Data Warehouse Definition
A fitting data warehouse definition would be “a place for storing data.” A data warehouse is a place that holds on to data, files, whose information is retrieved at any time via the Internet (whether through a wireless router, DSL, etc.). There are three constant activities of a data warehouse: staging, the process by which data is stored; integration, the meeting of data with operational systems; and access, the presentation of the data to machine users.
Data warehouses can be divided into subunits called data marts. A data mart is a component of the data warehouse that sends data to computer users; where data is presented in a comprehensible format to users. There can be numerous data mart subunits, but continual additions to data marts will create a complex structure that, once fallen, is impossible to rebuild.
Data warehouse concepts are certain terms and definitions about data warehousing that you need to know. One of these concepts, data warehouse dictionary, pertain to the data’s meaning, the data’s relationship to other data, its source of origin, its uses, and structure. While the data warehouse’s basic definition focus on the implantation of data, a neglected part of the definition is data extraction. Thus, business intelligence (BI) is also a part of the definition—when it is expanded to encompass all of the data warehouse operations.
The business intelligence definition refers to the gathering, storing, and releasing of data to better aid businessmen in the workplace. Applications of business intelligence include data mining, decision support systems, query and reporting, online analytical processing, statistical analysis, and forecasting.
Decision support systems collect information such as sales figures in a one-month period, projected figures based on current patterns of financial success or failure, and the consequences of certain actions should they be released. Business intelligence is “informational” as distinguished from other data that are “operational.” Operational data are needed to run machines and corporations, while informational data are present to help influence decisions. Business intelligence may often present decision results by way of graphs to show whether or not the decision helped the business or hurt the business.
Query and reporting refer to a search by a user on a search engine (query) as well as an announcement of results found by the search (reporting). This is done in cases where businesses examine what various population segments exist for the purpose of market segmentation—and how many individual age groups (for example) like a certain product. Forecasting involves the projection of a certain idea or concept into the future—that is, the direction of a certain thought if taken into the future. A good example of forecasting involves the acceptance of a certain marketing strategy for the next ten years (when the marketing strategy has proven ineffective for the last five years). Statistical analysis is simply the gathering of statistics, numbers and percentages regarding a given subject, topic, or product, assessment of the numbers and percentages and what they indicate for the business.
Data mining is an application of business intelligence—that is, the practice of decisions made via business intelligence. The data mining definition involves the use of data to draw correlations between different types of data. A good example would be to observe that people are buying products in larger sizes. An observer could then look to consumer buys before a certain time period and notice that Americans were not buying in bulk. He or she would then look at the economic decline of the last few years and conclude that the economy and financial hardship have forced consumers to buy products in bulk—buy large, save money. A person could observe celebrity donations to charity and see that they are in decline. The observer could then look back to a time before five years ago and notice that celebrities were more magnanimous with their donations. The observer could then conclude that economic hardship has forced celebrities to “penny-pinch” when it comes to their large donations.
Data mining is capable of far more than just targeting financial sales and the motivations for them. Data mining can also be used to assess what types of financial perks a card issuer like American Express could offer their customers to encourage them to spend more. Banks are now seeing the benefits of data mining, as they have started points rewards systems all across the country. The idea behind points rewards systems is that you must spend more to get more points, which then gets you higher-priced items for free with your bank.
A software data warehouse definition requires that data warehouses have both data-loading tools and data-extracting tools. The software or computer packages can range from one single operation to a myriad of operations within a complex system. Since data warehouse software can have such a variety of complex compositions, it is important to research well and plan what kind of software you would like to have before purchasing it and uploading it onto your business network.
The definition of a data warehouse may seem easy at first: it is all about storing and releasing data. However, if you peel the surface back, you will find that there is so much more than data at stake. For many companies, the question is, “What do we do with this data?”