The most common feeds into BI Systems are:
Operational databases
Social Data
Purchased data
Employee Knowledge
The four main uses of BI are:
Project management
Problem solving
Deciding
Informing
An example of using BI for project management is the creation of Gear UP determining ways to reduce costs determining # of soccer balls to sell comparing sales to forecast( the creation of Gear UP, determining ways to reduce costs, determining # of soccer balls to sell, comparing sales to forecast )
An example of how to use BI for problem solving is creating Gear Up determining how to reduce costs finding out # soccer balls can we sell comparing sales to forecast( creating Gear Up, determining how to reduce costs, finding out # soccer balls can we sell, comparing sales to forecast )
An example of using BI for making decisions is creating Gear up determining how to reduce costs finding out # soccer balls to sell comparing sales to forecast( creating Gear up, determining how to reduce costs, finding out # soccer balls to sell, comparing sales to forecast )
An example of using BI to inform is creating Gear up determining how to reduce costs finding out # soccer balls to sell comparing sales to forecast( creating Gear up, determining how to reduce costs, finding out # soccer balls to sell, comparing sales to forecast )
Typical uses for BI are
identifying changes in purchasing patterns
keeping track of people's reading, watching and listening habits
predictive policing
The three primary activities in the BI process are
Acquiring data
Performing analysis
Publishing results
The acquiring data activity also includes
cleansing
organizing and relating
cataloging
The analysis activity includes
data mining
knowledge management
Results may be published by
printing and distributing via email or a collaboration tool
publish on a web server or via SharePoint
publish on a BI server
automate results via a web service
Pushing information means automatically sending it to users
Pulling information means that the user must make a request for it
As long as two tables have a field in common, then the tables can be joined.
A data broker is also known as a data miner.
Data brokers gather data from
public records
retailers
internet cookie vendors
social medial trackers
others
Data brokers gather this data to create business intelligence that they then sell to organizations and the government.
Cloud processing has made processing consumer data cheaper and faster.
It is easy to request and make sense of data that has been gathered about yourself.
Cleansing data means scanning it for viruses.
BI analysis is performed on the same servers that run the operations of the organization.
The DBMS of a data warehouse consists of
Data warehouse metadata
Data warehouse database
From a data warehouse can be created one or several data marts
Examples of BI tools that can be used for specific data marts are
Clickstream analysis tools
store management tools
inventory management tools
In the US, it is not difficult for an aggregator to purchase information including an individual's name, address, phone number, income, marital status, name(s) of spouse and children, etc.
Source data can be too granular or not granular enough.
the basic operations of reporting applications are
Sorting
Filtering
Grouping
Calculating
Formating
The acronym RFM stands for .
Unsupervised data mining refers to data mining performed by an inexperienced analyst.
Cluster analysis is an example of unsupervised data mining
Supervised data mining uses an a priori model to compute the outcome of that model.
Market-basket analysis consists of
identifying sales patterns in large volumes of data
identifying products that customers tend to purchase together
probabilities of customer purchases
identification of cross-selling opportunities
Decision trees are a supervised data mining technique.
Big Data applications can only be used with structured data.
The storage part of Hadoop is called the HDFS - Hadoop Distributed File System.
The processing part of Hadoop is called MapReduce.
The goal of Knowledge Management is to use the organization's collective knowledge to improve process quality and team strength.
Knowledge management creates value from intellectual capital income( value from intellectual capital, income ) and allows for the sharing of knowledge with those that need that capital.
Expert systems are created by gathering knowledge from human experts. predicting the best outcome.( gathering knowledge from human experts., predicting the best outcome. )
Expert systems are cheap and easy to develop.
Expert systems are difficult to maintain.
Expert systems often disappoint because they can't duplicate the diagnostic abilities of humans.
CMS stands for .
Some of the challenges of content management systems include
huge databases
dynamic content
documents that don't exist in isolation
perishable content
documents in many languages
An organization can provide content management by
in-house development
off-the-shelf purchase
using a public search engine like Google
The main functions of a data warehouse are
obtain or extract data
cleanse data
organize and relate data
create and maintain the catalog
