Legal Department

1
Legal Department

• Case-Based Reasoning

2
Legal Department Members

• Melanie Austreng
• Julia Korpatcheva
• Alex Kovalenko
• Adena Oliver
• Susanna Szonyi

3
Case-Based Reasoning

• Description
• CBR is the process of solving new problems based
  on the solutions of similar past problems.
• It has been argued that case-based reasoning is
  not only a powerful method for computer
  reasoning, but also a universal behavior in
  everyday human problem solving.

4
Case-Based Reasoning

• Description
• CBR is a form of Artificial Intelligence (AI)
  that attempts to mirror the process of human
  reasoning, i.e., reasoning from experience
• CBR systems attempt to utilize previous
  experience in a specific area and thus to apply
  situations which are similar to the current
  problem to solve it. 
• CBR exhibits incremental, sustained learning from
  previous experiences, and retains these
  experiences in order to solve future problems.

5
Case-Based Reasoning

• Description
• Four Steps
• Retrievegiven a target problem, retrieve cases
  from memory that are relevant
• Reuse map the solution from the previous case to
  the target problem
• Revise test the new solution in the real world
  and if necessary, revise
• Retain store the resulting experience as new
  case in memory 

6
Case-Based Reasoning

• Advantages
• Sustained learning continually learn from
  capturing and retaining past solutions
• Improvement over time CBR grows and improves as
  further experience is gained and cases are added
• Integration with databases using a database
  provides a large memory for storing cases and
  also allows information to be shared with other
  departments

7
Case-Based Reasoning

• Disadvantages
• Optimal solution can't always be guaranteed
• Restricted to its case libraries and has only a
  limited degree of flexibility that can be
  achieved by adaptation (difficult to incorporate
  external data)
• Limitations associated with finding an indexing
  scheme and designing a set of adaptation rules
• Costly to build and maintain

8
Case-Based Reasoning

• Recommendations
• We do not recommend implementing case-based
  reasoning
• Expensive and time consuming to implement and
  maintain.
• Lacks the common sense of human beings
• Artificial intelligence technology is still in
  its infancy
• Using law students to research case information
  would be cheaper for 240co legal department

9
Case-Based Reasoning

• Recommendations
• The system is useless for e-commerce legal
  issues, because it has no past experience (no
  database) and there are no established laws at
  this time.
• If 240co could find a company that leased out
  case-based reasoning technology, it might be
  cheaper than developing our own systems, and we
  would then be able to use CBR to our advantage.