Semantic Information Processing

Rating:
Summary: Still the best introduction to the research in pure AI
Review: This book is a collection of PhD thesis from several fameous Artificial Intelligence researchers with couple of extra chapters by Minsky.

The chapters are: Introduction (by Minsky), SIR: A Computer Program for Semantic Information Retrieval (by Berthrand Raphael), Natural Language Input for a Computer Problem-Solving System (by Daniel Bobrow), Semantic Memory (by Ross Quillian), A Program for Solution of Geometry-Analogy Intelligence Test Question (by Thomas Evans), A Deductive Question-Answering System (by Fischer Black), Programs with Common Sense (by John McCarthy), Descriptive Languages and Problem Solving (by Minsky), and Matter, Mind, and Models (by Minsky).

This book might seem a bit old since it was originally published in 1968 (the copy I have is third printing from 1982) but it's still the best introduction to the approach that some of Artificial Intelligence pioneers had while working on some of the very first issues in the field.

This book gives clear guidances how one should design and write a thesis from Artificial Intelligence. The authors show where to draw the line between the theory and the practice and how to articulate research issues and present experimental results in such a fuzzy area.

I can only recommend obtaining a used copy of this wonderful book.