Preference Modelling preference, comparison, and alternatives

1
Preference Modelling---preference, comparison,
and alternatives

• Presented by Guixian Lin

2
Contents

• Motivation
• Relation among them
• Different views
• Preference modelling
• ConclusionFindings

3
Motivation

• Preference measurement is crucial to managerial
  decisions in marketing. Methods such as conjoint
  analysis, voice-of-the-customer methods, and,
  more recently, configurators, are used widely for
  product development, advertising, and strategic
  marketing.
• Typical situationAmong the alternatives, people
  need comparision, then get what they
  want(preference)

4
Different Views

• Psychologists
• study the process by which judgements of
  preference are formed, transformed and
  influenced.
• Economists
• preference for the participants to a market
  that will allow them to derive nice equilibrium
  results and useful comparative statistics
  propositions
• Philosophers
• the links between ethics and preferences
  and investigate the notion of rationality.
• Political scientists
• study decisions made by groups of people and
  design electoral systems.
• Operations researchers

5
Different Views

• List exhaustive
• computer scientists (incorporating notions of
  value in knowledge-based systems),
• statisticians (testing a hypothesis on collected
  preference information and building tools to
  represent it)
• marketers (optimising the characteristics of a
  product to increase its market share given what
  is known of the preferences of various segments
  of consumers)

6
Preference modelling

• Common Questions in preference modelling
• How to behave rationally in a specific
  situation?
• What are the preference models allowing us to
  reach this goal?),
• How do people make decisions in a specific
  situation?
• How to help someone to reach a decision in a
  specific situation?

7
Classical Preference modelling

• Use of a specific language
• E.g. reflexivity ,complete,transitive(A weak
  order).
• convex and monotonic(For utility function
  smooth downward sloping demand curves)
• Complete Preferences vs Incomplete
  Preferences
• Use of a specific syntax
• Emphasis on a number of particular situations
• decision-making with multiple criteria,
  decision-making under risk, decision-making under
  uncertainty

8
Preference its Theoretical basis

• Fuzzy logical connectives
• Valued binary relations
• Valued preference modelling
• Similarity relations and valued orders
• Aggregation operations
• Ranking procedures
• Multiple criteria decision making

9
Decision theory

• What is decision?A decision is a choice made by
  some entity of an action from some set of
  alternative actions.
• What Makes a Decision Good? A good decision
  identifies an alternative that the decision maker
  believes will prove at least as good as other
  alternative actions.
• How Should One Formalize Evaluation of
  Decisions? Good decisions are formally
  characterized as actions that maximize expected
  utility, a notion involving both belief and
  goodness.

10
Decision theory

• Multicriteria Decision Making
• "construction step" in which one or several
  "outranking relations" are built,
• "exploitation step" in which (in this
  particular problem formulation) outranking
  relations are used to derive a ranking of the
  alternatives
• Qualitative Decision Theory(in AI)
• The rationale for choosing among decisions no
  longer relies on probability theory or numerical
  utility functions.

11
ConclusionFindings

• Different situations need different preference
  models.
• Different preference model affect rankings
• A lot of work needs to do.