is a useful model for explaining bottom up thinking. It seeks out and makes squishy correlations between seemingly unrelated data. (Nexialism) Using this model the computer operator is the conscious (left mind) the computer is the sub-conscious, (right mind) (data storage and analysis) and the computer screen is the interface (GUI) The brain fart occurs when the operator confuses the picture on the screen for the data, which in this case is ones and zeros.
IE the computer screen shows an interpreted sub-set of the data. But you must ask a question (curiosity) for the computer to respond (insight, intuition) with dreams, or symbols that it puts on the screen. A person who shows this ability in front of a psychiatrist is called Bipolar Manic/Depressive and the plug is pulled on the computer with dopazine or smarticide. In the mind this process is transparant to the user.
This is what is actually happening in the computer (right mind) the left mind interprets this and puts a picture on the screen that you can make sense of. What this interpretation looks like depends on the software installed. (wetware(imprinting + learning)) The data range being analyzed determines the tunnel.
The following is from Oracle, the premier data mining technology. Although quite primitive compared to the mind, it is still pretty awesome.
Generally, data mining (AKA bottom up thinking) is the process of analyzing data from different perspectives and summarizing it into useful information – information that can be used to increase revenue, cuts costs, or both. Data mining software is one of a number of analytical tools for analyzing data. It allows users to analyze data from many different dimensions or angles, categorize it, and summarize the relationships identified. Technically, data mining is the process of finding correlations or patterns among dozens of fields in large relational databases called data warehouses, ( AKA right mind)
Supervised Learning has the goal of predicting a value for a particular
characteristic, or attribute that describes some behavior. For example: (AKA curiosity)
The attribute being predicted is called the Target Attribute.
Unsupervised Learning has the goal of discovering relationships and patterns
rather than of determining a particular value. That is, there is no target attribute. (AKA sub-conscious free running)
Examples S1, S2, S3 illustrate Binary Classification – the model predicts one of
two target values for each case (that is, places each case into one of two
classes, thus the term Classification).
Example S4 illustrates Multiclass Classification – the model predicts one of
several target values for each case.
Example S5 illustrates Regression – the model predicts a specific target value for
each case from among (possibly) infinitely many values.
Example S6 illustrates One-class Classification, also known as Anomaly
Detection – the model trains on data that is homogeneous, that is all cases are in
one class, then determines if a new case is similar to the cases observed, or is
somehow “abnormal” or “suspicious”.
Example U1 illustrates Clustering – the model defines segments, or “clusters” of
a population, then decides the likely cluster membership of each new case.
Example U2 illustrates Associations – the model determines which cases are
likely to be found together.