What is Business Intelligence ?

Typical use of BI :

In Identifying changes in purchasing patterns

–      Important life events change what customers buy.

In Entertainment

–      Netflix has data on watching, listening, and rental habits.

–      Classify customers by viewing patterns.

In Predictive policing

–     Analyze data on past crimes - location, date, time, day of week, type of crime, and related data.

Ethic Guide : Unseen Cyberazzi

Data broker or Data aggregator

–      Acquires and purchases consumer and other data from public records, retailers, Internet cookie vendors, social media trackers, and other sources.

–      Data for business intelligence to sell to companies and governments.

Cheap cloud processing of consumer data easier, less expensive.

Processing happens in secret.

Data brokers enable you to view data stored about you, but ...

–      Difficult to learn how to request your data,

–      Torturous process to file for it,

–      Limited data usefulness.

Functions of a data warehouse

–      Obtain data from operational, internal and external databases.

–      Cleanse data.

–      Organize and relate data.

–     

Catalog data using metadata.

Reporting Application

The purpose of reporting application is to create meaningful information from disparate data sources and deliver information to user on time.

Basic operations:

1. Sorting  
2. Filtering
3. Grouping          
4. Calculating      
5. Formatting

RFM Analysis :
-Recently

-Frequently

-Money

Unsupervised data mining

•       No a priori hypothesis or model.

•       Findings obtained solely by data analysis.

•       Hypothesized model created to explain patterns found.

Supervised data mining

•       Uses a priori model.

•       Prediction, such as regression analysis.

Market-Basket Analysis

–      Identify sales patterns in large volumes of data.

–      Identify what  products customers tend to buy together.

–      Computes probabilities of purchases.

–      Identify cross-selling opportunities.

Decision Tress

•       Unsupervised data mining technique.

•       Hierarchical arrangement of criteria to predict a value or classification.

•       Select attributes most useful for classifying “pure groups.”

BI for security trading

•       Quantitative applications using BigData and BI.

–      Analyze immense amounts of data over a broad spectrum of sources.

–      Build and evaluate investment strategies.

Knowledge Management (KM)

Creating value from intellectual capital and sharing knowledge with those who need that capital.

•       Preserving organizational memory

Capturing and storing lessons learned and best practices of key employees.

•       Scope of KM same as SM in hyper-social organizations.

Benefit of Knowledge Management

1.      Improve process quality.

2.      Increase team strength.

3.      Enable users to use organization’s collective knowledge.

Drawback of Expert System

1. Difficult and expensive to develop.

–      Labor intensive.

–      Ties up domain experts.

2. Difficult to maintain.

–      Changes cause unpredictable outcomes.

–      Constantly need expensive changes.

3. Don’t live up to expectations.

–      Can’t duplicate diagnostic abilities of humans.

Content Management Systems (CMS)

            Manage and delivery of documents, and other employees/users knowledge.

Challenge of CMS :
- Huge Databases.

- Dynamic Content.

Alternative of CMS

•       In-house custom development

–      Customer support develops in-house database applications to track customer problems.

•       Off-the-shelf

–      Horizontal market products (SharePoint).

–      Vertical market applications.

•       Public search engine

–      Google, Bing.

Hyper-social knowledge management

–      Social media, and related applications, for management and delivery of organizational knowledge resources.

Hyper-organization theory

–      Framework for understanding KM.

–      Focus shifts from knowledge and content to fostering authentic relationships among knowledge creators and users.

Guide:

            Semantic Security

1. Unauthorized access to protected data and information.
2. Physical security

Ø  Passwords and permissions.

Ø  Delivery system must be secure.

3. Unintended release of protected information through reports and documents.

Data Mining In Real World

Problems:

–      Dirty data

–      Missing values

–      Lack of knowledge at start of project

–      Over fitting

–      Probabilistic

–      Seasonality

–      High risk with unpredictable outcome