Reasoning system, reasoning method, and recording medium

2015 
Provided is a reasoning system whereby it is possible to perform reasoning even if there is lack of knowledge. In this reasoning system (100), an input unit (110) receives a start state and an end state. A rule candidate generation unit (120) identifies both a first state, which can be found by tracing one or more known rules from the start state in the forward direction, and a second state, which can be found by tracing one or more known rules from the end state in the reverse direction. The rule candidate generation unit (120) then generates one or more rule candidates associated between the first state and the second state, or generates one or more rule candidates associated with the first state and one or more rule candidates associated with the second state. A rule selection unit (130) then selects some of these rule candidates as new rules, on the basis of the feasibility of each rule candidate as calculated on the basis of known rules. A derivation unit (140) then performs a derivation process for deriving the end state from the start state on the basis of known rules and the new rules.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []