Data Mining: What is data mining?

elements of data mining

extract, store, provide, analyze, present.

data

facts, numbers or text that can be processed by a computer. the amount and variety is huge

continuous innovation: example

grocery chain. oracle to find local buying patterns. bought diapers and beer. when they did weekly shopping. when they rarely shopped. made an insight on buying beer for the coming week.

MPP

massively parallel processors have order of magnitude in improvements in query time.

types of data

operational: transactional data. sales, costs, inventory, payroll and accounting. nonoperational data: industry sales, forecasts, macro economic. meta data: data about the data. logical database design or data dictionary.

information

patterns, associations, relationships of data makes this.

internal factors

price, product positioning, staff skills.

nearest neighbor method

a level of data analysis. classifies each record in a dataset based on a combination of the classes of the k records.

rule induction

a level of data analysis. extraction of useful if-then rules from data based on statistics.

artificial neural networks

a level of data analysis. non-linear predictive models learn by training and look like biology neural networks.

genetic algorithms

a level of data analysis. optimization techniques that use processes such as genetic combination, mutation, natural selection in a design based on the concepts of natural evolution..

decision trees

a level of data analysis. tree-shaped structures that represent sets of decisions. rules for dataset classification. CART. CHAID. urels to apply to new data to predict outcomes.

data visualization

a level of data analysis. visual interpretation of complex relationships in multidimensional data. graphic tools.

walmart and datamining

a pioneer. transform supplier relationship. point of sale transactions captured and sent to warehouse suppliers access and analyze. suppliers find buying patterns. local improved.

data warehouses

a process of centralized data management and retrieval. central repository of organizational data.

what allows warehouses

advances in data capture, processing power, data transmission, storage abilities. allows integration of databases into these

store

an element of data mining. manage the data in multidimensional systems.

analyze

an element of data mining. the data by application software.

levels of data analysis

artificial intelligence networks. genetic algorithms. decision trees. nearest neighbor method. rule induction. data visualization

types of data relationships

classes. clusters. associations. sequential patterns.

who uses data mining

companies with a strong consumer focus. retail, financial, communication, marketing.

associations

data can be mined to identify associations. beer diaper correlation

sequential patterns

data is mined to anticipate behavior patterns and trends.

clusters

data item are grouped according to logical relations or customer preferences. market segments.

data mining

date or knowledge discovery. is the process of analyzing data from different perspectives and summarizing it into useful info.

external factors

economic indicators. competition. customer demographics.

useful information

information that can increase revenue, cut costs or both.

how does mining work

it analyzes relationships and patterns in stored transaction data based on open-ended user queries. there are four types of relationships.

what data mining does

allows companies to determine relationships of internal factors to external factors. determine the impact on sales, satisfaction and profits. drill down into summary info to view transactional data.

provide

an element of data mining. data access to business analysts and information technology professionals

present

an element of data mining. the data in a useful format such as a graph or table.

extract

an element of data mining. transform and load transaction data onto the warehouse system.

data mining software

analytical tool to use for analyzing data. analyze from different dimensions, angles, categorize it and summarize relationships.

knowledge

a conversion of information. based on historical patterns or future trends. summary information. used to make decisions.

classes

stored data is used to locate data in predetermined groups.

data mining and fundamentals

technically it is the process of finding correlations or patterns among dozens of fields in a large relational database.

continuous innovation

technology of mining is not new. computer processing power, disk storage and statistical software are increasing the accuracy of data analysis and lowering costs.

query complexty

the more complex the queries and the greater the number of queries being processed the more powerful the system need be

size of the database

the more data being processed and maintained the more power an infrastructure system is needed

centralization of data

used to maximize user access and analysis. users can access the data freely.