DATA WAREHOUSE AND BUSINESS INTELLIGENCE 6
DataWarehouse and Business Intelligence
Differencesbetween data warehouses and databases
Database is any collection of data organized for storage, accessibility,and retrieval. There occurs different types of databases but the mostcommon is the OLTP (on-line analytic processing). A data warehouse isan OLTP database itself. Databases are based on operationalprocessing while data warehouse is based on information processing.It can also be argued that databases store the current data whichalways promised to be up-to-date. Data warehouse on the other handstores information that is mainly historical and whose accuracy hasto be maintained over time. On the issue of functions, databases areused for day-to-day activities while data warehouses are usually usedfor long-term decision support. Data warehousing is associated withmanagers and analysts while databases are associated with commonusers such as database professionals.
Anotherdifference that occurs between the two is that databases haveorientation on transactions, but data warehouses have orientation inthe analysis. Data warehouses have their summarizations inconsolidated form, but databases have primitive and highly primitive.The above illustrations show that the database is appropriate for thetraditional system of data storage strategy while data warehouse is amodern form of data storage strategy which is used for producingenormous information.
Whatare the data warehouse technologies?
Datamining is the deliberative act of obtaining hidden predictiveinformation from large databases. It is one of the powerfultechnologies that help organisations to focus on the most importantdata in their data warehouses (Nick, 2004). Data mining equipmentsenable companies to be predictive about future trends and behavioursas it provides information that experts may not think about becauseit lies outside their expectations. Data mining saves time as it isfast and efficient.
Fordata mining to be productive, several applications have to beemployed including massive data collection, powerful multi-processorcomputers, and data mining algorithms. All of these technologies aimat generating new business opportunities by providing capabilitiessuch as automated prediction of trends and behaviours and automateddiscovery of previously unknown patterns. The most usefultechnologiesused in data mining include unnaturally neural networks, decisiondiagrams, rule induction and nearest neighbour method (Nick, 2004).
Customerrelationship management (CRM)
CRMis a devised way that integrates the concepts of data mining,knowledge management, and data warehousing in order to beef-up theorganisation’s decision making to regain the long-term andprofitable relationships with the customers. It normally deals withgathering all customer-related information into a single compositionand combines it with robust analysis of sales, marketing, and serviceproposals. The focus of customer relationship management is toautomate sales force by building separate applications for CRM havingtheir own databases. CRM has brought benefits such as minimizedETL-processing, more timely and high quality data, alignments ensuredwith business objectives, reducing the operational cost, improvingcustomer services, and customer retention. These benefits have beenmeasured through statistical data analysis to show the trends overthe years. CRM employs such tools as customer mart, marketing mart,sales mart, service mart, and partner for success.
On-lineanalytic processing (OLAP) for business intelligence
On-lineAnalytical Processing (OLAP) is often considered best if a companywishes to analyze data. OLAP is, therefore, a technology that is usedto organise large business databases and support businessintelligence (David, 2009). OLAP databases are organised into one ormore sections, and each section is organised and designed a sectionadministrator to fit the way that the company can retrieve dataeasily. OLAP databases comprise of two types of data: measures anddimensions. Moreover, OLAP databases help arrange data by many levelsof detail, using the categories that the company is familiar with toanalyze data. OLPA is dynamic as the company can change the directionof the analysis. OLAP framework, for instance, has been integratedinto some commercial e-commerce systems (Data Mining Page, 2014).
Whatare the relationships between data warehouse and businessintelligence?
Todetermine the relationship between data warehouse and business, it isadvisable to understand the two terms. Data warehouse is a placewhere storage of data occurs for it to be archived, to be analyzed,and for security purposes. Data is a technique used by many companiesto come up with facts, trends, or correlations. Data marts aresegments or levels of information and data that are grouped togetherto provide relevant information into that segment or level. Datamarts are usually smaller than data warehouse. As a result, reportingtools are made to generate reports from the data warehouse (David,2009). Data warehouse also manages cleansing, data transformation,data acquisition persistence management, and storing information.
Businessintelligence implies the tools and systems that play a crucial rolein strategic planning process of a corporation. The systems usedallow the company to collect, store, access, and analyze corporatedata to enhance decision making. Business mainly deals with areas ofcustomer profiling, customer support, market research, and productprofitability (David, 2009). Business intelligence, therefore, is theleveraging of data warehouse to aid make business decisions andrecommendations. Data and information rules are leveraged by businessintelligence to help to make decisions along statistical analysis,OLAP (online transaction and processing), and data mining tools.Business intelligence operations are more expensive than datawarehouse ones.
Thereexists a correlation between data warehousing and businessintelligence. Business intelligence is a term that is commonlyassociated with data warehousing. Business intelligence indulgesitself in various activities to gather today’s market informationthat can also constitute that of their competitors (David, 2009).Business intelligence relies on data warehouse to get data so thatthey can simplify information and crunch it. Data warehouse has torely also on business intelligence to be fed with the simplified andanalyzed data. The raw data in the data warehouse has to betransformed so as to meet requirements and objectives of the company.
Thetransformation is done through a process known as Extraction,Transformation, and Loading (ETL). Once the data has been convertedand fed into the databases, it is ready for the next processing.These databases are the one referred to as data warehouse databases.Building of data marts follows. The whole process, therefore, hasintelligence involved in business (David, 2009). This is datawarehousing and the system involved from the start to end is known asbusiness intelligence system. It can be deducted that datawarehousing and business intelligence are interrelated.
Insummary, it can be noted from the above discussion that datawarehousing helps a company to store data while business intelligencehelps the company in controlling the data for effective decisionmaking as well as forecasting. Although most business intelligencesystems are based on data warehouse systems, not all of them are datawarehousing, that is, they do not take data directly from the sourcesystem. It can also be noticed that the term data warehouse mayeither mean data warehouse system or data warehouse database.Business intelligence, on the other case, can either mean businessintelligence system or business intelligence application.
DataMining Page (2014). AnIntroduction to Data Mining. RetrievedFrom:
David,T. (2009). Progressivemethods in data warehousing and business intelligence : concepts andcompetitive analytics.Hershey: Information Science Reference.
Nick,C. (2004). InternationalJournal of Intelligent Systems-Granular Computing and Data Mining.NewYork: John Wiley & Sons.