Applications of artificial intelligence for chemical inference. IX. Analysis of mixtures without prior separation as illustrated for estrogens

DENDRAL: A case study of the first expert system for scientific hypothesis formation

“…If chemists have experience with this class, they may know enough about the fragmentations that are most likely to have occurred to allow them to predict efficiently the behavior of an arbitrary compound of that class. This class-specific information--even in its predictive form--can be used to produce constraints that will reduce the DENDRAL generator's output, as was done for a few subclasses of steroids [49][50][51].…”

“…The processing speed and clerical superiority of the computer in fact amplify, in many cases, the power of the scientist's informal knowledge. A particular case in point is the PLANNER's analysis of spectra taken from unseparated mixtures of estrogens, described in [49]. The record-keeping and cross-checking demands of this task are too great to permit a thorough job by hand, yet the program is able to identify correctly the number and structure of compounds in mixtures.…”

Comparison of various K-nearest neighbor voting schemes with the self-training interpretive and retrieval system for identifying molecular substructures from mass spectral data

“…The popularity of combined gas chromatography-mass spectrometry has greatly increased the need for computeraided interpretation of mass spectral data (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14). A variety of techniques, including pattern recognition and artificial intelligence methods, have been proposed, but little has been done to assess their relative merits on a quantitative basis.…”

Recent Research in Artificial Intelligence, Heuristic Programming, and Network Protocols

“…(1) A general proof program that has been developed for lesearch in different areas of mathematical problems This is based on the Resolution principle and rules for equality program below (2) A fast special purpose prover (called the " capability aimed at eliminating the "easy work" and this has turned out in experiments to be a powerful component of the verification system (see below)A usei's manual for the general theorem provei is available [i], and publications ^. 'M.fO deal with interactive applications of this program to mathematics and mfoimation retnrvalRemnt expenmrnts in using the prover to obtain new characterizations of varieties of Croups and to check tedious proofs in Euclidean Geometry are given in[6]…”