【正文】 ata mining attempts to discover statistical rules and patterns automatically from , data mining differs from machine learning in that it deals with large volumes of data, stored primarily on discovered from a database can be represented by a set of can discover rules from database using one of two models:In the first model, the user is involved directly in the process of knowledge the second model, the system is responsible for automatically discovering knowledge from the database, by detecting patterns and correlations in the on automatic discovery of rules has been influenced strongly by work in the artificialintelligence munity on machine main differences lie in the volume of data handled in databases, and in the need to access datamining algorithms have been developed to handle large volumes of diskresident data manner in which rules are discovered depends on the class of datamining illustrate rule discovery using two application classes: classification and and Geographic DatabasesSpatial databases store information related to spatial locations, and provide support for efficient querying and indexing based on spatial types of spatial databases are particularly important:Design databases, or puteraideddesign(CAD)databases, are spatial databases used to store design information about how objectssuch as buildings, cars or aircraftare important examples of puteraideddesign databases are integratedcircuit and electronicdevice databases are spatial databases used to store geographic information, such as databases are often called geographic information data are spatial in nature, but differ from design data in certain and satellite images are typical examples of geographic may provide not only location informationsuchas boundaries, rivers and roadsbut also much more detailed information associated with locations, such as elevation, soil type, land usage, and annual data can be categorized into two types: raster data(such data consist a bit maps or pixel maps, in two or more dimensions.), vector data(vector data are constructed from basic geographic objects).Map data are often represented in vector DatabasesRecently, there has been much interest in databases that store multimedia data, such as images, audio, and multimedia data typically are stored outside the database, in files the number of multimedia objects is relatively small, features provided by databases are usually not functionality bees important when the number of multimedia objects stored is such as transactional updates, querying facilities, and indexing then bee objects often have descriptive attributes, such as those indicating when they were created, who created them, and to what category they approach to building a database for such multimedia objects is to use database for storing the descriptive attributes, and for keeping track of the files in which the multimedia objects are , storing multimedia outside the database makes it harder to provide database functionality, such as indexing on the basis of actual multimedia data can also lead to inconsistencies, such a file that is noted in the database, but whose contents are missing, or vice is therefore desirable to store the data themselves in the and Personal DatabasesLargescale mercial databases have traditionally been stored in central puting the case of distributed database applications, there has usually been strong central database and network technology trends have bined to create applications in which this assumption of central control and administration is not entirely correct: increasingly widespread use of personal puters, and, more important, of laptop or “notebook” development of a relatively lowcost wireless digital munication infrastructure, base on wireless localarea networks, cellular digital packet networks, and other puting creates a situation where machines no longer have fixed locations and network plicates query processing, since it bees difficult to determine the optimal location at which to materialize the result of a some cases, the location of the user is a parameter of the example is a traveler’s information system that provides data on hotels, roadside services, and the like to about services that are ahead on the current route must be processed based on knowledge of the user’s location, direction of motion, and (battery power)is a scarce resource for mobile limitation influences many aspects of system the more interesting consequences of the need for energy efficiency is the use of scheduled data broadcasts to reduce the need for mobile system to transmit amounts of data may reside on machines administered by users, rather than by database , these machines may, at times, be disconnected from the Decisionsupport systems are gaining importance, as panies realize the value of the online data collected by their online transactionprocessing extensions to SQL, such as the cube operation, help to support generation of summary mining seeks to discoverknowledge automatically, in the form of statistical rules and patterns from large visualization systems help humans to discover such knowledge databases are finding increasing use today to store puteraided design data as well as geographic data are stored primarily as vector data。下面我們要研究幾個新的應(yīng)用，近年來它們變得越來越重要。我們可以通過使用簡單的SQL查詢語句提供大量用于決策支持的信息。數(shù)據(jù)庫應(yīng)用從廣義上可分為事務(wù)處理和決策支持兩類。數(shù)據(jù)挖掘這個概念廣義上講是指從大量數(shù)據(jù)中發(fā)現(xiàn)有關(guān)信息，或“發(fā)現(xiàn)知識”。但是，數(shù)據(jù)挖掘和機器學(xué)習(xí)的不同在于它處理的是大量數(shù)據(jù)，它們主要存儲在磁盤上。我們用如下兩個模型之一從數(shù)據(jù)庫中發(fā)現(xiàn)規(guī)則：● 在第一個模型中，用戶直接參與知識發(fā)現(xiàn)的過程● 在第二個模型中，系統(tǒng)通過檢測數(shù)據(jù)的模式和相互關(guān)系，自動從數(shù)據(jù)庫中發(fā)現(xiàn)知識。其主要的區(qū)別在于數(shù)據(jù)庫中處理的數(shù)據(jù)量，以及是否需要訪問磁盤。規(guī)則發(fā)現(xiàn)的方式依賴于數(shù)據(jù)挖掘應(yīng)用的類型?？臻g和地理數(shù)據(jù)庫空間數(shù)據(jù)庫存儲有關(guān)空間位置的信息，并且對高效查詢和基于空間位置的索引提供支持。另一個計算機輔助設(shè)計數(shù)據(jù)庫的重要例子是整合電路和電子設(shè)備設(shè)計圖。地理數(shù)據(jù)庫常稱為地理信息系統(tǒng)。地圖和衛(wèi)星圖像是地理數(shù)據(jù)的典型例子。地理數(shù)據(jù)可以分為兩類：光柵數(shù)據(jù)（這種數(shù)據(jù)由二維或更高維的位圖或像素圖組成）、矢量數(shù)據(jù)（由基本幾何對象構(gòu)成）。多媒體數(shù)據(jù)庫最近，有關(guān)多媒體數(shù)據(jù)（如圖像、聲音和視頻）的數(shù)據(jù)庫的研究很熱門。當(dāng)多媒體對象的數(shù)目相對較少時，數(shù)據(jù)庫提供的特點往往不那么重要?？傊?，事務(wù)更新、查詢機制和索引也開始變的很重要。構(gòu)造這種多媒體對象的數(shù)據(jù)庫的方法之一是用數(shù)據(jù)存儲描述屬性，并且跟蹤存儲這些媒體對象的文件。此外這種情況還會造成不一致，譬如一個文件在數(shù)據(jù)庫中做了記錄，但其內(nèi)容卻丟失了；或