Page 263 - DCAP310_INTRODUCTION_TO_ARTIFICIAL_INTELLIGENCE_AND_EXPERT_SYSTEMS
P. 263
Unit 14: Types of Learning
Notes
Example: Over the set of all people we wish to automatically distinguish some person
as either male or female based on a collection of n observations such as height, weight, pitch of
voice, etc. We stack these observations into a feature vector x.
If we have chosen the types of these observations intelligently then the distributions of these
characteristics may tell us how to classify some person described by x.
Example: Males will on average be taller, heavier, and have a lower pitched voice than
females. So based on some instance of these characteristics, the hope is that we will be able to
predict the sex of a subject with some amount of certainty.
Self Assessment
State whether the following statements are true or false:
3. Learning can be accomplished using same methods.
4. The information conveyed to the learner component is used to create and modify
knowledge structures in the knowledge base.
14.3 Performance Measures
A performance measure is a numeric description of an agency’s work and the results of that
work. Performance measures are based on data, and tell a story about whether an agency or
activity is achieving its objectives and if progress is being made toward attaining policy or
organizational goals. In technical terms, a performance measure is a quantifiable expression of
the amount, cost, or result of activities that indicate how much, how well, and at what level,
products are provided to customers during a given time period. “Quantifiable” means the
description can be counted more than once, or measured using numbers. “Activities” mean the
work, business processes and functions of Washington state government agencies. “Results” are
what the agency’s work is intended to achieve or accomplish for its customers. Performance
measurement is the process of collecting, analyzing and/or reporting information regarding
the performance of an individual, group, organization, system or component. It can involve
studying processes/strategies within organizations, or studying engineering processes/
parameters/phenomena, to see whether output are in line with what was intended or should
have been achieved. There are several reasons to measure, monitor and report performance of
our work. It’s the right thing to do: We measure many things in our lives outside work. Our
children regularly bring home objective measures of their performance at school (i.e. test scores
and report cards). We get monthly measures of performance at home: investment performance,
water and electricity usage, and so on. We monitor our health through a variety of measures of
how well our body is performing: weight, blood pressure, cholesterol levels. Work performance
is another aspect of our lives, and measuring it should be what we do.
Self Assessment
State whether the following statements are true or false:
5. Performance measures are not based on data.
6. Quantifiable means the description can be counted more than once, or measured using
numbers.
LOVELY PROFESSIONAL UNIVERSITY 257