| Trade-off
Manipulations in the Development of Emilia Bellucci and John Zeleznikow
Introduction Negotiation is a process by which two or more parties conduct communications or conferences with the view to resolving differences between them. Negotiation Support Systems (NSS) are programs that assist users in the negotiation process. In comparison, Negotiation Decision Support Systems (NDSS) extend the operation of NSS to include an element of decision support. In this paper, we present a system that we have implemented, Family_Winner, as a NDSS that uses trade-off manipulations to propose settlements to disputes. We present a survey of existing NSS and NDSS. In the latter category, we did not find any systems that use trade-off manipulations to settle disputes, even though our research suggests that the use of trade-offs in negotiation is widespread . We discuss Family_Winner (Bellucci, 2004) a Negotiation Decision Support System that uses Trade-off Maps (a variant of Constraint Diagrams) to represent trade-off opportunities inherent in the issues of a dispute. The system acts upon trade-offs once an issue has been allocated, resulting in compensation and rewards to the utilities of issues remaining in dispute. The amount by which a party is compensated is decided through a complex set of formulae that have been derived empirically. The Issue Decomposition Hierarchy imbedded in the system allows for the incorporation of sub-issues, which forms our attempt to increase the number of issues in dispute. Family_Winner operates on a two-party basis. Although Family_Winner was initially developed in the domain of Australian Family Law, we argue Family_Winner is not domain dependent, and is flexible with regard to the type of issues it is able to process. The system was evaluated against negotiation case studies from various domains. We conclude the paper by mentioning our future directions in the development of On-line dispute resolution systems and other projects involving the extension of Family_Winner as form of On-line dispute resolution. Negotiation theory Numerous models have been developed from detailed studies of the way people negotiate. There is a significant difference between the formal models of negotiation and practical approaches to developing negotiation strategies. Formal models have been derived from Game Theory, Multi-Criteria Decision Making, negotiation analysis, and economics. Negotiation approaches, for example Positional Bargaining, collaboration and Principled Negotiation, arrive from behavioural research. The difference between the two is that the former use all or selected assumptions of economic rationality while the latter are based on individual and social norms and behaviours, whether they be actual or postulated. Our research has focused on behavioural negotiation approaches, and in particular we isolate Principled Negotiation as the theory most suited to our requirements. (Fisher et al., 1994) has conceptualised value-based negotiation into Principled Negotiation, developed under the Harvard Negotiation Project. Its' emphasis is that parties look for mutual gains, wherever possible, and when interests conflict, parties should come to a ruling that is independent of the beliefs of either side. The essential features of Principled Negotiation as a problem-solving task are as follows: Separate the people from the problem. This is to ensure that persons with stronger personalities cannot influence others into a decision that is biased towards a party or group of parties. Focus on interests, not on positions. Participants must distinguish and make known their underlying values in order to justify their position. In most negotiations, each party will have interests they would like satisfied by settlement, and it is important these be understood as separate from their positions. Invent options for mutual gain. There are a number of strategies that enable option generation. Expanding the pie is one strategy, where new issues are added to the dispute in an attempt to locate new resources. Another is compensation, in which payment is made through another issue or entirely new case and log-rolling, where disputants form agreement by taking into account the differences between multiple issues. In providing decision support, Family_Winner uses a strategy integrating the principles of logrolling and compensation to approach settlement. Expanding the pie is interpreted in Family_Winner as increasing the number of issues in dispute - which is enabled by disputants decomposing issues into sub-issues and storing these into the Issue Decomposition Hierarchy. Insist on objective criteria. Some negotiations are not susceptible to a win-win situation. The most obvious of these is haggling over the price of an item: since the more money one side negotiates, the less their opponent receives. In similar cases, unbiased independent evaluations of an item may provide guidance in setting a mutually agreeable settlement. 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. When a person is wishing to buy a used car, they will usually refer to a commonly accepted set of approximate automotive prices. Using this initial figure and considering other variables such as new components, the distance travelled by the car and its current condition, the negotiator then decides the value they wish to place on a car. BATNAs in negotiations are therefore generally used to form a basis on which fair agreements can be argued. Family_Winner uses Principled Negotiation as its' foundation negotiation theory. In providing decision support, it uses a trade-off and compensation strategy in which to invent options for mutual gain. An analysis of current NSS and NDSS The majority of traditional NSS have been restricted to informing parties of past and present preferences and on the progress made within the negotiation. We have classed these as template-based NSS. Examples of NSS include Negotiator Pro, The Art Of Negotiating (Eidelman, 1993) and DEUS (Zeleznikow et al., 1995). Systems implementing on-line negotiation exhibiting the properties of template systems include SmartSettle (www.smartsettle.com), INSPIRE (Kersten, 1997) and CBSS (Yuan et al., 1998). INSPIRE (Kersten, 1997) used utility functions to graph offers; while in DEUS (Zeleznikow et al., 1995) the goals and beliefs of parties were set on screen side by side. SmartSettle performs similar analysis to that performed by INSPIRE, by assessing the viability of packages through satisfaction graphs. The Art of Negotiating asked questions from disputants to help in assessing the extent of agreement and differences. Similar to INSPIRE, SmartSettle, CBSS, and DEUS; Negotiator Pro and The Art of Negotiating are programs that assist in the preparation and planning stages of negotiation. The aim of this paper is to demonstrate use of decision making support in negotiation, in particular through our system, Family_Winner. NSS that extend the primary role of template-based systems to incorporate a decision-making aspect are classified as Negotiation Decision Support Systems (NDSS). NDSS are equipped with the tools to propose sample solutions based on the inputs representative of disputant preferences. Early decision-support negotiation systems primarily used Artificial Intelligence techniques to model negotiation. LDS (Peterson and Waterman, 1985) used rule-based reasoning to assist legal experts in settling product liability cases. SAL (Waterman et al., 1986) also used rule-based reasoning to help insurance claim adjusters evaluate claims related to asbestos exposure. The SAL and LDS systems were important as they represent the first steps in recognising the value of settlement-oriented decision support systems. NEGOPLAN (Matwin et al., 1989) is a rule based system written in PROLOG which advised upon industrial disputes in the Canadian paper industry. Mediator (Kolodner and Simpson, 1989) used case retrieval and adaptation to propose solutions to international disputes, while PERSUADER (Sycara, 1993) integrated case based reasoning and decision-theoretic techniques to provide decision support to United States' industrial disputes. Family_Negotiator (Bellucci and Zeleznikow, 1997), a hybrid rule-based and case-based system implemented in Family Law, provides disputants with suggestive advice as how best to resolve the issues in dispute. Whilst evaluating the Family_Negotiator system, we discovered that Family Law negotiation was not an appropriate domain in which to apply either Case-based or Rule-based Reasoning, due principally to the open textured nature , of the domain. The overall framework of Family_Negotiator did not provide in-depth solutions expected from real-life negotiations. AdjustWinner (Bellucci and Zeleznikow, 1998), uses a utility function to achieve equal distribution of the common pool. The algorithm used in the system was the Adjusted Winner procedure (Brams and Taylor, 1996). AdjustWinner resolves a dispute by dividing issues and items among disputants, through a mathematical manipulation of numeric preferences. Mediator, Persuader, NEGOPLAN and Family_Negotiator are considered to be intelligent systems since they can generate solutions using the system's internal knowledge as well as users input. All incorporate some level of negotiation support, together with the ability to provide users with a resolution to the current problem. Artificial Intelligence techniques such as case-based, rule-based and hybrid reasoning have had mixed degrees of success. The Mediator proved quite successful in its retrieval and adaptation of previous cases. NEGOPLAN used rule-based reasoning to model its domain successfully, while Persuader successfully modelled its domain using a hybrid case and rule-based methodology. Family_Negotiator however, did not perform to its initial expectations, primarily due to its relatively simple modelling of the domain. Negotiation decision support through the use of Trade-Off manipulations Decision-making is a knowledge-intensive activity that alters an organisation's state of knowledge. A decision is defined "a piece of knowledge indicating a commitment to some course of action" (Holsapple and Whinston, 1996). The decision support process not only introduces a new piece of knowledge (the decision), but the process itself may result in the addition of new knowledge. Decision support in negotiation involves a number of complex variables, which include the number of issues, the number of parties to the dispute, and to some extent, the complexities inherent in the domain. Family_Winner's method of decision support involves a complex number of techniques, including the incorporation of an Issue Decomposition Hierarchy, a Compensation and Trade-Off strategy and an Allocation strategy. The trade-offs pertaining to a disputant are graphically displayed through a series of trade-off maps, while an Issue Decomposition Hierarchy enables disputants to decompose issues to any required level of specification. Although it may not appear intuitive, the number of issues involved will influence the success of the negotiation, as it is assumed, based on observations and results from data analysis, that the greater the number of issues, the greater the scope and opportunity for a mutual agreement. Principled Negotiation advocates use of 'Expanding the pie' as a method of option generation. Family_Winner will usually present for discussion the least contentious issue first. Increasing the number of issues in dispute and the implementation of a method to order the presentation of issues enables the fostering of a positive environment; one in which agreement on any issue will lead to increased prospects of a successful negotiation. Compensation is considered as an external reward, one that is not related to the issues on the table. For the purposes of this study, we consider a trade-off as a strategy combining log-rolling and compensation strategies. (Pruitt, 1981) believes that log-rolling, where participants look collectively at multiple issues, can only occur if the dispute revolves around the resolution of many issues. When a conflict consists of several issues in which some issues are considered more important to party A than to party B, and other issues are considered more important to party B, then log-rolling is appropriate. For log-rolling to be successful, we need to pay careful attention to the importance value of issues, as it is expected that in a negotiation, each side will concede issues to which they give low values. The issues involved in trade-offs are determined by the principles of log-rolling, essentially those issues valued highly by party A are allocated to party A, providing there are no issues that both parties value equally. The amount of any compensation resulting from the triggering a trade-off has been empirically determined from an analysis of data. A trade-off is formed after a comparison between the ratings of two issues has been conducted. The value of a trade-off relationship is determined by analysing the differences between the parties (Mnookin et al., 2000). We implemented a trade-off strategy based on extracting the differences between the ratings of opposing parties. We hence present a theory of decision support encompassing these theories, implemented in the Family_Winner system. Family_Winner This section outlines the major components of Family_Winner through a comprehensive flow chart, displayed in Figure 1. The input data consists of several variables (including issue names and associated ratings), which all directly contribute to the outcome of the current case. The system uses the Issue Decomposition Hierarchy in which to store all issues (and sub-issues) and makes ample use of Trade-off Maps to mimic a compensation strategy. The output consists of a list of allocations, which forms the basis of the advice provided by the system. The flowchart in Figure 1 identifies the sequence of actions, decisions and branching points in the negotiation process implemented in Family_Winner. The system accepts input from both parties involved in the dispute. This data is then analysed and transformed into information required by the functions inherent in the system. The first major process is that of forming and displaying Trade-off Maps. These diagrams are indicative of possible trade-offs between pairs of issues. Two maps are drawn side by side, each one representing a party's view of the negotiation. They consist of a series of circles (indicating issues) and lines connecting two issues together, (indicating a trade-off relationship). Trade-off relationships translate to a trade-off opportunity. Issues are labelled by their name and current rating. The value of an issue can be that directly entered by the party, or a rating modified as a result of a previous allocation. The trade-off relationships between pairs of issues are labelled by the numerical difference between the two ratings. This newly devised numeral is used in calculations to determine appropriate compensation awarded to the parties after the allocation of an issue. As the program progresses, the parties are asked to build on an Issue Decomposition Hierarchy by decomposing issues, which allows for the current pool of issues to be expanded. The parties are asked if the issues should be sub-divided into smaller issues. If the disputants answer yes, then the system suggests the first issue to be decomposed. This recommendation has been based on the understanding that a large difference between the ratings of parties is indicative of an issue that is most likely to be resolved quickly . Once the issue to be decomposed has been decided upon, parties are required to enter new sub-issues in the same format as parent issues were entered. When this task is completed, the new details are stored in the Issue Decomposition Hierarchy under the appropriate primary (parent) issue. On the flowchart, development of the hierarchy is shown by the line labelled level + 1 issues to be input. If an issue does not require decomposition or has been sub-divided appropriately, the issue is allocated according to the issue's importance rating. The ratings of issues are hence compared. The party that values the issue more highly is most likely to receive the issue. After an allocation, the ratings of the remaining issues may be modified through compensation, to influence future issue allocation. The amount of compensation awarded is calculated through graphs that were derived from data obtained from domain experts, and is explained in full in (Bellucci, 2004). Once Family_Winner allocates an issue, a summary outlining the allocation is presented to users. Information presented at this time includes the allocated issue and the party to whom it is allocated, previously allocated issues (and the parties they have been allocated to), and the value of rating changes made to the subsequent issues. This information enables users to gain an insight into the reasoning behind the allocation and the relative impact of the allocation on the remaining issues in dispute. The process of allocation and issue decomposition continues until there are no more issues to allocate, at which point the program ceases execution. Family_Winner in Operation We now detail a hypothetical family law trial case using Family_Winner to provide negotiation support. This exercise demonstrates the system's operation in practice. A hypothetical Family Law case Suppose Cassandra (Wife) and Paul (Husband) Jones have been married for fifteen years and have two sons aged thirteen and eleven. Cassandra wants a divorce and an immediate property settlement. She also believes that although she received income from employment throughout her marriage, her principal role was as a homemaker and a nurturer. Both agree to the distribution of the joint marital property consisting of a house, his Mitsubishi car, and her Holden car. In addition, she believes she is entitled to a portion of her Husband's share portfolio and of his superannuation entitlements. She wishes to retain the house and the Holden car, while Paul wishes to retain his Mitsubishi car and agrees on an equal share of the share portfolio and his superannuation entitlements. Cassandra believes she should receive primary custody of the children. She consults a lawyer who advises her that as the parent with current primary custody of the children, she should seek 60% of the marital property and adequate child allowance. The 60% mainly consists of the matrimonial home and the holiday house. She wishes to retain both of these properties. The above case will be used to highlight several important theories used by Family_Winner in determining negotiation advice about this case. These include the areas of input, the Issue Decomposition Hierarchy's development, the display of Trade-off Maps, the allocation of issues and their effect on issue ratings and Trade-off Maps. Input to Family_Winner system In Family_Winner, the disputants enter information regarding the dispute at hand. Disputants enter the issues in no particular order. Since issues will be stored in a hierarchy, it is important that issues on the same level of decomposition should be entered at the same time. Equally as important are indications of the importance of an issue to each party, represented in the form of a numerical rating between 0 and 100 inclusive. The case is presented to Family_Winner, using the following data as input.
Table 1. Initial input of Issues and ratings for use in the hypothetical Family Law Negotiation. This information is then analysed by a number of functions. These functions include the translation of data into Trade-off Maps, the relaying of information to the database, forming issue allocations and modifying the ratings of the issues in the negotiation to reflect allocations. Trade-off Maps Once the user has entered the data appropriately, the next screen displays Trade-off Maps generated by the system. The elements of a Trade-off Map are: (i) The nodes (or issues in this case); (ii) The strength of connections between these nodes (reflective of the trade-off opportunities); and (iii) A rating figure for each issue. The issues and their ratings are retrieved directly from user input. Figures 2 and 3 are the Trade-off Maps displayed to disputants following the input of issues listed in Table 1.
Figure 2. The Husband's Trade-off Map after the initial input of the primary issues.
Figure 3. The Wife's Trade-off Map after the initial input of the primary issues. Formation of the Issue Decomposition Hierarchy The disputants are asked to decompose an issue into many smaller sub-issues. Sub-issues are then incorporated into the dispute through the formation of an Issue Decomposition Hierarchy. Child-related Issues is the first issue to be considered for decomposition or allocation. Table 2 lists the point allocations (ratings) given to each issue by the Husband and the Wife, and the ratings used in the dispute (p-ratings), which represent the influence of Child-Related Issues on the sub-issue's initial point allocation. P-ratings are calculated as a ratio of the parent issue's rating. For instance, Party A gives issue1 a rating of 60, and issue2 a rating of 40. Issue11 has a p-rating of 10 (10% of 60) = 6, and Issue12 a p-rating of 90 (90% of 60) = 54.
Table:2. Ratings and p-ratings for the sub-issues of Child-Related Issues. The Trade-off Map is now altered to include the sub-issues of the primary issues. The modified Trade-off Maps of both parties are detailed in Figures 4 and 5.
Figure 4. The Husband's Trade-off Map incorporating the sub-issues of Child-related Issues.
Figure 5. The Wife's Trade off Map incorporating the sub-issues of Child-Related Issues. Family_Winner allocates a parent issue through the allocation of its sub-issues. Therefore, in this example, one of the issues listed in Table 2 will be allocated next. All the sub-issues of Child-related Issues will be allocated before the negotiation moves to consider other issues. Commencing the allocation of issues The system allocates an issue to one of the parties. The party whose rating is greatest for the issue is allocated the issue. If the issue is valued equally (by the disputants), then the next issue to be allocated replaces the issue in question. The current rating of issues connecting to the issue allocated is revised, based on mathematical functions derived empirically from data used in our study. (Bellucci, 2004) details the source of this data and subsequent functions used in Family_Winner. The allocation of an issue involves removal of the issue from the Trade-off Maps, and making appropriate changes to the ratings of affected issues. The first issue in this example to be allocated is Visitation Rights. It is awarded to the Husband, as his rating of 35 is greater than the Wife's equivalent of 5. As a result of the Husband's allocation, the ratings of remaining issues are changed. The following table lists all existing issues, their updated ratings and the percentage change resulting from the allocation of Visitation Rights to the Husband.
Table 3. Changes made to the ratings of issues following the allocation of Visitation Rights to the Husband. As a result of the Husband's allocation of an issue he considered important (valued at 35 points), his ratings did not change considerably. The Wife was duly compensated for her loss of Visitation Rights, valued relatively unimportant at 5 points. The relative Trade-off Maps of each party, shown in Figures 6 and 7, can be interpreted to explain the amount of change each rating experienced as a result of the allocation. The Husband's ratings experienced little change as the issue's rating was considered by the system to be of great importance to the Husband. All ratings experienced an increase of 5%, as the relationship figures between the issues and Visitation Rights were all similar in number. Their relationship figures were 17 between Child Support, 17 between Residency, 25 between Monetary Issues and 15 between Property. The Wife was compensated for her loss of Visitation Rights (valued at 5 points) through those issues whose relationship with Visitation Rights is of relatively greater significance. The trade-offs between Visitation Rights and Monetary Issues, and Visitation Rights and Residency held relationship values of 30 and 25 respectively. These issues were the only ones whose ratings increased, with increases of 50% and 37.5% respectively. Property Issues and Residency did not change their ratings, as their relationships with Visitation Rights were valued at 10 points each. Changes to Trade-off Maps as a result of the allocation of issues. Trade-off maps display the trade-offs currently applicable to the dispute. Once an issue is removed from a dispute through allocation, the Trade-off Map is modified to reflect this change. The issue is removed from the map, and the ratings of the remaining issues are re-calculated according to the values dictated by the applicable trade-off relationships. The resulting Trade-off Maps following the allocation of Visitation Rights are demonstrated in Figures 6 and 7.
Figure 6. Husband's Trade off Map after the allocation of Visitation Rights.
Figure 7. Wife's Trade-off Map after the allocation of Visitation Rights. The system continues to traverse the hierarchy, by either allocating or decomposing issues, until all issues have been allocated. A summary of subsequent allocations is found in table 4.
Table 4. Allocation table for the hypothetical Family Law Dispute. Family_Winner was evaluated using the Context, Criteria, Contingency Evaluation Framework (Hall et al., 2003) for evaluating legal knowledge-based systems. Although the strategy has components specifically developed for the requirements of legal systems, we were able to develop an evaluative strategy suited to Family_Winner's requirements. Family_Winner is a negotiation decision support system that was initially built to resolve disputes in Australian Family Law. From the system's evaluation, we concluded its' use is of greater significance in domains other than Family Law. In (Bellucci 2004) we discuss how Family_Winner has been used in a variety of negotiation domains; for example in Family Law, Enterprise Bargaining Agreements, International disputation and company negotiations. An investigation of these examples (Zeleznikow and Bellucci 2003) 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 points can be allocated to issues. Conclusions Our survey of existing Negotiation Support Systems isolated two major streams of negotiation support: template-based systems and decision support systems. The major role of a template system is to provide tools and graphical aids for keeping parties informed on past preferences and other issues concerning progress made in a negotiation. Whilst most template systems successfully support the negotiation, they assume the negotiation continues passively after the initial intake of preferences and issues. Negotiated Decision Support Systems attempt to model the negotiation dynamically, through the incorporation of decision support. We have presented Family_Winner as a Negotiated Decision Support System that provides a step-by-step approach to a negotiated settlement, based on a series of trade-offs and compensation to provide decision support. In addition, the system utilises the principles of Principled Negotiation to model the negotiation process. SmartSettle (Thiessen and McMahon 2000) assists parties to overcome the challenges of conventional negotiation through a range of analytical tools to clarify interests, identify tradeoffs, recognise party satisfaction and generate optimal solutions. The aim is to better prepare parties for negotiation or to support them during the negotiation process. We are working at incorporating SmartSettle strategies into the Family_Winner system. Our evaluation of the system resulted in positive feedback regarding its use in domains other than family law. We have obtained a grant to extend the applicability of the Family_Winner system by developing an on-line version. Bibliography Bellucci, E. 2004. Developing Compensation Strategies for the construction of Negotiation Decision Support Systems. PHD thesis, La Trobe University, Bundoora 3086, Victoria, Australia Bellucci, E. and Zeleznikow, J. 1997. Family-Negotiator: an intelligent decision support system for negotiation in Australian Family Law. Proceedings of the Fourth Conference of the International Society for Decision Support Systems, Lausanne, International Society for Decision Support Systems: 359-373. Bellucci, E. and Zeleznikow, J. 1998. A comparative study of negotiation decision support systems. Proceedings of the Thirty-First Hawaii International Conference on System Sciences. IEEE Computer Society, Los Alamitos, Cal: 254-262. Brams, S. J. and Taylor A. D., 1996. Fair Division, From cake cutting to dispute resolution. Cambridge University Press, Cambridge, UK. Eidelman, J.A. 1993. Software for Negotiations, Law Practice Management, 19(7): 50-55. Fisher, R., Ury, W. and Patton, B. 1994. Getting to Yes, Century Business, Sydney. Hall, M. J. J., Hall, R. and Zeleznikow, J. 2003. A method for evaluating legal knowledge-based systems based upon the Context Criteria Contingency-guidelines framework. Proceedings of the Ninth International Conference on Artificial Intelligence and Law, Edinburgh, Scotland: ACM Press, 274-283. Holsapple, C. W. and Whinston, A. B. 1996. Decision Support Systems - a Knowledge Based Approach, West Publishing Company, St. Paul, Minnesota. Kersten, G. E. 1997. Support for Group Decisions and Negotiations, in: J. Climaco (Ed.) An Overview, in Multiple Criteria Decision Making and Support, (Heidelberg: Springer Verlag). Kolodner, J.L. and Simpson, R. L., 1989. The Mediator: Analysis of an Early Case-Based Problem Solver, Cognitive Science, 13:507-549. Matwin, S., Szpakowicz, S., Koperczak, Z., Kersten, G. E. and Michalowski, G. 1989. NEGOPLAN: An Expert System Shell for Negotiation Support, IEEE Expert 4:50-62 Mnookin, R., Peppet, S. R. and Tulumello, A. S. 2000. Beyond Winning: Negotiating to Create Value in Deals and Disputes, The Belnap Press of Harvard University Press. Peterson, M. and Waterman, D. 1985. Evaluating Civil Claims: An Expert Systems Approach to Evaluating Product Liability Cases. In Computer Power and Legal Reasoning (Ed, Walter, C.) West Publishing Company, pp. 627 - 659. Pruitt, D. G. 1981. Negotiation Behaviour, Academic Press Inc. Sycara, K. 1993. Machine Learning for Intelligent Support of Conflict Resolution, Decision Support Systems, 10:121-136. Thiessen, E. M. and McMahon, J. P. 2000. Beyond Win-Win in Cyberspace. Ohio State Journal on Dispute Resolution, 15: 643. Waterman, D. A., Paul, J. and Peterson, M., 1986. Expert systems for legal decision making, Expert Systems 3 (4): 212-226. Yuan, Y., Rose, B. J., Archer, N. and Suarga, H. (1998) A Web-Based Negotiation Support System, EM - Electronic Markets, 8. Zeleznikow, J. and Bellucci, E. 2003. Family_Winner: integrating game theory and heuristics to provide negotiation support. Proceedings of Sixteenth International Conference on Legal Knowledge Based System. IOS Publications, Amsterdam, Netherlands: 21-30. Zeleznikow, J., Meersman, R., Hunter, D. and van Helvoort, E. 1995. Computer tools for aiding legal negotiation. ACIS95 - Sixth Australasian Conference on Information Systems, Curtin University of Technology, Perth, Western Australia: 231-251 |
