IRE Index index Butterworth & Company Karen Sparck Jones All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, including photocopying and recording, without the written permission of the copyright holder, application for which should be addressed to the Publishers. Such written permission must also be obtained before any part of this publication is stored in a retrieval system of any nature. Office for Scientific and Technical Informa- tion, 285 Off-shelf test, 137, 142, 149, 150, 151 Online systems, 72, 249, 250, 286 observing user behaviour, 131 record of searches, 84 software, 77 testing, 128[OCRerr]1 35 aspects that warrant testing, 132 constraints on, 129 database search and print, 133 environment of, 129 index search, 133 opportunities of real life systems, 131 search effectiveness, 133 system and database restraints, 130 search formulation and checking, 132 system effectiveness, 134 types of, 133 user constraints, 130 Operational definitions, 49 Operation tests, 11 Optical character recognition cards, 84 Optimization, 180 ORBIT, 129 Outcome of service, 108 Output, form of, 65 from computer, 83 indicativity of field, 65 measurement of, 108 ranking, 226, 231, 234 variety of, 78 People, collection of data, 84 training and experience of, 65 Performance, absolute and relative, 320 criteria, 113 limits, 16 measurement of, 226 Personnel, 65 Pertinence, 47, 68 Philosophy of information, 180 Portable test collections, 22 Post-coordinate indexing, 218, 219, 226 uniterm system, 136 Post-coordinate natural language terms, 2 Post-coordinate searching, 2 Post-coordination, 249 PRECIS, 142, 144 Precision, 32,36,66, 111, 113, 128, 143 analysis of failures, 118 averaging, 86 calculation of, 32, 66 problems, 67 Cranfield 1, 268, 271 Cranfield 2, 271, 273, 276, 278, 280, 282 definition, 19,34 `12 Index 349 Precision (cont.) differences in, 223 evaluation, 241 in experimental environment, 120 in SMART, 318, 320, 323 limits, 40 relation to recall, 67, 224, 263 results, 242 sensitivity of, 188 test ranges, 234 use of, 189 user's requirement, 112 values of, 223 Precision-boosting power, 206 Precision Engineering study, 282 Precision ratio, 111, 117 in evaluation, 120 limitations, 112 selected, 150 Prediction, 98 Presentation clarity, 142 Pretests, value of, 118 Principles, 143 fundamental, 143 Probabilistic systems, 21 explicit, 203 Probabilistic theory, 157, 169, 187, 202 advantages of, 208 Probability change factor, 206 Probability ranking principle, 35, 40 Problematic situation, 46 Profile compilation, 290 modification times, 294 standard tasks, 291 times, 293, 309 Programs, structured, 77 testing and proving, 174 Proof[OCRerr]n'ented studies, 122 Quantum mechanics, 33 Queries, 13, 44, 46, 63, 222 artificial, 27 formation of, 28 bogus, 225 definition, 10 formulation of, 180 for tests, 287 negotiation, 66 number of, 81 optimization, 180, 182 processing, 75 response of searcher, 64 source of, 73 statement of, 75 statistics, 25 treatment of, 243 Query-term incidence array, 90 Query terms, 231 selection, 239