This paper is an excellent in-depth survey of the most important developments in semantic network research. After a brief introduction that explains the importance of such research to the area of natural language understanding, Mac Randal describes the main developments in the field from 1966 to 1986. He also provides lengthy and detailed descriptions of some of the principal semantic network systems, such as Brachman’s KL-ONE.
The presentation is always clear and focused and has enough detail to make it interesting even for people already knowledgeable in the subject. This paper could serve as an introduction to the main concepts of semantic network research. Both students and practitioners should find it enjoyable and quite useful. The length of the material is appropriate, and the level of discussion is right on target and never gets lost in technical details.
Overall, I found this paper well designed and quite interesting. The presentation is clear and logical, and the emphasis is on the truly influential developments in the area. The limitations of the formalism are also clearly explained. I would not hesitate to suggest this paper to anyone interested in learning about the main features of semantic networks and obtaining a clear, historical perspective on the field.