Knowledge based system

A method of constructing a ripple down rules knowledge based system for generating an output conclusion in response to an input case, said method including the steps of providing a database arranged to receive and store a plurality of conclusions and a plurality of rules, associating each new conclusion added to the database with a previously stored conclusion as a refinement of the previously stored conclusion, associating each new conclusion added to the database with a rule, providing input means for facilitating input of a case, providing output means for outputting at least one previously stored conclusion in response to the input case, arranging said system such that when multiple identical conclusions are returned by the system in response to an input case and a new conclusion associated with the input case is added as a refinement of said identical conclusions, a list of cornerstone conflict cases associated with all of said identical conclusions is generated, arranging said system to receive an input rule indicative of a difference between the input case and the cornerstone conflict cases, and associating the input rule with each of said identical conclusions.

A computer usable medium having computer readable program code embodied therein for causing a computer to operate in accordance with a ripple down rules knowledge based system for generating an output conclusion in response to an input case, said system including a database arranged to receive and store a plurality of conclusions and a plurality of rules, each new conclusion added to the database being associated with a previously stored conclusion as a refinement of the previously stored conclusion and each new conclusion added to the database being associated with a rule, input means for facilitating input of a case, and output means for outputting at least one previously stored conclusion in response to the input case, said system being arranged such that when multiple identical conclusions are returned by the system in response to an input case and a new conclusion associated with the input case is added as a refinement of said identical conclusions, a list of cornerstone conflict cases associated with all of said identical conclusions is generated, and said system being arranged to receive an input rule indicative of a difference between the input case and the cornerstone conflict cases, and to associate the input rule with each of said identical conclusions.

A ripple down rules knowledge based system for generating an output conclusion in response to an input case, said system including a database arranged to receive and store a plurality of conclusions and a plurality of rules, each new conclusion added to the database being associated with a previously stored conclusion as a refinement of the previously stored conclusion and each new conclusion added to the database being associated with a rule, input means for facilitating input of a case, and output means for outputting at least one previously stored conclusion in response to the input case, said system being arranged such that when a new conclusion identical with a previous conclusion is added to the database, simultaneous output of the new conclusion and the previous conclusion is restricted.

A method of constructing a ripple down rules knowledge based system for generating an output conclusion in response to an input case, said method including the steps of providing a database arranged to receive and store a plurality of conclusions and a plurality of rules, associating each new conclusion added to the database with a previously stored conclusion as a refinement of the previously stored conclusion, associating each new conclusion added to the database with a rule, providing input means for facilitating input of a case, providing output means for outputting at least one previously stored conclusion in response to the input case, and arranging said system such that when a new conclusion identical with a previous conclusion is added to the database, simultaneous return of the new conclusion and the previous conclusion is restricted.

A computer usable medium having computer readable program code embodied therein for causing a computer to operate in accordance with a ripple down rules knowledge based system for generating an output conclusion in response to an input case, said system including a database arranged to receive and store a plurality of conclusions and a plurality of rules, each new conclusion added to the database being associated with a previously stored conclusion as a refinement of the previously stored conclusion and each new conclusion added to the database being associated with a rule, input means for facilitating input of a case, and output means for outputting at least one previously stored conclusion in response to the input case, said system being arranged such that when a new conclusion identical with a previous conclusion is added to the database, simultaneous return of the new conclusion and the previous conclusion is restricted.

A ripple down rules knowledge based system for generating an output conclusion in response to an input case, said system including a database arranged to receive and store a plurality of conclusions and a plurality of rules, each new conclusion added to the database being associated with a previously stored conclusion as a refinement of the previously stored conclusion and each new conclusion added to the database being associated with a rule, input means for facilitating input of a case, output means for outputting at least one previously stored conclusion in response to the input case, and determining means for determining a preferred order of execution of rules of the database.

A method of constructing a ripple down rules knowledge based system for generating an output conclusion in response to an input case, said method including the steps of providing a database arranged to receive and store a plurality of conclusions and a plurality of rules, associating each new conclusion added to the database with a previously stored conclusion as a refinement of the previously stored conclusion, associating each new conclusion added to the database with a rule, providing input means for facilitating input of a case, providing output means for outputting at least one previously stored conclusion in response to the input case, and determining a preferred order of execution of rules of the database.

A computer usable medium having computer readable program code embodied therein for causing a computer to operate in accordance with a ripple down rules knowledge based system for generating an output conclusion in response to an input case, said system including a database arranged to receive and store a plurality of conclusions and a plurality of rules, each new conclusion added to the database being associated with a previously stored conclusion as a refinement of the previously stored conclusion and each new conclusion added to the database being associated with a rule, input means for facilitating input of a case, output means for outputting at least one previously stored conclusion in response to the input case, and means for determining a preferred order of execution of rules of the database.

A ripple down rules knowledge based system for carrying out an action in response to an input case, said system including a database arranged to receive and store a plurality of action instructions and a plurality of rules, each new action instruction which is added to the database being associated with a previously stored action instruction as a refinement of the previously stored action instruction and each new action instruction which is added to the database being associated with a rule, input means for facilitating input of a case, and implementation means for carrying out an action corresponding to an action instruction when said corresponding action instruction is returned by the system in response to an input case.

A method of constructing a ripple down rules knowledge based system for carrying out an action in response to an input case, said method including the steps of providing a database arranged to receive and store a plurality of action instructions and a plurality of rules, associating each new action instruction which is added to the database with a previously stored action instruction as a refinement of the previously stored action instruction, associating each new action instruction which is added to the database with a rule, providing input means for facilitating input of a case, and implementing an action corresponding to an action instruction when said corresponding action instruction is returned by the system in response to an input case.

A computer usable medium having computer readable program code embodied therein for causing a computer to operate in accordance with a ripple down rules knowledge based system for carrying out an action in response to an input case, said system including a database arranged to receive and store a plurality of action instructions and a plurality of rules, each new action instruction which is added to the database being associated with a previously stored action instruction as a refinement of the previously stored action instruction and each new action instruction which is added to the database being associated with a rule, input means for facilitating input of a case, and implementation means for carrying out an action corresponding to an action instruction when said corresponding action instruction is returned by the system in response to an input case.

A knowledge based system for generating an output conclusion in response to an input case, said system including a database arranged to receive and store a plurality of conclusions and a plurality of rules, each rule having at least one condition, each new conclusion added to the database being associated with a previously stored conclusion as a refinement of the previously stored conclusion and each new conclusion added to the database being associated with a rule, input means for facilitating input of a case, output means for outputting at least one previously stored conclusion in response to the input case, and means for facilitating direct editing by a user of conclusions, rules and conditions stored in the database, said system being arranged such that when a conclusion, rule or condition is changed by a user, a list of cornerstone conflict cases affected by the changed conclusion, rule or condition is generated, and said system being arranged to receive at least one input rule to distinguish between cornerstone cases in relation to which the change should apply and cornerstone cases in relation to which the change should not apply.

A method of constructing a knowledge based system for generating an output conclusion in response to an input case, said method including the steps of providing a database arranged to receive and store a plurality of conclusions and a plurality of rules, each rule having at least one condition, associating each new conclusion added to the database with a previously stored conclusion as a refinement of the previously stored conclusion, associating each new conclusion added to the database with a rule, providing input means for facilitating input of a case, providing output means for outputting at least one previously stored conclusion in response to the input case, facilitating direct editing by a user of conclusions, rules and conditions stored in the database, arranging said system such that when a conclusion, rule or condition is changed by a user, a list of cornerstone conflict cases affected by the changed conclusion, rule or condition is generated, and arranging said system to receive at least one input rule to distinguish between cornerstone cases in relation to which the change should apply and cornerstone cases in relation to which the change should not apply.

A computer usable medium having computer readable program code embodied therein for causing a computer to operate in accordance with a ripple down rules knowledge based system for generating an output conclusion in response to an input case, said system including a database arranged to receive and store a plurality of conclusions and a plurality of rules, each rule having at least one condition, each new conclusion added to the database being associated with a previously stored conclusion as a refinement of the previously stored conclusion and each new conclusion added to the database being associated with a rule, input means for facilitating input of a case, output means for outputting at least one previously stored conclusion in response to the input case, and means for facilitating direct editing by a user of conclusions, rules and conditions stored in the database, said system being arranged such that when a conclusion, rule or condition is changed by a user, a list of cornerstone conflict cases affected by the changed conclusion, rule or condition is generated, and said system being arranged to receive at least one input rule to distinguish between cornerstone cases in relation to which the change should apply and cornerstone cases in relation to which the change should not apply.

A ripple down rules knowledge based system for generating an output conclusion in response to an input case, said system including a database arranged to receive and store a plurality of conclusions and a plurality of rules, each new conclusion added to the database being associated with a previously stored conclusion as a refinement of the previously stored conclusion and each new conclusion added to the database being associated with a rule, input means for facilitating input of a case, output means for outputting at least one previously stored conclusion in response to an input case, means for generating a prudence profile, means for generating a prudence identifier using the prudence profile, said prudence identifier being indicative of whether said output previously stored conclusion is likely to be incorrect for the input case, and means for generating a profile maturity value indicative of the likely accuracy of the prudence identifier.

A method of predicting the accuracy of a ripple down rules knowledge based system for generating an output conclusion in response to an input case, said system including a database arranged to receive and store a plurality of conclusions and a plurality of rules, each new conclusion added to the database being associated with a previously stored conclusion as a refinement of the previously stored conclusion and each new conclusion added to the database being associated with a rule, input means for facilitating input of a case, and output means for outputting at least one previously stored conclusion in response to an input case, said method including the steps of generating a prudence profile, generating a prudence identifier using the prudence profile, said prudence identifier being indicative of whether said output previously stored conclusion is likely to be incorrect for the input case, and generating a profile maturity value indicative of the likely accuracy of the prudence identifier.

A computer usable medium having computer readable program code embodied therein for causing a computer to operate in accordance with a ripple down rules knowledge based system for generating an output conclusion in response to an input case, said system including a database arranged to receive and store a plurality of conclusions and a plurality of rules, each new conclusion added to the database being associated with a previously stored conclusion as a refinement of the previously stored conclusion and each new conclusion added to the database being associated with a rule, input means for facilitating input of a case, output means for outputting at least one previously stored conclusion in response to an input case, means for generating a prudence profile, means for generating a prudence identifier using the prudence profile, said prudence identifier being indicative of whether said output previously stored conclusion is likely to be incorrect for the input case, and means for generating a profile maturity value indicative of the likely accuracy of the prudence identifier.

A ripple down rules knowledge based system for generating an output conclusion in response to an input case, said system including a database arranged to receive and store a plurality of conclusions and a plurality of rules, each new conclusion added to the database being associated with a previously stored conclusion as a refinement of the previously stored conclusion and each new conclusion added to the database being associated with a rule, input means for facilitating input of a case, and output means for outputting at least one previously stored conclusion in response to the input case, said system being arranged such that when a conclusion is output by the system in response to an input case and a new conclusion associated with the input case is added as a refinement of the output conclusion, the system determines whether a cornerstone case already exists which would cause output of the new conclusion and associates said determined cornerstone case with the new conclusion.

A method of constructing a ripple down rules knowledge based system for generating an output conclusion in response to an input case, said method including the steps of providing a database arranged to receive and store a plurality of conclusions and a plurality of rules, associating each new conclusion added to the database with a previously stored conclusion as a refinement of the previously stored conclusion, associating each new conclusion added to the database with a rule, providing input means for facilitating input of a case, providing output means for outputting at least one previously stored conclusion in response to the input case, and arranging said system such that when a conclusion is output by the system in response to an input case and a new conclusion associated with the input case is added as a refinement of the output conclusion, the system determines whether a cornerstone case already exists which would cause output of the new conclusion and associates said determined cornerstone case with the new conclusion.

A computer usable medium having computer readable program code embodied therein for causing a computer to operate in accordance with a ripple down rules knowledge based system for generating an output conclusion in response to an input case, said system including a database arranged to receive and store a plurality of conclusions and a plurality of rules, each new conclusion added to the database being associated with a previously stored conclusion as a refinement of the previously stored conclusion and each new conclusion added to the database being associated with a rule, input means for facilitating input of a case, and output means for outputting at least one previously stored conclusion in response to the input case, said system being arranged such that when a conclusion is output by the system in response to an input case and a new conclusion associated with the input case is added as a refinement of the output conclusion, the system determines whether a cornerstone case already exists which would cause output of the new conclusion and associates said determined cornerstone case with the new conclusion.

A ripple down rules knowledge based system for generating an output conclusion in response to an input case, said system including a database arranged to receive and store a plurality of conclusions and a plurality of rules, each new conclusion added to the database being associated with a previously stored conclusion as a refinement of the previously stored conclusion and each new conclusion added to the database being associated with a rule, input means for facilitating input of a case, output means for outputting at least one previously stored conclusion in response to the input case, generating means for generating a statistical value indicative of conclusions previously output by said system, and selecting means for selecting a percentage value indicative of the proportion of subsequent input cases to be presented for verification.

A method of constructing a ripple down rules knowledge based system for generating an output conclusion in response to an input case, said method including the steps of providing a database arranged to receive and store a plurality of conclusions and a plurality of rules, associating each new conclusion added to the database with a previously stored conclusion as a refinement of the previously stored conclusion, associating each new conclusion added to the database with a rule, providing input means for facilitating input of a case, providing output means for outputting at least one previously stored conclusion in response to the input case, generating a statistical value indicative of conclusions previously output by said system, and selecting a percentage value indicative of the proportion of subsequent input cases to be presented for verification.

A computer usable medium having computer readable program code embodied therein for causing a computer to operate in accordance with a ripple down rules knowledge based system for generating an output conclusion in response to an input case, said system including a database arranged to receive and store a plurality of conclusions and a plurality of rules, each new conclusion added to the database being associated with a previously stored conclusion as a refinement of the previously stored conclusion and each new conclusion added to the database being associated with a rule, input means for facilitating input of a case, output means for outputting at least one previously stored conclusion in response to the input case, generating means for generating a statistical value indicative of conclusions previously output by said system, and selecting means for selecting a percentage value indicative of the proportion of subsequent input cases to be presented for verification.