Transcript CAST - Criminal Analysis System Techniques

Building Negotiation Support
Systems to support On Line
Dispute Resolution
John Zeleznikow
School of Information Systems
Victoria University
[email protected]
The rising tide of pro se
representation
• A recent American Bar Association study
has shown
• 1) at least one of the parties were selfrepresented in over 88% of domestic
relations cases; and
• 2) both parties were self-represented in 52%
of the cases
Negotiation Support Systems and On
Line Dispute Resolution
The rising tide of pro se
representation
• (Quatrevaux 1996) notes that there is a shortfall in
legal systems for poor persons in the United
States.
• (Branting 2001) claims that domestic abuse
victims are particularly likely to have few
resources and little opportunity to obtain the
services of a lawyer.
• He states that the growth of the consumer
movement has increased the trend for pro se
litigation to organise their own litigation.
Negotiation Support Systems and On
Line Dispute Resolution
The rising tide of pro se
representation
• The growing availability of books, document kits
and computerised forms, together with the
increasing availability of legal materials on the
World Wide Web, has increased the opportunities
for pro
• This swelling tide of pro se litigants constitutes a
growing burden not only to the judiciary but the
entire legal process.
Negotiation Support Systems and On
Line Dispute Resolution
The rising tide of pro se
representation
• Typically, unrepresented litigants:
• 1) Extend the time taken for litigation – due to
their lack of understanding of the process.
• 2) Place themselves at a disadvantage compared to
their opponent(s).
• 3) Place the judicial decision-maker in the difficult
position of deciding how much support and
forbearance the decision-maker should offer to the
pro se litigant.
Negotiation Support Systems and On
Line Dispute Resolution
The rising tide of pro se
representation
• In this seminar we shall illustrate how web-based legal
decision support systems can provide important support for
pro-se litigants.
•
•
•
•
Examples will be taken from:
Eligibility for Legal Aid
Property distribution in Family Law
Negotiation
• We shall also consider how NSS can help provide useful
ODR advice.
Negotiation Support Systems and On
Line Dispute Resolution
Decision Support Systems
• When considering decision making as a
knowledge-manufacturing process, the purpose of
a decision support system is to help the user
manage knowledge.
• A decision support system fulfils this purpose by
enhancing the user's competence in representing
and processing knowledge.
• It supplements human knowledge management
skills with computer-based means for managing
knowledge.
Negotiation Support Systems and On
Line Dispute Resolution
Decision Support Systems
• A decision support system accepts, stores, uses,
receives and presents knowledge pertinent to the
decisions being made.
• Its capabilities are defined by the types of
knowledge with which it can work, the ways in
which it can represent these various types of
knowledge, and its capabilities for processing
these representations.
Negotiation Support Systems and On
Line Dispute Resolution
Legal Decision Support Systems
• The development of our legal decision support
systems has led to:
• (i) Consistency – by replicating the manner in
which decisions are made, DSS are encouraging
the spreading of consistency in legal decisionmaking.
• (ii) Transparency – by demonstrating how legal
decisions are made, legal DSS are leading to a
better community understanding of legal domains.
Negotiation Support Systems and On
Line Dispute Resolution
Legal Decision Support Systems
• (iii) Efficiency - One of the major benefits of
decision support systems is to make firms more
efficient.
For example, since VLA handles a large number of cases
(and rejects even more) and unlike most private legal
firms, it does not bill either by the hour, or even by each
individual case, it is most concerned about being efficient.
About 60% of VLA’s labour costs are devoted to assessing
whether applicants are eligible for legal aid
Negotiation Support Systems and On
Line Dispute Resolution
Legal Decision Support Systems
Clearly, it would be preferable if VLA lawyers spent their
time providing clients with legal aid, rather than assessing
the eligibility of clients for grants.
• (iv) Enhanced support for dispute resolution
- Users of legal decision support systems
are aware of the likely outcome of litigation
and thus are encouraged to avoid the costs
and emotional stress of legal proceedings
Negotiation Support Systems and On
Line Dispute Resolution
ODR in Australia
• The Victorian Department of Justice has
comissioned a report on Online Alternative
Dispute Resolution (www.justice.vic.gov.au).
• They classify the following forms of online ADR
• A) Facilitated negotiation e.g I-Courthouse;
• B) Automated negotiation e.g. Cybersettle
• C) Negotiation Support Systems e.g. Smartsettle
Negotiation Support Systems and On
Line Dispute Resolution
Types of Disputes Resolved
• Family Law
• Internet Name Disputes
• Consumer Transactions
• Insurance Disputes
Negotiation Support Systems and On
Line Dispute Resolution
Cases attracted and outcomes
achieved
• Most successful sites were:
• www.squaretrade.com
• www.cybersettle.com
• Funding came from:
•
•
•
•
•
Philanthropic grants;
Government funding;
User fees;
Membership fees
Advertising revenue.
Negotiation Support Systems and On
Line Dispute Resolution
Limitations to access to ODR sites
• Access to a computer with minimum
hardware and software support;
• Accessibility of sites to people with
disabilities and slow on-line connection
speeds;
• Language of services offered.
Negotiation Support Systems and On
Line Dispute Resolution
Australian ODR sites
• www.notgoodenough.com.au and
www.complain.com.au are consumer complaints
sites.
• www.adronline.com.au is a private venture
offering automated negotiation, mediation and
arbitration.
• www.retailtenancy.nsw.gov.au permits an
applicant to register for mediation services and
lodge forms, pay fees and engage in a bulletin
board discussion with either an advice officer or a
mediator
Negotiation Support Systems and On
Line Dispute Resolution
UN Forum on ODR
• Interested in Online Dispute Resolution?
• Always wanted to visit Australia?
• … Now here’s your excuse
The Third Annual Forum on Online Dispute
Resolution will be held in Melbourne, Australia 56 July 2004
• hosted by the International Conflict Resolution
Centre at the University of Melbourne
• in collaboration with the United Nations
Economic and Social Commission for Asia and the
Pacific
Negotiation Support Systems and On
Line Dispute Resolution
Forum Aims
To report on online dispute resolution activities at the
regional and global level
To take stock of accumulated knowledge and expertise
To increase understanding the current state of technology
and its likely future directions
To discuss the major barriers to online dispute resolution
and ways of expanding its benefits.
Negotiation Support Systems and On
Line Dispute Resolution
Plan early
Participation is free but is limited to 150 participants worldwide.
Participants are responsible for their own travel. However, discounted
accommodation is available for Forum participants. There are a limited
number of stipends covering travel and accommodation for residents of
less developed countries.
To register your interest in either participating or presenting a paper,
please fill in the form attached and return to [email protected].
To ensure you don’t miss out, return your expression of
interest today!
Negotiation Support Systems and On
Line Dispute Resolution
GETAID - a Web based decision support
system for determining eligibility for legal aid
• After passing a financial test, applicants for legal
aid must pass a merits tests.
• The merits test involves a prediction about the
likely outcome of the case if it were to be decided
by a Court.
• VLA grants officers, who have extensive
experience in the practices of Victorian Courts,
assess the merits test.
Negotiation Support Systems and On
Line Dispute Resolution
GetAid
• A decision tree is a directed graph in which the
nodes represent domain concepts and possible
values for each concept are captured in arcs
emerging from each node.
• Leaf nodes represent conclusions.
• Figure 1 depicts a decision tree that represents
reasoning used by VLA lawyers, to determine
whether an applicant for legal aid, who is
scheduled to appear in a minor (Magistrates)
court, has met statutory guidelines.
Negotiation Support Systems and On
Line Dispute Resolution
Figure 1 Decision tree for eligibility for legal aid.
Acquittal
prospects
2. not reasonable
1. reasonable
Likely
penalty
Severity of
fine
1. is at least $750
2. is less than $750
Special
circumstan
ces
1. no
Guidelines not satisfied.
5. community based
order for less than 200
hous
3. a community based
order for more than
2. intensive
200 hours
correction order
4. a community based order
1. actual or suspended
and the defendent cannot
imprisonment
communicate
Guidelines satisfied
2. yes
Refer to specialist
Negotiation Support Systems and On
Line Dispute Resolution
Sequenced Transition Networks
• Sequenced Transition Networks (STNs) extend
Decision Trees with labeled nodes and arcs.
• The variation, illustrated in Figure 2, enables user
prompts and explanations to be more easily
generated directly from information in the tree and
facilitates the maintenance of the system.
Negotiation Support Systems and On
Line Dispute Resolution
0
02
2. does not have
reasonable prospects
for acquital on the
primary matter
The
applicant
Not sure :
Acquittal prospects
diagram
1. lead to an actual or
suspended term of
imprisonment
2. lead to an
021,
intensive
022,
023,
correction order
1. has reasonable
prospects for acquital on
the primary matter
024,
011
01
The penalty
Not sure :
Penalty summary
A conviction offences diagram
is likely to
1. is at least as
severe as a fine
of $750
Guidelines satisfied
3. lead to a
community based
order for more than
200 hours
4. lead to a community
based order and the
defendent cannot
communicate their
rehabilitation needs to the
Court
Not sure :
Penalty summary
offences $750
diagram
2. is not as severe as a fine
of $750
5. lead to a fine or community work order of less thn 200 hours
1. are no special
circumstances
0121
0251
0121 R550
Guidelines not satisfied.
0251 R552
012,
025
0122
0252
Refer to specialist
There
2. may be special circumstances
Not sure :
Special circumstances
diagram
Negotiation Support Systems and On
Line Dispute Resolution
Sequenced Transition Networks
• The STN based approach can facilitate the development of
knowledge based systems for knowledge that can readily
be captured as a decision tree.
• However, not all knowledge is procedural.
• For example, the node in Figure 2 relates to the prospects
for acquittal of an applicant for legal aid.
• Determining the possibility of a successful defence to a
charge requires legal expertise.
• Such knowledge cannot readily be modeled using decision
trees.
Negotiation Support Systems and On
Line Dispute Resolution
Toulmin Argument Structures
• Toulmin concluded that all arguments, regardless of the
domain, have a structure, which consists of six basic
invariants: claim, data, warrant and backing (as well as
modality and rebuttal which we do not use).
• Every argument makes an assertion. The assertion of an
argument stands as the claim of the argument.
• A mechanism is required to act as a justification for the
claim, given the data— the warrant
Negotiation Support Systems and On
Line Dispute Resolution
Toulmin Argument Structures
The backing supports the warrant and in a legal argument
is typically a reference to a statute or precedent case.
• Toulmin structures provide a representation, which allows
for a separation of inferencing from explanation.
• An explanation is generated by reproducing the data,
warrant or backing and is performed after, and
independently from the claim inference.
Negotiation Support Systems and On
Line Dispute Resolution
Toulmin Argument Structures
• Since experts could not readily represent knowledge about
an applicant's prospects for acquittal as a decision tree, we
decided to model the process as a tree of Toulmin
arguments.
• The first of these is illustrated in the following figure.
• During knowledge acquisition, the expert is prompted to
articulate factors (data items) that may be relevant in
determining a prospect for an acquittal claim, without any
consideration about how the factors may combine to
actually infer a claim value.
Negotiation Support Systems and On
Line Dispute Resolution
overall strength of
evidence on key
elements of charge
overall strength of
circumstantial
evidence
Police
engaged in
illegal activity
without
authorisation
overwhelming
extremely strong
strong
moderately strong
weak
overwhelming
extremely strong
strong
moderately strong
weak
strength of crown case
direct evidence exists
compelling indirect evidence exists
some indirect evidence exists
no evidence exists
client's instruction
Accused not
informed of
right to
silence
direct evidence exists
compelling indirect evidence exists
some indirect evidence exists
no evidence exists
direct evidence exists
Admission compelling indirect evidence exists
obtained
some indirect evidence exists
under duress
no evidence exists
Evidence was obtained
illegaly
Propensity for the
evidence to be unduly
prejudicial to the
accused
very likely
likely
arguable
not likely
Liklihood that crucial
crown evidence will be
ruled inadmissable
very likely
likely
arguable
not likely
Negotiation Support Systems and On
Line Dispute Resolution
overwhelming
extremely strong
strong
moderately strong
weak
full admission
partial admission
complete denial
positive defense offerred
no instructions
very likely
likely
arguable
not likely
The applicant
has reasonable prospecst for
acquital on the primary matter
does not have reasonable
prospecst for acquital on the
primary matter
Argument Tree for Acquittal
Prospects
• It was difficult for the principal domain expert to
articulate the ultimate argument (the argument on
the extreme right of the previous figure).
• She could not express her heuristic as rules
because the way in which the factors combine is
rarely made explicit.
• Her expertise was primarily a result of the
experience she had gained in the domain.
Negotiation Support Systems and On
Line Dispute Resolution
Argument Tree for Acquittal
Prospects
• Although it is feasible to attempt to derive heuristics, the
approach we used was to present a panel of experts with an
exhaustive list of all combinations of data items as
hypothetical cases and prompt the panel for a decision on
acquittal prospects.
• Six experts and the knowledge engineer were able to
record their decision in all of the exhaustive hypothetical
cases (for that argument) in approximately 40 minutes.
The decisions from each rater were merged to form a
dataset of 600 records that were used to train neural
networks.
Negotiation Support Systems and On
Line Dispute Resolution
GetAid
• The GetAid system is currently being used by Victorian
solicitors on-line, in conjunction with web based
lodgement of applications for legal aid (see
http://www.vla.vic.gov.au/upload/briefcase_12-2002.pdf) .
• It can be viewed at http://203.143.70.182/getaid/login/
• Toulmin Argument Structures have also been used to build
up Split-Up – a system which advises upon property
distribution in Ausstralian Family Law.
Negotiation Support Systems and On
Line Dispute Resolution
Alternative Dispute Resolution
• (Katsh and Rifkin 2001) state that compared to litigation,
negotiation has the following benefits
• Lower cost
• Greater speed
• More flexibility in outcomes
• Less adversarial
• More informal
• Solution rather than blame-oriented
• Private
Negotiation Support Systems and On
Line Dispute Resolution
On Line Dispute Resolution
• On Line Dispute Resolution has additional
benefits:
• Disputants do not have to meet face-to-face: an
important factor if there has been a history of
violence;
• Mediation can occur at any time, with participants
located in different countries
Negotiation Support Systems and On
Line Dispute Resolution
Computer Tools for Supporting
Legal Negotiation
• [Chung and Pak 1997] stress that although negotiation is a
human problem, computers are already at the bargaining
table to transform the negotiation process.
• The Harvard Negotiation Project of [Fisher and Ury 1981]
claimed that the few generally accepted principles of
successful negotiation require world knowledge beyond
the scope of existing artificial intelligence methods.
• Principled negotiation advocates separating the problem
from the people.
• It promotes a focus on interests of the party rather than
allowing negotiation to deteriorate into a contest of 'who
will back down first' as occurs in Positional Negotiation.
Negotiation Support Systems and On
Line Dispute Resolution
Computer Tools for Supporting
Legal Negotiation
• [Fisher and Ury 1981] advocate the notion of Know your
best alternative to a negotiated agreement (BATNA) —
The reason you negotiate with someone is to produce
better results than would otherwise occur.
• If you are unaware of what results you could obtain if the
negotiations are unsuccessful, you run the risk of entering
into an agreement that you would be better off rejecting;
OR Rejecting an agreement you would be better off
entering into.
• Split-Up uses KDD (Knowledge Discovery from
Databases) to provide advice upon obtaining your
BATNA
Negotiation Support Systems and On
Line Dispute Resolution
Australian Family Law
• Family-Winner uses artificial intelligence and game theory
to advise in what order mediators should attempt to settle a
dispute and how they should perform trade-offs
• Family law varies from other legal domains in that in
general:
• There are no winners or losers — in most common law
domains one party to a legal dispute wins a case whilst the
other loses;
• There are a vast amount of litigated Family law cases each
year
• Parties to a family law case often need to communicate
after the litigation has concluded.
Negotiation Support Systems and On
Line Dispute Resolution
Australian Family Law – Property
Distribution
• In determining the distribution of property under
Australian divorce law a judge performs the following
functions:
• She determines assets of the marriage the Court is
empowered to distribute — common pool determination;
• She determines what percentage of the common pool each
party to the marriage receives — percentage split
determination; and
• She determines a final property order in line with decisions
made in the percentage split and common pool
determination.
Negotiation Support Systems and On
Line Dispute Resolution
Modeling the Common Pool
Distribution
• Using a jurisprudential theory that we have developed, the
task of common pool determination was classified as
being suited to modeling using rule based reasoning
• We have modeled the task using Sequenced transition
networks (STN).
• The next diagram is a STN which indicates whether a
car used by one of the partners to the marriage, can
be considered marital property.
Negotiation Support Systems and On
Line Dispute Resolution
Initial Node
The vehicle
was
2. a gift or
inheritance
1. neither a
gift nor
inheritance
The gift
benefited
1. only the
recipient
It was
2. both parties or
the family
1. acquired
during the
marriage
2. acquired
before or after
the marriage
The vehicle
was
2. maintained or
improved during
the marriage
There is
2. a party outside
the marriage
The vehicle is
registered to
1. no
evidence of
a sham
The vehicle will be
included in the pool but
the Court cannot order
a transfer
2. evidence of a
sham
1. the husband or
wife
The vehicle will be included in
the pool and the Court can
order a transfer
1. not maintained or
improved during the
marriage
The vehicle will be
excluded from the pool
Negotiation Support Systems and On
Line Dispute Resolution
Discretionary Reasoning in Split-Up
• The section of the Family Law Act dealing with the
percentage of the common pool each partner receives is
highly discretionary.
• Ascertaining knowledge about how a judge weighs and
combine factors is difficult in that a guessed numerical
weighting is unlikely to represent the weight of the factor
in the context of a large number of interdependent factors.
• We wished to generate a descriptive model of the domain,
and neural networks combined with a rule based reasoner
was deemed appropriate
Negotiation Support Systems and On
Line Dispute Resolution
Discretionary Reasoning in Split-Up
• Although the Act presents a flat list of relevant factors
without specifying how these factors relate to each
other, we believe domain specific knowledge is
crucial in specifying relationships between factors
and for the elicitation of the other factors which are
relevant but not explicitly mentioned in the statute.
• The Split—Up system has a hierarchy of 99 factors which
domain experts indicated were relevant for a percentage
split prediction.
Negotiation Support Systems and On
Line Dispute Resolution
Discretionary Reasoning in Split-Up
Ninety-four variables were identified as relevant
for a determination in consultation with experts.
• The way the factors combine was not elicited from
experts as rules or complex formulas
• Split Up is on line at
http://www.ballarat.edu.au/~astranieri/splitup/splitup.php
Negotiation Support Systems and On
Line Dispute Resolution
Discretionary Reasoning in Split-Up
• The choice of inferencing mechanism chosen
depended upon jurisprudential concepts we
developed.
• We use the argumentation theory of Stephen Toulmin
to justify our decisions.
• The Family Law Act (1975) directs a decision maker to
take into account the past contributions of each party to a
failed marriage in addition to their resources for coping
with life into the future.
Negotiation Support Systems and On
Line Dispute Resolution
Discretionary Reasoning in Split-Up
• Values on the 94 variables were to be extracted from cases
previously decided, so that a neural network could learn to
mimic the way in which judges had combined variables.
• Rather than offering one definition for contributions and
one for needs, the statute presents a 'shopping list' of
factors to be taken into account in arriving at a property
order.
• Although the statute presents a flat list of relevant factors
without specifying how these factors relate to each other,
we realised that the factors could be placed in a hierarchy.
Negotiation Support Systems and On
Line Dispute Resolution
Discretionary Reasoning in Split-Up
• The development of the hierarchy required specific
knowledge supplied by domain experts.
• A sophisticated hierarchy of ninety-four factors was
elicited.
• The hierarchy provides a structure that was used to
decompose the task of predicting an outcome into thirtyfive sub-tasks.
• Outputs of sub-tasks further down the hierarchy are used
as inputs into sub-tasks higher in the hierarchy.
Negotiation Support Systems and On
Line Dispute Resolution
Percent time he worked full time
Percent time he worked part time
Percent time she worked part time
Percent time she worked full time
Cash from her
Cash from him
Dollars brought in by husband for
businesses
Dollars brought in by wife for
businesses
Dollars brought in by husband for
assets
Dollars brought in by wife for
assets
Relative effort by way
of salary
Relative contributions
by way of cash
Relative contributions
by way of business
Relative contributions
by way of assets
Significance of salary
to the marriage
Significance of
business
Significance of cash
Relative direct
contributions
Relative indirect
contributions
Length of
marriage
Relative home
maker
contribution
Significance of assets
Relative labor contribution
Relative superannuation
Contributions
of husband
relative to
wife
Number of years of marriage
Number of years of cohabitation
Common pool
value
Relative household duties
Relative child rearing
His age
group
His health
His skills
are in
demand
His work
experience
He currently
works
He has capacity
to work
His educational
level
Employment is
available
Level of
wealth
His age group
His health
His employment
prospects
His personal
prospects
His obligation to
dependents
His future needs
Her future needs
His financial
resources
Number of his pre primary dependents
Number of his primary dependents
Number of his secondary dependents
Number of dependents with severe health
problems
His access obligations
Negotiation Support Systems and On
Line Dispute Resolution
Future needs
of husband
relative to
wife
Percentage of
assets to the
husband
His access obligations
His salary amount
Current realisable value
Payout date
His salary resources
His business asset
bracket
His superannuation
His business resources
His help from a
partner
His business income
His help from family
Her age
group
Her health
Her skills
are in
demand
Her work
experience
She currently
works
She has
capacity to
Her educational
level
Employment is
available
Her age group
Her health
Her employment
prospects
Her personal
prospects
Her obligation to
dependents
Her financial
resources
Number of her pre primary dependents
Number of her primary dependents
Number of her secondary dependents
Number of dependents with severe health
problems
Her access obligations
Her salary amount
Current realisable value
Payout date
Her business asset
bracket
Her business income
Her salary resources
Her superannuation
Her business resources
Her help from a
partner
Negotiation Support
Systems and
Her help
from
family
Line
Dispute
Resolution
On
Modeling Family Law Property
Distribution
• The Split Up project applies Knowledge
Discovery from Databases (KDD) to predict
property division decisions made by judges
of the Family Court of Australia following a
divorce.
• We have noted that:
• KDD is particularly suited to the discovery
of knowledge in discretionary domains
Negotiation Support Systems and On
Line Dispute Resolution
Modeling Family Law Property
Distribution
• KDD is particularly suited to the discovery of
knowledge in discretionary domains.
• The discernment of tasks suited to KDD from
those more appropriately suited to other methods
relies heavily on the jurisprudential concept of
open texture.
• The argumentation theory proposed by Toulmin
can be used for the representation of domain
knowledge in the data transformation phase.
Negotiation Support Systems and On
Line Dispute Resolution
Modeling Family Law Property
Distribution
• In the Split-Up project we wished to model how
Australian Family Court judges exercise discretion
in distributing marital property following divorce.
• Section 79(1) of the Family Law Act (1975)
empowers the Family Court to make orders
altering the property interests of parties to the
marriage but does not lay down procedural
guidelines for judicial decision makers.
• In practice, judges of the Family Court follow a
five-step process in order to arrive at a property
order:
Negotiation Support Systems and On
Line Dispute Resolution
Modeling Family Law Property
Distribution
• 1. Ascertain the property of the parties.
• 2. Value all property of both parties.
• 3. Determine which assets will be paramount in
property considerations. This is referred to as
common pool property.
• 4. Determine a percentage of the property to be
awarded to each party.
• 5. Create an order altering property interest to
realise the percentage.
Negotiation Support Systems and On
Line Dispute Resolution
Modeling Family Law Property
Distribution
• The Split-Up system implements steps 3 and 4
above, the common pool determination and the
prediction of a percentage split.
• According to domain experts, the common pool
determination task (Step 3) does not greatly
involve the exercise of discretion, in stark contrast
to the percentage split task (Step 4).
Negotiation Support Systems and On
Line Dispute Resolution
Modeling Family Law Property
Distribution
• Consequently, Split-Up implements the
common pool determination by eliciting
heuristics as directed graphs from domain
experts using a methodology we have called
sequenced transition networks.
• The following represents knowledge about
whether a vehicle is considered marital
property.
Negotiation Support Systems and On
Line Dispute Resolution
VEHICLES
Initial Node
The vehicle
was
2. a gift or
inheritance
1. neither a
gift nor
inheritance ?
The gift
benefited
1. only the
recipient
It was
2. both parties or
the family
1. acquired
during the
marriage
2. acquired
before or after
the marriage
The vehicle
was
2. maintained or
improved during
the marriage
There is
2. a party outside
the marriage
The vehicle is
registered to
The vehicle will be
included in the pool but
the Court cannot order
a transfer
1. no
evidence of
a sham
2. evidence of a
sham
1. the husband or
wife
The vehicle will be included in
the pool and the Court can
order a transfer
1. not maintained or
improved during the
marriage
Negotiation Support Systems and On
Line Dispute Resolution
The vehicle will be
excluded from the pool
Modeling Family Law Property
Distribution
• Domain expertise in family law is represented in
the Split-Up system as Toulmin arguments.
• The Toulmin Argument Structure we use
facilitates the development of hybrid systems and
the generation of explanations.
• Ninety-four variables were identified as relevant
for a determination in consultation with experts.
• The way the factors combine was not elicited from
experts as rules or complex formulas
Negotiation Support Systems and On
Line Dispute Resolution
Modeling Family Law Property
Distribution
• Values on the 94 variables were to be extracted
from cases previously decided, so that a neural
network could learn to mimic the way in which
judges had combined variables.
• Toulmin concluded that all arguments, regardless
of the domain, have a structure that consists of six
basic invariants: claim, data, modality, rebuttal,
warrant and backing.
• Every argument makes an assertion based on some
data.
Negotiation Support Systems and On
Line Dispute Resolution
Toulmin Arguments
• The assertion of an argument stands as the
claim of the argument.
• Knowing the data and the claim does not
necessarily convince us that the claim
follows from the data.
• A mechanism is required to act as a
justification for the claim.
• This justification is known as the warrant.
Negotiation Support Systems and On
Line Dispute Resolution
Toulmin Arguments
• The backing supports the warrant and in a legal
argument is typically a reference to a statute or a
precedent case.
• The rebuttal component specifies an exception or
condition that obviates the claim.
• The Family Law Act (1975) directs a decision
maker to take into account the past contributions
of each party to a failed marriage in addition to
their resources for coping with life into the future.
Negotiation Support Systems and On
Line Dispute Resolution
Split-Up
• Rather than offering one definition for
contributions and one for needs, the statute
presents a 'shopping list' of factors to be taken into
account in arriving at a property order.
• Although the statute presents a flat list of relevant
factors without specifying how these factors relate
to each other, we realised that the factors could be
placed in a hierarchy.
• The development of the hierarchy required
specific knowledge supplied by domain experts.
Negotiation Support Systems and On
Line Dispute Resolution
Split-Up
• A sophisticated hierarchy of ninety-four factors was
elicited.
• The hierarchy provides a structure that was used to
decompose the task of predicting an outcome into thirtyfive sub-tasks.
• Outputs of sub-tasks further down the hierarchy are used
as inputs into sub-tasks higher in the hierarchy.
• Solid arcs in the following figure represent inferences
performed with the use of rule sets
• Whereas dashed arcs depict inferences performed using
neural networks (or indeed any other KDD technique).
Negotiation Support Systems and On
Line Dispute Resolution
Toulmin Arguments
• The classification scheme has been used to
classify tasks in the domain of
• family law (35 arguments),
• refugee law (200 arguments),
• copyright law (50 arguments),
• eligibility for legal aid (8 arguments) and
• the evaluation of eye-witness evidence
Negotiation Support Systems and On
Line Dispute Resolution
DEUS – a template-based NSS
• DEUS calculates the agreement and disagreement
between the litigants’ goals at any given time
• It helps mediators understand what issues are in
dispute and the extent of the dispute over these
issues.
• Split-Up is a hybrid rule-based/neural network
systems that advises upon property distribution
following divorce in Australia
Negotiation Support Systems and On
Line Dispute Resolution
Split-Up
. Whilst Split-Up is not a NSS, it can be used
to determine one’s BATNA for a
negotiation and hence provides an important
starting point for negotiations.
• Split-Up first shows both litigants what they
would be expected to be awarded by a court
if their relative claims were accepted.
• It gives them relevant advice as to what would
happen if some or all of their claims were rejected.
Negotiation Support Systems and On
Line Dispute Resolution
Split-Up as a Negotiation Support System
Users are then able to have dialogues with the
system to explore hypothetical situations to
establish clear ideas about the strengths and
weaknesses of their claims.
• Suppose the disputants' goals are entered into the
system to determine the asset distributions for
both parties.
• The Split-Up system provided the following
answers as to the percentages of the marital assets
received by each party in a hypothetical example:
Negotiation Support Systems and On
Line Dispute Resolution
W’s % H’s %
Given one accept’s W’s beliefs
65
35
Given one accept’s H’s beliefs
42
58
Given one accept’s H’s beliefs but
gives W primary residence of the
children
60
40
Negotiation Support Systems and On
Line Dispute Resolution
Split-Up as a Negotiation Support System
• Clearly custody of the children is very significant
in determining the husband’s property distribution.
• If H is unlikely to win custody of the children, the
husband would be well advised to accept 40% of
the common pool (otherwise he would also risk
paying large legal fees and having on-going
conflict).
• AdjustedWinner is a point allocation procedure
that distributes items or issues to people on the
premise of whoever values the item or issue more.
Negotiation Support Systems and On
Line Dispute Resolution
Adjusted Winner
• The two players are required to distribute 100
points across the range of issues in dispute.
• The Adjusted Winner paradigm is a fair and
equitable procedure. At the end of allocation of
assets, each party accrues the same number of
points.
• Although the system suggests a suitable allocation
of items or issues, it is up to the human mediators
to finalise the agreement acceptable to both
parties.
Negotiation Support Systems and On
Line Dispute Resolution
Adjusted Winner
• Arising from our work on the AdjustedWinner
algorithm, we noted that
• 1) The more issues and sub-issues in dispute, the
easier it is to form trade-offs and hence reach a
negotiated agreement;
• 2) We should choose as the first issue to resolve
the issue on which the disputants are furthest apart
- one wants it greatly, the other considerably less
so.
Negotiation Support Systems and On
Line Dispute Resolution
Family_Winner
• Family_Winner supports the process of negotiation by
introducing importance values to indicate the degree to
which each party desires to be awarded the issue being
considered.
• The system uses this information to form trade-off rules.
• As an allocation enables trade-offs to be executed, the
details of an allocation will have some input into the
resulting changes.
• It was necessary to obtain some negotiation specific data to
provide an insight into what change, and what amount of
change was expected as a result of trade-off execution.
Negotiation Support Systems and On
Line Dispute Resolution
Family_Winner
• 650 negotiated and litigated cases, provided by the
Australian Institute of Family Studies, were analysed.
• It was hence discovered the shift in most issues appeared
to be similar in number.
• We analysed the data and set numbers to each issue and
position pair, and developed trade-off rules.
• The trade-off rules are used to allocate issues according to
the logrolling strategy.
The system makes this analysis by transforming user input
into trade-off values, which show the effect of an issue’s
allocation on all unallocated issues.
Negotiation Support Systems and On
Line Dispute Resolution
Family_Winner
• To use Family_Winner we must assume
• A) The dispute can be modeled using principled
negotiation;
• B) That weights can be assigned to each of the
issues in dispute; and
• C) That sufficient issues are in contention to allow
each side to be compensated for losing an issue.
Negotiation Support Systems and On
Line Dispute Resolution
Family_Winner
• Users of the Family_Winner system enter
information such as the issues in dispute,
indications of each issue’s importance to the
respective parties and how the issues relate to each
other.
• An analysis of the information is compiled, which
is then translated into graphical trade-off maps.
• The maps illustrate the relevant issues, their
importance to each party and trade-off capabilities
of each issue.
Negotiation Support Systems and On
Line Dispute Resolution
Family_Winner
• The user is asked if the issues can be resolved in
its current form.
• If this is the case, the system then proceeds to
allocate the issue as desired by the parties.
• Otherwise, the user is asked to decompose an
issue chosen by the system as the least
contentious.
• The issue on which there is the least disagreement
is chosen to be the issue first considered.
Negotiation Support Systems and On
Line Dispute Resolution
Family_Winner
• Users are asked to enter sub-issues.
• As issues are decomposed, they are stored in a
decomposition hierarchy, with all links intact.
• This process of decomposition continues through the one
issue, until the users decide the current level is the lowest
decomposition possible.
• At this point, the system calculates which issue to allocate
to each party, then removes this issue from the each of the
party’s respective trade-off maps, and makes appropriate
numerical adjustments to remaining issues linked to the
issue just allocated.
Negotiation Support Systems and On
Line Dispute Resolution
Law and Negotiation
• Alexander has illustrated that women tend to be more
reluctant than men to continue conflict and are more likely
to wave their legal rights in a mediation session.
• If their major goal is to be the primary care giver for their
children, they may reach a negotiated settlement, which
whilst acceptable to them is patently unjust.
• The wife may for example, give the husband the bulk of
the property, in return for her being granted the primary
care of the children.
• Whilst such an arrangement may meet the goals of both
parents, it does not meet the paramount interests of the
children, who will be deprived of subsequent financial
resources.
Negotiation Support Systems and On
Line Dispute Resolution
Law and Negotiation
Family Law is one domain where mediation conflicts with
notions of justice.
Bargaining theory has its limitation in providing decision
support for family law negotiation.
However, we have noticed that various negotiation
domains are far more suitable than family law, for
modeling using integrated game theory and knowledge
based systems to advise upon trade-offs.
Thus, we decided to evaluate the use of Family_Winner in
other negotiation domains.
Negotiation Support Systems and On
Line Dispute Resolution
The Extension of Family_Winner to non
Family Law negotiation domains
• Since Family_Winner has been designed with a
view to being utilised in many negotiation
domains, there are no domain-specific
requirements that prospective cases need to
exhibit.
• We hence used Family_Winner in:
• A) Panama Canal dispute of 1974;
• B) Negotiation with terrorists;
• C) Company Mergers.
Negotiation Support Systems and On
Line Dispute Resolution
Panama Canal Treaty
• Family_Winner’s advice varied considerably from the settlement
obtained from negotiations held between the parties.
• As Family_Winner only has the ability to allocate issues to one of the
parties, a comparison based upon creative settlements is not possible.
• Also, there needs to be a better understanding of exactly what issues
are in contention, and how these issue definitions can be translated into
issues ready for allocation by Family_Winner.
• The Panama Canal treaty was particularly difficult to resolve, since the
disputants changed their goals during the dispute.
• Unfortunately, Family_Winner has no temporal aspects.
Negotiation Support Systems and On
Line Dispute Resolution
1988 hijacking of Kuwait Airlines flight 422
• Family_Winner advised upon negotiations
between the Kuwaiti government and the
hijackers.
• Issues covered were allowing the plane to land,
fuel, food, access to media, release of hostages,
release of convicted terrorists and the possible
conviction of the terrorists.
• Family_Winner’s advice
coincided with the
eventual outcome of the siege.
Negotiation Support Systems and On
Line Dispute Resolution
A merger negotiation between
companies
• Issues in dispute were, name, headquarters, chairman,
chief executive officer and layoffs.
• The settlement reached by Brams and Taylor was identical
to that achieved using Family_Winner.
• The results obtained from this example demonstrate the
effectiveness of point assignments to show the importance
value of an issue to a party, coupled with trade-off
equations to assist in the allocation of issues valued closely
by the respective parties.
Negotiation Support Systems and On
Line Dispute Resolution
Evaluating Family_Winner
• An investigation of our examples has shown the benefit of
Family_Winner for advising upon trade-offs,
compensation and the sequencing of negotiations as long
as the issues can be described and remain static and points
can be allocated to issues.
• The system cannot develop creative solutions nor
(currently) deal with temporal changes.
• Problems can occur in legal domains where a negotiated
settlement conflicts with our notion of justice.
Negotiation Support Systems and On
Line Dispute Resolution
Future Work
• We are developing systems for
• A) the UK building industry and
• B) plea-bargaining in Victoria, Australia.
– We have commenced a project on the Development and
testing of a United Kingdom web-based decision
support system for use in improving the consistency
and predictability of adjudicators’ decisions in building
construction disputes.
– We will build a web-based decision support system by
Negotiation Support Systems and On
Line Dispute Resolution
Building Industry Advisory System
– 1. Combining the records of project partners;
electronically publishing these records; creating a
standard hub where stakeholders can record
adjudication data; and data mining the records;
– 2. Decision modeling of the domain of building
industry dispute resolution by developing a web-based
model of legal reasoning in adjudication;
– 3. Commissioning a tool for predicting the course of
building dispute adjudications.
Negotiation Support Systems and On
Line Dispute Resolution
Building Industry Advisory System
– The project has as objectives:
• a.
The data mining of adjudication records.
• b. The development of new web-based, analytical tools
to identify the predictability of decisions involving many
variables, some of which will depend on legal reasoning,
and some on purely statistical analysis. Accordingly, a
decision model will be created based on a legalistic
approach (rule based and case based) to adjudication, as
derived from decisions that have been made by the UK
courts.
Negotiation Support Systems and On
Line Dispute Resolution
Building Industry Advisory System
• C. Testing the decision model against new adjudications which are
being handled by project partners, and the identification of issues
where the model aids predictability, and which may be used to
facilitate settlement and reduce conflict.
• D. The impact of the achieving of these objectives will be to have
greater transparency as to the cause of disputes which go to
adjudication and the likely result of such disputes.
• E. The project will also provide an easily accessed, one-stop decision
support system with information on procedure and case law and is
capable of being used by adjudicators to test reasoning before
publishing a decision.
Negotiation Support Systems and On
Line Dispute Resolution
Building Industry Advisory System
• The project will develop
• (1) Ability to test reasoning;
• (2) Ability to test likelihood of success;
• (3) Information and guidance on procedure, thereby reducing the
danger of procedural error by adjudicators;
• (4) Up-to-date information and guidance on matters that frequently
come to adjudication;
• (5) Up-to-date information and guidance on case law;
• (6) Training tool;
• (7) Reporter on trends
Negotiation Support Systems and On
Line Dispute Resolution
Future Work
• In conjunction with VLA and the Victorian Office of
Public prosecution we are developing ODR systems to
help with regard to plea bargaining about sentencing.
• In evaluating Family_Winner, we noted that disputant’s
preferences might well conflict with notions of justice.
• This is particularly true in Family Law, where the interests
of children are paramount.
Negotiation Support Systems and On
Line Dispute Resolution
Conclusion
• We hence realized certain limitations of bargaining-type
negotiation support systems such as Family_Winner.
•
An evaluation of Family_Winner in diverse domains such
as enterprise bargaining, company mergers and
international disputes has shown that it is useful for
advising upon trade-offs, compensation and the sequencing
of negotiations as long as the issues can be described and
remain static and points can be allocated to issues.
• The system cannot develop creative solutions nor
(currently) deal with temporal changes.
Negotiation Support Systems and On
Line Dispute Resolution