Jusletter IT

The aim of this paper is to present the Parent Plan Support System (hereinafter: the PPSS): a negotiation support system project designed to help divorcing parents to prepare the so-called parent plan, that is, an agreement between them concerning the exercise of their parental power over their children, and contacts between them and children, after the divorce is granted by the court. The legal context for the system is given by Polish family law.

The main ideas of the PPSS have been already presented in [Araszkiewicz, Łopatkiewicz, Zienkiewicz 2013a and 2013b]. The objective of this paper is more specific. The paper focuses on the structure of the system’s knowledge base and inference engine with particular emphasis on Case-Based Reasoning (CBR) structures, that is, on the role played by dimensions and factors in the system and on the operation of rule-based inference engine on the information provided by CBR knowledge representation structures.

The order of investigations is as follows. In the second Section the legal, dispute-theoretical and psycho-pedagogical background that motivated the development of the PPSS is given. Section 3 is devoted to detailed description of the model content of any parent plan. This content is vital to the structure of the knowledge base of the PPSS. In Section 4 the structure of the knowledge base is described with emphasis on CBR structures. Section 5 is devoted to discussion of an example of application of the inference engine of the system to a case. Section 6 concludes.

The problems of divorce proceedings, including the issue concerning parent plan, are regulated in the Polish law in the Code of Family and Custody law (hereinafter: the Code). According to the provision of art. 57 § 1 and 1a of the Code, the divorce court decides, amongst others, the issues concerning parental authority over minor children of the divorcing couple as well as the issue of contacts between them and their minor children. If the parents decide to prepare the so-called parent plan, that is, an agreement concerning the rules of conduct as regards their custody over the minor children after the divorce, the court is authorized take it into account, however, if and only if this agreement is compatible with the well-being of the child.

«The well-being of the child» is a very general concept that is evaluation-sensitive, context-sensitive and open-textured. It is sensitive for evaluation, because the answer to the question what is good for the child may differ with regard to system of values possessed by the person who makes this assessment. The concept is context-sensitive, because many circumstances concerning for instance the past events that occurred in the family may affect the assessment. Finally, the concept of open-textured, because the changes in social environment (in general) may lead to change of answers to the question what counts as realization of well-being of the child. The evaluation is not purely subjective, however, because the concept has been the subject of numerous judgments of the Polish Supreme Court (the PSC). Although case law is generally not formally binding in Polish legal system (as in other jus civile jurisdictions), as a practical matter it should be and it is actually taken into account by the judiciary.

Compatibility of the parent plan with the requirements of the well-being of the child is a necessary criterion of acceptance of the parent plan by the court (although it is not the only criterion, for instance, public interest should also be taken into account). On the other hand, the court is not bound by the parent plan. Therefore, it is necessary to persuade the court that the agreement prepared by the parents is actually a reasonable and applicable one, and, most importantly, that it promotes the well-being of the child (children). The tool that is presented in the present paper aims at facilitation of achieving of this objective.

However, the background of the idea is not created by legal issues only: there are further social and psycho-pedagogical implications of the idea of the parent plan. The situation of the divorce is in many cases one of the most stressful situations for both divorcing parents and their children. The divorce proceedings may lead to escalation of the conflict between the parents to the detriment of their children. The very idea of the parent plan is to grant the parents certain scope of autonomy as regards their decisions concerning exercise of parental authority and contacts with children after the divorce. In this way, this institution is in principle able to mitigate the adverse effects stemming from the formalism of civil procedure.

As for the character of the dispute that may arise between the parties to the negotiated agreement it is necessary to emphasize that due to the fact that the divorce proceedings are adversary, the parents may be naturally inclined to adopt an adversarial attitude also as regards the work on the parent plan. However, as the main premise concerning acceptability of their agreement is its compatibility with the well-being of the child, the optimal strategy for the parents would be to adopt a cooperative attitude leading to a win-win solution [Folberg et al. 2005, 79–80]. The process of transformation of the dispute may be difficult in this context due to emotional distress and mutual lack of trust of the parties. This opens a rational possibility of adoption of either a classical conflict resolution procedure (for instance, mediation), or, if simultaneous presence of parties is either not possible or not advisable, a usage of negotiation support systems as for instance the PPSS may be the justified choice. The system described in the present contribution aims at facilitation of communication between the parties, providing information about the scope of decisions that have to be made to finalize the work on the agreement, establishing relations between parents and their children and, finally, at contribution to the decision-making process of the parties.

As regards the issues related to developmental psychology and pedagogy that are involved in the system, the knowledge extracted from research on these subjects provides much valuable information for assessment of certain fact situations. Undoubtedly, divorce is one of the most distressing events that may affect the psychological development of minor children and their educational chances. One should also be aware of the fact that the PPSS performs educational functions as regards the divorcing parents, developing their awareness of complexity of issues related to proper upbringing of their children after the divorce.

In summing up the above considerations, it should be stated that the PPSS is an interdisciplinary project, comprising issues of law, psychology, pedagogy, dispute resolution theory and last but not least from the AI and Law research. The latter layer is the main topic of the present contribution.

The structure of an idealized parent plan has been prepared for the sake of systematization of knowledge in the PPSS. The tool presented in this paper should be able to facilitate generation of relatively exhaustive agreements between the parties, because leaving certain important issues outside of the scope of the parent plan increases probability of its rejection by the court.

Although, under the Polish law, there is no unified official template for parent plans, its content may be reconstructed on the basis of the set of issues that have legal and practical relevance as regards exercise of parental authority, including upbringing of children and development of relations in the family.

The structure of the parent plan begins with preliminary and meta-level information comprising three types of essential information: (1) formal information including the names of parents and children, their place of residence, health insurance information etc.; (2) declarations concerning their mutual attitude, including cooperation for the sake of the well-being of their children and (3) declarations and statements concerning realization of the plan in the future, including methods for introduction of modifications in the content of the plan.

The principal part of a parent plan are provisions that are grouped into subjects referred to as categories. The subject matter of the parent plan is divided into ten Categories: (1) contacts between the parents and the child; (2) contacts between the child, family members and other people; (3) education; (4) livelihood; (5) leisure, holidays and ceremonies; (6) worldview and raising issues; (7) healthcare issues; (8) contacts between the parents and information flow; (9) mode of conduct in emergency cases and (10) remaining issues. The extent of specificity of contractual rules that are established in each Category depends on the context that is given by the fact situation of any family. In the next Section it is presented how the structure of the idealized parent plan is represented in the PPSS knowledge base.

The PPSS knowledge system is based on propositional elements that are represented by both CBR structures (dimensions and factors) as well as by defeasible rules. Almost all of this knowledge is representable in predicate logic which is compatible with the possibility of PROLOG implementations. This Section is devoted to presentation of each type of knowledge representation structure in the system (cf. [Araszkiewicz, Łopatkiewicz, Zienkiewicz 2013a and 2013b]) with particular focus on CBR structures and their discussion in the light of relevant AI and Law literature.

Options. Options (O) are simple (mostly atomic) propositions representing states of affairs that can be chosen by the parents working of the agreement. Each Parent Plan generated in the PPSS is a subset of the set of all Options. The Options are systematized in sets referred to as Questions (Q). In each Parent Plan, exactly one Option should be chosen for each Question. We state that a given Parent Plan is complete if and only if it contains an Option for each Question. We also define a relation of Incompatibility (Inc) over a set of Options, and we state that if two Options from one and the same Parent Plan, o₁ and o₂, ∊ Inc, then this Parent Plant is inconsistent. A set of Questions forms a Subcategory (S). A set of Subcategories forms a Category. The set O is equivalent to the sum of the Categories. The definition of the set of Options enable us to define the Parent Plan as an outcome of the PPSS. Let O be the finite set of all Options (o₁… o_n), where:

∀_o ∃_Qi (o ∊ Q_i), and

 

∀oi ∀Qi, Qj (((oi ∊ Qi) ∧ (oi ∊ Qj)) → (Qi = Qj)).

 

A Parent Plan (PP) is a subset of O (PP ⊂ O), which is a selection of single elements from Questions:

 

∀oi ((oi ∊ PP) iff ∀Qi ((oi ∊ Qi) → (¬ ∃oj ((oj ∊ Qi) ∧ (oi ≠ oj))))

 

A Parent Plan is complete if and only if the following condition holds:

 

∀Qi, Qi ⊂ O, ∃ oi ((oi ∊ PP)).

 

Let Inc be a binary relation ranging over O (Inc ⊂ O²). We state that a Parent Plan is consistent if and only if the following condition holds:

 

∀oi, oj ((oi, oj ∊ PP) → (¬(Inc) oi, oj)).

 

A set of Options that is presented to the parent while constructing a concrete Parent Plan is referred to as a feasible set of Options.

Dimensions. The systems contain a library of dimensions: scalable knowledge representation structures [Ashley 1990]. Each dimension has two extreme points that represent states of affairs supporting opposite outcomes (classically, these extreme points may favor either a plaintiff or a defendant in the proceedings before the court; however, in the present system the extreme points of each Dimension point to the least and most favorable outcome as regards the realization of the well-being of the child). Each Dimension in the PPSS comprises the following elements:

• The name of the Dimension (identical to the name of Category C)
• A scale encompassing Valuations of the Dimension:
• Unsatisfactory (0)
• Sufficient (1–3)
• Good (4–6)
• Excellent (7–9).
• The alternative sets of Options that by default qualify as realization of the value of the child’s well-being.

There are 10 Dimensions in the system, one for each basic Category of the system. There are important differences regarding this account of Dimensions and the classical account developed by Rissland and Ashley, most authoritatively presented in [Ashley 1990]. The first important difference is that the extreme points of the dimensions do not support the opposing parties. The second difference concerns the structure of the dimension, which is much more complex here. Each point on the scale of the dimension may be satisfied by different Options (or sets of Options). Indeed, there may be two or more different Options that will be classified on a given point of scale of a given Dimension. The third difference is that each Parent Plan must be indexed by all of the Dimensions (in HYPO for instance, a case could be characterized by one or more dimensions). Fourth, the function of Dimensions here is not to give a basis for any analogical reasoning, but rather to represent the degree to which the value of the child’s well-being is satisfied by a given Parent Plan.

On a general level, however, there are also important similarities between the structure and the role of the Dimensions in this system and in HYPO. Both are scalable (although in PPSS, unlike in HYPO, only one type of scale is employed). Different Parent Plans, like cases in HYPO, may be compared regarding the realized point of scale of a given Dimension. Further, similarly to the HYPO program, the PPSS provides hypothetical suggestions concerning the assessment of the Parent Plan, if another and not the actually chosen Option is picked by the parent.

Defeasible Rules Set. The system comprises a huge (>100) set of defeasible rules extracted from the knowledge on pedagogy and developmental psychology as well as from the corpus of judgments of the PSC related to the concept of the well-being of the child. These rules are functions which accept Options as input and valuation on the scales given by Dimensions as output. As Options may be incompatible with each other, the system consists also of the rules for priority assignment between conflicting defeasible rules [Prakken and Sartor 1998].

Environmental Factors Set. The set of Environmental Factors (EFs) encompasses contextual information that is introduced by the users of the system in the initial stage of its operation. The factors employed here are binary knowledge representation structures: they may be either present or absent in a given case [Aleven 1997]. This set encompasses data such as distance from the residence of each parent to the schools in the surrounding area, the parents’ monthly salaries, as well as relevant data concerning the parents’ and the child’s personal characteristics. Like Options, there are Environmental Factors that are mutually incompatible (as for an obvious example, one child cannot have different ages at the same time). Certain EFs may also be incompatible with certain Options. Importantly, EFs may change the valuation function expressed by the default Defeasible Rules.

Cases and Legal Factors (LFs). The PPSS contains a database (>50) of cases decided by the PSC, related to the concept of well-being of the child. This case law corpus performs two important functions in the PPSS’ knowledge base. First, the legal cases serve as the source of extreme points on the Dimensions in the system, that is, such stereotypical fact patterns which violate well-being of the child or realize it to great extent. This is due to the fact that the PSC typically formulates statements about violations of necessary conditions of realization of this value. Second, LFs strengthen certain Options in the system by giving additional information concerning, for instance, favoring of this type of choice in one on the PSC judgments. Thus, legal cases help to attain the main goal of the PPSS, that is, development of such agreement that could be accepted by the court in divorce proceedings.

This Section is devoted to presentation of the operation of components of the PPSS. For the sake of brevity, we do not comment on the functions of purely technical components such as the Tutorial and Terms and Conditions Acceptance module.

The first component of the PPSS that is relevant from the perspective of computing knowledge is the Questionnaire. Before entering into their negotiations concerning the content of a parent plan, the parents have to give contextual information in the Questionnaire, for instance, information concerning their living conditions, the age of their children, the distance from their place of residence to school etc. The aim of using of the Questionnaire is to generate a set of Environmental Factors from both parties. The system detects incompatible EFs and signalizes the occurrence of them to the parties.

If the generated set of EFs is consistent, two important computational results are generated in the system. First, the set of EFs is mapped on the set of all Options in the system and the Options incompatible with the given EFs are marked as irrelevant as regards preparation of the parent plan in question. Second, the valuations given by the Defeasible Rules are changed in accordance with the defined Influence relation, if there are any such EFs that influence the default valuations prescribed by the rules.

The most important components of the system are: the Option Choice Module (OCM) and the Option Deliberation Module (ODM). They will be described in detail with the use of an example.

The function of the OCM is to present the set of possible choices to the negotiating parents and to facilitate the process of choice of Options. The parties go simultaneously through the set of all Categories in the system and are obliged to choose exactly one Option in each Question. Let us present an example of such a choice situation. The end of the example sections are signaled by █.

Example 1.

Let us assume that two parents, approximately 35 years of age and well educated, are intending to generate a Parent Plan. They are living separately in a big city and they have one child, a 4-year-old daughter, living with her mother.

The Question currently considered by the parents concerns the child’s contact with her maternal grandparents.

The following information is presented to each parent:

QUESTION [number]. Choose an Option concerning your child’s contact with her maternal grandparents. Choose a feasible Option that, according to your opinion, benefits the child to the greatest extent.

O (1). No contact with grandparents.

 

O (2). Rare contact with grandparents – a few times a year (for instance during holidays, family meetings).

 

O (3). More frequent contact (up to a few times a month).

 

O (4). Frequent contact (up to a few times a week), both planned and spontaneous.

 

O (5). Daily contact (including temporal exercise of custody by grandparents while parents are absent).

 

█

The choice of Options by the parents yields three computational consequences in the PPSS. Any choice made by the parent is stored in the catalogue of Options chosen by a given negotiating party. After that, the pairs is Options are classified as Matching, Neutral or Incompatible. Finally, as the Options chosen by the parents are antecedents of Defeasible Rules in the system, the defeasible rules engine assigns valuations to the chosen option. Note that this valuation may not be the default valuation, but, instead, a valuation generated due to the operation of the Influence relation that connects the EFs and the Defeasible Rules.

Example 1 continued.

The set of Defeasible Rules as regards the considered Question could appear as follows. Each Option produces a valuation on a scale of a relevant Dimension (Family contacts):

R1: O (1) => U (0)

 

R2: O (2) => S (1)

 

R3: O (3) => G (4)

 

R4: O (4) => G (6)

 

R5: O (5) => E (9)

This assignment may seem arbitrary, and to a certain extent this opinion would be justified. Such situations almost always occur when measure scales are applied to legal contexts [Alexy 2003]. However, in this context, the valuation is suggested by the exact wording of the judgment of the PSC, III CZP 42/88, which is quoted further in later investigations.

█

If a parent is not able to find an acceptable Option in the list provided by the system, he or she may type it in natural language. However, this Option cannot be evaluated by the system as regards the value on the scale of a given Dimension. The parent is given the option of assigning this valuation manually; however, the information concerning manual assignment of valuation is presented to the other parent. A parent may also find an Option provided by the system acceptable and at the same time disagree with its valuation. In such situations, the parent is again authorized to assign a valuation manually and to type in the reason for doing so, either by choosing a factor from a set suggested by the system, or by formulating it in natural language. The information about these actions is presented to the other party.

Example 1 continued.

Let us assume here that a parent chose an Option O (2) and received the following information from the system:

THE PARENT PLAN SUPPORT SYSTEM KINDLY INFORMS YOU THAT YOUR OPTION IS ASSESSED AS SATISFACTORY, YET THERE ARE BETTER OPTIONS. CONTACT WITH GRANDPARENTS IS VITAL TO THE CHILD’S WELL-BEING.

ACCORDING TO THE POLISH SUPREME COURT:

The parents are obligated, for the sake of the child’s well-being, to care for his or her development. In order to ensure the full development of his or her personality, they should facilitate his or her contact with his or her relatives (grandparents) if their attitude is right and they have a positive influence on the child. These contacts, often resulting from close emotional relationships, particularly as regards grandchildren and grandparents bringing the child up from their infancy, may influence the child positively and support the development of good relations in a multi-generation family. Such behavior of the grandparents, showing their close relationship and care for the child, is not against his or her well-being, but it promotes it (judgment of the PSC, III CZP 42/88).

WHAT IS YOUR DECISION?

• I AGREE WITH THIS ASSESSMENT
• I DISAGREE WITH THIS ASSESSMENT

If the parent chooses to disagree with this assessment, he or she is asked to assign a valuation for this decision manually, and to provide a reason for this disagreement. A list of reasons suggested by the system encompasses the Legal Factors extracted from the quoted decision of the PSC; however, the parent is given the option of typing in a reason manually. Let us assume that he or she disagrees with the assessment:

YOU HAVE JUST DISAGREED WITH THE ASSESSMENT OF YOUR OPTION PROVIDED BY THE PARENT PLAN SUPPORT SYSTEM. PLEASE CHOOSE ONE OF THE REASONS FROM THE FOLLOWING LIST:

• THE GRANDPARENTS DON’T KNOW THE CHILD
• THE GRANDPARENTS ARE NOT WILLING TO HAVE CONTACT WITH THE CHILD
• THE GRANDPARENTS COULD HAVE A BAD INFLUENCE ON THE CHILD OR TYPE IN YOUR OWN REASON.

If a parent decides to choose one of the Legal Factors presented to him or her, the valuation of an Option is determined by the Influence relation defined between this factor and the relevant Defeasible Rule. If a parent decides to assign his or her own valuation of his or her own reason, this choice is made visible to the other party.

The ODM enables the parents to negotiate the choices they made by means of the OCM. Let us remind that the Options chosen by the parents are classified as Matching, Neutral or Conflicting.

Let us assume that parent (1) chooses Option x to a given Question and that parent (2) chooses Option y to the same Question.

If Options are classified as Matching, then they have identical valuation. The system informs the parents about this result; however, if in the space of agreement there is a pair of Matching Options that are valued higher, the parties are informed about the suboptimal character of their choice.

If Options are classified as Neutral, it means that they are in the space of agreement between the parents; however, they can possess different valuation. A suggestion is given to the parent whose Option is valued lower (referred to as Parent #2) to choose another Option.

If Options are classified as Conflicting, then they cannot be accepted simultaneously. The parents are informed that they are not within the space of agreement. If the valuation of the chosen Option is unequal, the parent who chose the lower valued Option (Parent #2) is asked to choose another Option.

The negotiating parents are given several possibilities of reaction to the information about the assessment of the chosen Options. For the elaboration of these possibilities, cf. [Araszkiewicz, Łopatkiewicz, Zienkiewicz 2013a and 2013b].

As the Option Deliberation is concluded, the parties may further bargain on the Options either through a communication device or by means of automated bargaining algorithm. Due to the fact that these issues are not directly related to the CBR knowledge representation structures, we do not comment on them here. For an elaboration, cf. [Araszkiewicz, Łopatkiewicz, Zienkiewicz 2013a]. There are important similarities between the adopted algorithm and the algorithm employed, for instance, by [Lodder and Zeleznikow 2005] or [Zeleznikow, Stranieri and Gawler 1996].

After the bargaining phase, the parents may move on to the final, acceptance phase. In this phase, the parents may ultimately accept the generated Parent Plan or choose to return to one of the earlier phases. In typical situations, the system may be helpful in generating one or a few such optimal plans, and therefore it should facilitate the process of negotiations between the parents. What is particularly important here is that the structure of the system (Dimensions representing the degree of satisfaction of the child’s well-being in certain respects) puts emphasis on realizing this value, and not on the interest of the parents, which is however taken into account in the bargaining phase.

The Parent Plan Support System is a multi-layered negotiation support system in which Case-Based Reasoning knowledge representation structures play a pivotal role. In this paper it was shown how both scalable representations of certain features of cases (dimensions) are used together with (binary) factors to achieve an aim of high degree of context-sensitivity in a complex and ill-defined [Lynch et al. 2009] domain. The next step in the PPSS project involves PROLOG implementation of defeasible rules engine of the system.

Aleven, Vincent, Teaching case-based argumentation through a model and examples. (Unpublished doctoral dissertation). University of Pittsburgh Graduate Program in Intelligent Systems (1997).

Alexy, Robert, On Balancing and Subsumption. A Structural Comparison, Ratio Juris Vol. 16, pp. 433–449 (2003).

Ashley, Kevin, Modeling legal argument: Reasoning with cases and hypotheticals. MIT Press, Cambridge, Mass (1990).

Araszkiewicz, Michał/Łopatkiewicz, Agata/Zienkiewicz, Adam, Factor-Based Parent Plan Support System. In: Proceedings of the 14th International Conference on Artificial Intelligence and Law (ICAIL 2013), (Francesconi, E., Verheij B. eds.), ACM, New York, pp. 171–175 (2013a).

Araszkiewicz, Michał/Łopatkiewicz, Agata/Zienkiewicz, Adam, Parent Plan Support System – Context, Functions and Knowledge Base. In: Abramowicz W. (ed.), Business Information Systems Workshops, Lecture Notes in Business Information Processing Vol. 160, Springer, pp. 160–171 (2013b).

Folberg, Jay/Golann, Dwight/Stipanowich, Thomas/Kloppenberg, Lisa, Resolving disputes: Theory, practice, and law. Aspen Publishers, New York (2005).

Lodder, Arno/Zeleznikow, John, Developing an online dispute resolution environment: Dialogue tools and negotiation support systems in a three-step model. Harvard Negotiation Law Review Vol. 10, pp. 287–337 (2005).

Lynch, Colin/Ashley, Kevin/Pinkwart, Neil/Aleven, Vincent, Concepts, structures and goals: Redefining ill-definedness. International Journal of Artificial Intelligence in Education Vol. 19, pp. 253–266 (2009).

Prakken, Henry/Sartor, Giovanni, Modelling Reasoning with Precedents in a Formal Dialogue Game. Artificial Intelligence and Law Vol. 6, pp. 231–287 (1998).

Zeleznikow, John/Stranieri, Andrew/Gawler, Mark, Project report: Split-Up: A legal expert system which determines property division after divorce. Artificial Intelligence and Law Vol. 3, pp. 267–275 (1996).

---

 

Michał Araszkiewicz

Adjunct, Jagiellonian University, Faculty of Law and Administration, Department of Legal Theory

Bracka 12, 31-007 Kraków, PL

michal.araszkiewicz@uj.edu.pl

 

Agata Łopatkiewicz

PhD researcher, Faculty of Philosophy, Institute of Education

Batorego 12, 31-135 Kraków, PL

agata.lopatkiewicz@uj.edu.pl

 

Adam Zienkiewicz

Adjunct, University of Warmia and Mazury, Department of Theory and Philosophy of Law and State

Warszawska 98, 10-702 Olsztyn, PL

adam.zienkiewicz@uwm.edu.pl

 

---