Keywords: certifications certification certificate certificates online certify low cost business automation workflow courses course software process system systems specialit specialists

S.  Maurer,
Academic Director

It is best to rëmember that no single method is a universal solution and some are better with particular types of data.  Often,  using more than one method for data mining yields the best results.  In doing so,  you can use different data mining tools or choose a software package that incorporates more than one method.

Know Your Visitor

Decision Trees.  A decision tree is essentially a flow chart of questions or data points that ultimately leads to a decision.  For example,  a car-buying decision tree might start by asking whether you want a 1999 or 2000 model year car,  then ask what type of car,  then ask whether you prefer power or economy,  and so on,  until IT¹ determines what might be the best car for you.  Decision tree systems try to create optimized paths,  ordering the questions so a decision can be made in the least number of steps.

After you've set up and configured the sample databases and the Lotus Workflow Architect,  you'll find examples of many of the concepts and elements that process designers can use in building workflow processes with Lotus Workflow.  

The workflow processes are started by employees that have a job to do.  For example,  the employee receives a letter that starts the process.  Or,  the employee has just received a loan application to process,  a software-bug report,  or expense account to process.  After an employee has the information to start the process,  there is typically a form that must be filled out.  This form then flows through the corporation.

More abstractly,  a workflow is a pattern of activity enabled by a systematic organization of resources,  defined roles and mass,  energy and information flows,  into a work process that can be documented and learned.

In order to accommodate the handling of many diverse processes in a workflow,  each different approval application is given an application profile.  You could have,  for example,  a profile for a special bid process,  a different one for employee travel expenses,  another for purchase requisitions,  and so forth.  All of these approval processes can reside in the same database with different sets of approvers and users.

Relational database storage and Management technology is adequate for many data mining applications less than 50 gigabytes.  However,  this infrastructure needs to be significantly enhanced to support larger applications.  Some vendors have added extensive indexing capabilities to improve query performance.  

BPM can be used to understand organizations through expanded views that would not otherwise be available to organize and present.  These views include the relationships of processes to each other which,  when included in the process model,  provide for advanced reporting and analysis that would not otherwise be available.  BPM is regarded by some as the backbone of enterprise content Management.

Whether or not existing processes are considered,  the aim of this step is to ensure that a correct and efficient theoretical design is prepared.

Estimation and Prediction.  Estimation guesses an unknown value,  such as income,  when you know other things about a person.  Prediction guesses a future value,  such as the probability of buying a car next year,  when a person hasn't done IT¹ yet,  or the expected number of stocks that a person will trade in the coming year.  The same algorithms can perform estimation and prediction.

It also involves running what-if analysis on the processes: What if I have 75% of resources to do the same task? What if I want to do the same job for 80% of the current cost?

Business Process Management [BPM] IDE is an integrated design environment used to design processes,  rules,  events and exceptions.  Creating a structured definition of each process is very important to any business and the IDE enables a business user to design all processes with no help from IT¹.

Use caution when listening to traditional offline data-mining professionals.  IT¹ is likely that your Web site operates at a faster rate,  involves more data,  and is more mission critical than anything they've done.  Traditionalists are familiar with a more relaxed world: where data mining is used once per month,  rather than once per click; where data accumulates in gigabytes per year rather than gigabytes per month; and where a crashed application needs to be fixed in the morning,  rather than instantly by redundant machines and fail-safe rollover.

The development of the concept of workflow occurred over a series of loosely defined,  overlapping,  eras.