Issue No. 03 - March (1996 vol. 29)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/2.485894
<p>The Neural Logic Network, or Neulonet, is an artificial neural network that represents a variety of logical operations. Neulonet emulates a range of human decision-making behaviors by combining aspects of rule-based expert systems and neural networks. Neulonet demonstrates intuitive and biased decision logic, as well as conventional logic. In addition, it is amenable to training and parallelism. The authors used Neulonet to build a shell that appears as a rule-based expert system capable of doing logical inferencing, but its underlying architecture is connectionist. With the shell, they developed two financial advisory systems that performed better in tests than pure rule-based and pure neural net systems. </p>
C. Lim Tan, H. Heng Teh and T. Seng Quah, "An Artificial Neural Network that Models Human Decision Making," in Computer, vol. 29, no. , pp. 64-70, 1996.