Proceedings of the International Conference on Scientific Information Two Volumes (1959) / Chapter Skim
Currently Skimming:
Classification with Peek-a-boo for Indexing Documents on Aerodynamics: An Experiment in Retrieval
Classification with Peek-a-boo for Indexing Documents on Aerodynamics: An Experiment in Retrieval
Pages 771-802
The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.
From page 771... ...
This index, which might later be extended to include other series of reports on aerodynamic subjects, would be based on a suitable classification system permitting conjunctive specification, and would use a clerical device of the peek-a-boo type all. The system would employ a minimum of expensive equipment and, if successful, might be applicable to other fields requiring cletailed analytical indexing of a comparatively small collection.
|
From page 772... ...
index sheets one for each subject heading and for each designated aircraft and aerofoil; register serial numbers of documents to be indicated by the positions of punched hc,] ~s For testing purposes, however, it was decided to use only the classification schedule itself and standard 80-column Hollerith cards.
|
From page 773... ...
PROCEDURE For test indexing the selection of documents was not limited to R.A.E. documents, some Institute of Aeronautical Sciences preprints and National Advisory Committee for Aeronautics Research Memoranda being included in order to achieve wider subject coverage, particularly of recent material.
|
From page 774... ...
At first, indexers were taking as long as 30 minutes for each document: practice, familiarity with the schedules and increasing knowledge of the subject soon reduced this to 15-20 minutes per document. It is doubtful whether an average of about 15 minutes can be much reduced, in view of the careful study of each document which is necessary to any system using highly analytical indexing.
|
From page 775... ...
In 7 cases the search was "partly" successful (subject document retrieved, but with more than 5% of the collection) , and in 15 cases the search failed completely.
|
From page 776... ...
A total of all the above single or contributory causes of complete or partial failure may be helpful in pinpointing the chief weaknesses of the system. Semantics Misleading question Question insufficiently specific 2 Question misinterpreted Input stage Indexing omission Schedule inadequacy Clerical error Output stage Faulty search Search too specific Total 3 8 34 6 2 41 6 55 It wig be noted that the largest single factor was indexing omission, and this in turn is believed to be attributable mainly to lack of subject knowledge.
|
From page 777... ...
ANALYSIS BY INDEXER Indexer Questions Documents indexed A B Successful first search Successful second search Successful third search Partly successful Failed Total Successful first search Successful second search Partly successful Failed Total C, D, E, and F Successful first search Successful second search Successful third search Partly successful Failed Total 9, 10, 14, 18, 19, 20, 21, 27,29,42,44,45,46,47, 50,52,62,63,64,71,75, 79, 80, 82, 83, 85, 89, 90, 91, 95, 98, 99, 100 4, 5, 11, 22, 25, 31, 33, 35, 55, 65, 72, 84, 86, 97 48, 67 28, 61 7, 32, 39, 40, 51, 53, 66, 81, 88 3, 6, 38, 59, 68, 73, 74, 92, 94 60, 76, 96 17, 24, 70 1 2, 8, 13, 15, 16, 30, 43, 54, 58, 69, 77, 78 49, 56, 87 37, 41 23, 26 12, 34, 36, 57, 93 33 (55%)
|
From page 778... ...
In 20 cases, therefore, questioners were presented with abstract cards or documents and they decided whether the documents retrieved, in addition to the ones on which their questions were based, were truly relevant in terms of their questions as set. Total documents retrieved for the 20 questions came to 146 (the number of aspect cards correlated varied from 1 to 5, but was usually between 2 and 4~.
|
From page 779... ...
The second phase consists of removing the relevant serially numbered abstract cards from the file, and offering them to the questioner, who may then wish to see some of the actual documents represented by the cards. The total output stage should not normally exceed 15-20 minutes.
|
From page 780... ...
R., Test of N.L.L. card catalogue of aerodynamic measurements.
|
From page 781... ...
WRIGHT and WILSON Classification with Peek-A-Boo 3 Air ~ ~ ~ ~ x cot ~ ~ O ~ ~ I= ^~ ~ o E ~ ~ ~ ~ ~ E ~= 5 D ~ ~ ~ ° ~ ~ ~ 2 ~ D D Cd Cal Cal o ~ .O' ._ .= ~ _ ~ = a,= V' A I _ C~ U)
|
From page 782... ...
782 Comparative Characteristics of Existing Systems AREA 4 )
|
From page 783... ...
WRIGHT and WILSON CIassi~cation with Peek-A-Boo 783 ^ A, 0 0 0 ·— ~ ~ 0 a- x O O ~ .
|
From page 784... ...
784 s: w .0 ¢ 0~% 2 Comparative Characteristics of Existing Systems AREA 4 As s ~ on no - Cal ~ Cal -5: q.
|
From page 785... ...
WRIGHT =] WILSON CIassi~cation with Peek-A-Boo 78S ~ 0 c ~ ~ ~ tO 0 ~ i?
|
From page 786... ...
786 Comparative Characteristics of Existing Systems AREA 4 ^ s .
|
From page 787... ...
WRIGHT and WILSON CIassi~cation with Peek-A-Boo ._ o of .= so 0 ~ To 00 cat up 787 ._ o _` to Cot US Ott I.= ~ ' ¢ ~ ~ to Go ~1 ~ ~ ~ ~ Cal U
|
From page 788... |
From page 789... ...
WRIGHT and WILSON CIass~cation with Peek-A-Boo 789 o en .
|
From page 790... ...
790 Comparative Characteristics of Existing Systems AREA4 ~ ~ ~ X Cal =o ~ i= X o ~ .
|
From page 791... ...
WRIGHT and WILSON Classification with Peek-A-Boo 791 ~ ._ 0 ~ — o 4 4 - _ · ~ cat · == .~ · ¢ V ~ x ~ ~ ~ ~ _ 5 u' cat ~ cat ~ ~ ~ ~ v, cot cat ~ ~ ¢ ¢ ¢ ~ ¢¢ ¢ ~ ~ cat .
|
From page 792... ...
792 Comparative Characteristics ojExisting Systems AREA 4 o en o o -1 1 Cal ~ X .
|
From page 793... ...
WRIGHT and WILSON Classification with Peek-A-Boo cd x ._ o 4 ·s o Hi cat cot ¢ - ~ ~ ~ ~4 en to cat cot cot ~ ~ ¢ ~ ¢ cat ~ ¢ ¢ cat ~ cot ED cot ~ cot en ._ 0 ~ .
|
From page 794... ...
794 Comparative Characteristics of Existing Systems AREA 4 ~ _ .
|
From page 795... ...
WRIGHT an] WILSON CIa~s~cation with Peek-A-Boo 795 ~ ~ , ~ Cal Ce , X 0 - ~ 0 .
|
From page 796... ...
796 Comparative Characteristics of Existing Systems AREA 4 X .= X U)
|
From page 797... ...
. Immersion Other/Free surface/ Mixed flow In own right In relation to solids Boundary Layer 110 18 111 0 112 524 113 230 114 493 115 203 116 3 117 621 118 3 119 77 120 107 121 176 122 253 123 280 124 393 125 324 126 236 127 12 128 359 129 20 130 328 131 18 132 8 133 8 134 38 135 26 136 279 137 94 a Tinges used in the indexing of 700 documents.
|
From page 798... ...
(not Planform 222 89 helicopters) Aspect ratio 223 28 Composite aircraft 186 0 Sweep 224 37 No wing 187 14 Taper 225 19 No body 188 5 Other/span 226 25 No sailplane/No 189 50 Section 227 63 foreplane Camber 228 25 No fin 108 20 Thickness 229 34 With multiple fuselage 190 8 Other/profile 230 35 With fins other than 191 15 Twist 231 22 single tail fin Dihedral/anhedral 232 4 With foreplane/Canard 192 6 Leading edge 233 42 Trailing edge 234 61 Components present and/or ~ Tip 235 42 studied Complete aircraft 193 286 Tailplane/Foreplane 236 79 Other 194 324 Fin 237 35 Wing 195 361 Monoplane 196 229 Propeller 238 32 Biplane 197 16 Cruciform, Y
|
From page 799... ...
Drag increase Other purpose Boundary layer control Wing-located T.E. region L.E.
|
From page 800... ...
used 381 16 382 25 383 17 384 45 385 20 386 47 387 35 Electronics /Radar/Radio 3 88 24 Mathematics Principal test equipment/ Technique Described Wind tunnel Continuous/Fan drive ~ . Intermittent Shock tube Special purpose tunnel/Test rigs Spinning tunnel Water tunnel Ballistic range Analogy/Simulator Ditching tank Towing tank/Moving channel Models mounted on aircraft Tracks/Sleds Free flight Piloted aircraft Free flight Models/ unpiloted Components Power/Flow production Other Auxiliary Apparatus/ Technique Model Design/ Construction Balances 389 51 391 133 392 261 393 45 394 8 395 4 396 23 397 10 398 4 399 4 400 4 401 2 402 12 403 1 404 1 405 92 406 41 408 24 409 15 410 13 411 58 412 49 413 18
|
From page 801... ...
WRIGHT and WILSON Classification with Peek-A-Boo TABLE 2 (continued) Code Timesa No.
|
Key Terms
This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More
information on Chapter Skim is available.