Uniterm indexing principles, problems and solutions

IN INFORMATION systems built on the fundamental principles of Uniterm indexing, operational difficulties can and do develop. Some Uniterm installations were abandoned after development to no more that laboratory scale because the individuals responsible for system operation neglected to, or elected not to develop the simple refinements necessary to solve these problems in their own local situations. Others which could visualize the benefits to be derived from the great depth of indexing possible with Uniterms, face the problems and developed simple but effective solutions. Smoothly functioning Uniterm systems many locations have developed: simplified posting procedures; vocabulary treatment and controls (such as for synonyms and homographs); syntactical control techniques; controls for providing automatic indexing or search on generic levels when indexing provided only narrow specificity; and thesauri word‐association matrices to provide to indexers and searchers suggested additional Uniterms for expressing concepts from other viewpoints. In many organizations, solutions to problems inherent indexing via the Uniterm approach have demonstrated that effective, rapid, low‐cost output is an assures end result of multi‐dimensional analysis and description at input, coupled with selective or parallel storage and search techniques.