A neural network approach to inference mechanism for logic programming language

Presents an inference mechanism for logic programming languages using neural networks that is flexible and suited for fine-grain parallel computing. The authors approach is radically different from the conventional methods based on refutation processes. Programs written in the logic programming language are transformed into a Hopfield-type neural network and relaxation techniques are applied to this network to inference solutions. The authors propose an algorithm to transform logic programs into Hopfield-type neural networks and implement a prototype of the inference system based on this mechanism. The authors tested the system with some preliminary problems. Preliminary results confirm that the algorithm is correct.