【文章內(nèi)容簡介】 riented (OO) databases store both data and procedures acting on the data, as objects. ? Encapsulation Capability – The OO database can be particularly helpful in multimedia environments, such as in manufacturing sites using CAD/CAM. – OO databases can be particularly useful in supporting temporal and spatial(時空） dimensions. ? Terminology in the OO model includes: – objects, attributes, classes, methods, and messages. HUANG Lihua, Fudan University Hypermedia Database Model ? The hypermedia database model stores chunks of information in the form of nodes connected by links established by the user. ? The nodes can contain text, graphics, sound, fullmotion video, or executable puter programs. ? Users can branch to related information in any kind of relationship. HUANG Lihua, Fudan University A hypermedia database HUANG Lihua, Fudan University Multidimensional Database ? A variation of the relational model ? Use multidimensional structures to anize data and express the relationship between data. ? A dimension of the data : a side of a cube. ? ① 多維數(shù)組 – (北京， 1999年，彩電， 10000) – (地理位置，年份，產(chǎn)品類型，銷售額 ) ? ② 維的層次 – 例如：年、季度、月份、日期 – 國家、地區(qū)、省、城市 ? ③維內(nèi)元素的類 – 例如： 按產(chǎn)品的價格分成高、中、低檔。 – 按原材料的成本價格分類 HUANG Lihua, Fudan University Multidimensional data model HUANG Lihua, Fudan University Specialized Databases ? There are many specialized databases, depending on the type or format of data stored. – A geographical information database contains locational data for overlaying on maps or images. – A knowledge database stores decision rules used to evaluate situations and help users make decisions like an expert. – A multimedia database stores data on many media—sounds, video, images, graphic animation, and text. HUANG Lihua, Fudan University Content ? Database Concepts amp。 Technology –Experiment: ACCESS ? Database Trends ? Data Trends of Application –Data Warehouse –OLAP –DATA Mining ? Creating Database Environment HUANG Lihua, Fudan University 3. Data Trends of Application ? Data Warehouse ? OLAP ? DATA Mining HUANG Lihua, Fudan University From Database to Data Warehousing ? 隨著信息技術(shù)在企業(yè)的廣泛應(yīng)用 ， 企業(yè)積累了大量數(shù)據(jù) ? 企業(yè)所面對的問題不是簡單地處理數(shù)據(jù) ， 而是如何使用數(shù)據(jù) 從操作處理 (Operational Processing) 到分析處理 (Analytical Processing) HUANG Lihua, Fudan University Operational vs. Decision Support Systems ? Operational Processing in Operational Systems – Support day to day transactions – Contain current, ―up to date‖ data – Examples: customer orders, inventory levels, bank account balances ? Analytical Processing in Decision Support Systems – Support strategic decision making – Contain historical, ―summarized‖ data – Examples: performance summary, customer profitability, market segmentation HUANG Lihua, Fudan University Why data warehouses? ? Decision Support Data – Are found in many different databases ? within the pany ? outside the pany – In practical terms, locating and integrating all this information in real time is very difficult ? Solution: – Create separate repositories of data for decision support – = data warehouses HUANG Lihua, Fudan University Data Warehousing ? Stores current and historical data – Consolidates data for management analysis and decision making – Supports reporting and query tools ? ―數(shù)據(jù)倉庫之父” W. H. Inmon給的定義：數(shù)據(jù)倉庫就是一個用以更好地支持企業(yè)或組織的決策分析處理的，面向主題的，集成的，不可更新的，隨時間不斷變化的數(shù)據(jù)集合。 HUANG Lihua, Fudan University Characteristics of Data Warehouses 1) Organization. Data are anized by detailed subjects. 2) Consistency. Data in different operational databases may be encoded differently. In the warehouse they will be coded in a consistent manner. 3) Time variant. The data are kept for 5 to 10 years so they can be used for trends, forecasting, and parisons over time. 4) Nonvolatile. Once entered into the warehouse, data are not updated. 5) Relational. The data warehouse uses a relational structure. 6) Client/server. The data warehouse uses the client/server to provide the end user an easy access to its data. HUANG Lihua, Fudan University Data Warehouse Suitability Data warehousing is most appropriate for anizations in which some of the following apply. ? Large amounts of data need to be accessed by endusers. ? The operational data are stored in different systems. ? An informationbased approach to management is in use. ? There is a large, diverse customer base. ? The same data are represented differently in different systems. ? Data are stored in highly technical formats that are difficult to decipher. ? Extensive enduser puting is performed. HUANG Lihua, Fudan University Comparison with DB Systems ? DB Systems – 數(shù)據(jù)庫：操作型數(shù)據(jù)(Operational Data)，增、刪、改操作頻繁 – 數(shù)據(jù)庫核心：功能強大，面向 OLTP (Online Transaction Processing) 應(yīng)用 – 數(shù)據(jù)庫工具：以查詢工具為主 ? DWS Systems – 數(shù)據(jù)倉庫：分析型數(shù)據(jù) (Analytical Data)，極少有更新操作 – 數(shù)據(jù)倉庫管理系統(tǒng)：因極少有更新操作，故功能簡單 – 數(shù)據(jù)倉庫工具：以分析工具為主 HUANG Lihua, Fudan University OLTP 在線業(yè)務(wù)處理 Database OLAP 在線分析處理 DSS 決策支持系統(tǒng) Data Mining 數(shù)據(jù)挖掘 Data Warehouse HUANG Lihua, Fudan University 數(shù)據(jù)倉庫的層次結(jié)構(gòu) 檔案數(shù)據(jù) 原始數(shù)據(jù) 綜合數(shù)據(jù) 高度 綜合數(shù)據(jù) 銷售數(shù)據(jù) 匯總的銷售數(shù)據(jù) 分產(chǎn)線的 周銷售數(shù)據(jù) 分產(chǎn)線的 月銷售數(shù)據(jù) HUANG Lihua, Fudan University 數(shù)據(jù)倉庫的實現(xiàn) 取 出 數(shù) 據(jù) 數(shù) 據(jù) 轉(zhuǎn) 換 主題數(shù)據(jù) 主題 1 S S 主題數(shù)據(jù) 主題 2 S S 主題數(shù)據(jù) 主題 3 S S Infomart Datamart Infomart(信息超市） : Datamart（數(shù)據(jù)超市） : 這是一個應(yīng)用功能（或從應(yīng)用功能產(chǎn)生的輸出）可多次運行 數(shù)據(jù)倉庫，它是對應(yīng)業(yè)務(wù)問題的答案。 這是數(shù)據(jù)倉庫的一個數(shù)據(jù)子集，它對應(yīng)最終用戶的信息需 求。它比之?dāng)?shù)據(jù)倉庫中的數(shù)據(jù)要更加歸納、匯總一些。 Infomart Datamart Data Warehouse Operational Databases Internal data source External data source HUANG Lihua, Fudan University Data Marts Data Marts are an alternative used by many other firms is creation of a lower cost, scaleddown version of a data warehouse. They refer to small wa