛菊+馬文越+黃繼海



摘 要: 為了提高對Web異常數(shù)據(jù)的檢測及挖掘能力,保障Web網絡數(shù)據(jù)庫的安全穩(wěn)定運行,進行Web異常數(shù)據(jù)挖掘的軟件開發(fā),提出一種基于堆棧彈出中斷屏蔽的Web異常數(shù)據(jù)挖掘方法,并在Bootloader 程序開發(fā)平臺上進行軟件開發(fā)。首先構建Web異常數(shù)據(jù)挖掘系統(tǒng)的總體結構模型,采用post關鍵字編譯方法進行Web異常數(shù)據(jù)的堆棧彈出設計,軟件模塊化設計包括程序加載模塊、數(shù)據(jù)寄存模塊、異常數(shù)據(jù)交互式編譯模塊和中斷屏蔽模塊,創(chuàng)建LabWindows/CVI工程文件進行軟件面板開發(fā),生成用戶界面文件,實現(xiàn)異常數(shù)據(jù)挖掘。測試結果表明,該系統(tǒng)能有效實現(xiàn)Web數(shù)據(jù)挖掘,準確挖掘概率有所提升。
關鍵詞: Web網絡; 異常數(shù)據(jù)挖掘; 軟件開發(fā); 堆棧彈出; LabWindows/CVI
中圖分類號: TN911?34; TP311 文獻標識碼: A 文章編號: 1004?373X(2017)18?0032?03
Development and improvement of Web abnormal data mining software
DING Ying1, WANG Aiju1, MA Wenyue2, HUANG Jihai1
(1. Zhengzhou Institute of Technology, Zhengzhou 450044, China; 2. College of Communication Engineering, Hunan University, Changsha 410000, China)
Abstract: In order to improve the detection and mining ability of Web abnormal data, and ensure the safe and stable operation of Web network database, the software of Web abnormal data mining is developed, a Web abnormal data mining method based on the stack pop?up interrupt mask is presented, and software development is conducted in the Bootloader program development platform. The overall structure model of the Web data mining system is constructed. The post keyword compiling method is used to carry out the stack pop?up design of Web abnormal data. The software module design includes program loading module, data storage module, abnormal data interactive compiling module and interrupt mask module. A LabWindows/CVI project file was created to develop the software panel, generate the user interface file and realize the abnormal data mining. The test results show that the system can effectively achieve Web data mining. The accurate mining probability has been improved.
Keywords: Web network; abnormal data mining; software development; stack pop?up; LabWindows/CVI
0 引 言
網絡技術的不斷普及和應用,催生了Web數(shù)據(jù)庫的跨越發(fā)展。Web數(shù)據(jù)庫存儲大量的網絡數(shù)據(jù),實現(xiàn)信息及時傳輸和定點調度,Web數(shù)據(jù)發(fā)生故障和遭受到網絡入侵時,可能產生Web異常數(shù)據(jù),通過對Web異常數(shù)據(jù)有效挖掘,保障Web網絡數(shù)據(jù)庫的可靠性運行,確保網絡安全和信息挖掘[1]。Web異常數(shù)據(jù)挖掘與常規(guī)大數(shù)據(jù)信息挖掘相比較,具有自身的特點和差異性[2],Web異常數(shù)據(jù)具有隱蔽性強和時間窗口較短的特點,常規(guī)采用預判篩選的關聯(lián)規(guī)則挖掘方法難以有效滿足準確挖掘的需求。本文進行Web異常數(shù)據(jù)挖掘的軟件開發(fā)與改進,通過對異常數(shù)據(jù)挖掘系統(tǒng)的改進設計,大大地縮短了收集數(shù)據(jù)的時間,提高了數(shù)據(jù)挖掘的工作效率。
1 軟件開發(fā)實現(xiàn)
1.1 Web異常數(shù)據(jù)的堆棧彈出設計
在上述進行Web異常數(shù)據(jù)挖掘的總體設計和軟件設計原理分析的基礎上,進行Web異常數(shù)據(jù)挖掘的開發(fā)設計。本文提出一種基于堆棧彈出中斷屏蔽的Web異常數(shù)據(jù)挖掘方法,并在Bootloader 程序開發(fā)平臺上進行軟件開發(fā),采用post關鍵字編譯方法進行Web異常數(shù)據(jù)的堆棧彈出設計。首先建立post關鍵字編譯決策數(shù)據(jù)集如表1所示。
在海量的Web數(shù)據(jù)背景下,選用SuperViVi作為Bootloader[3],通過設定鎖相環(huán)倍頻數(shù)識別這些指令的操作數(shù),結合表1建立后綴項表,在Web網絡數(shù)據(jù)庫中,建立Web異常數(shù)據(jù)挖掘的特征檢測模型,異常數(shù)據(jù)挖掘的傳輸模塊采用50 MHz參考時鐘作為調制信號,A/D分辨率可達0.45 Hz,兼容性估計模型的一個通信傳輸組件使用post關鍵字mach?mini2440.c進行編譯[4],進行Web異常數(shù)據(jù)的堆棧彈出設計,編譯代碼如下:endprint