Login

Publications  •  Project Statistics

Glossary  •  Schools  •  Disciplines
People Search: 
   
Title/Abstract Search: 

Dissertation Information for Maria Ana Cleveland

NAME:
- Maria Ana Cleveland
- (Alias) Maria Ana Divino Cleveland

DEGREE:
- Ph.D.

DISCIPLINE:
- Library and Information Science

SCHOOL:
- Case Western Reserve University (USA) (1976)

ADVISORS:
- William Goffman

COMMITTEE MEMBERS:
- Jeffrey A. Schultz
- Thomas Gleason Morris
- Tefko Saracevic

MPACT Status: Fully Complete

Title: The effect of coding in information retrieval a study of two searching strategies

Abstract: Information retrieval is a complex, subtle intellectual task involving communication between the one who creates information and the person who needs to use the information. There are several basic functions involved in the process: (1) The creation of information, (2) Acquisition, (3) Information record proxy (abstracting, extracting), (4) Representation (indexing, classifying), (5) Coding, (6) Query formation, (7) search strategies, and (8) Dissemination.

Of these basic functions, coding is usually taken for granted and very little has been done to determine what effect a machine-oriented coding system has on the other functions. Coding is too often assumed to be a trivial operation that has little part to play and retrieval effectiveness depends on indexing language and methods, file organization, and search strategy. It is the thesis of this study that a coding is a critical variable.

Coding is the symbolic representation of data. Once the informational content of documents have been analyzed and index terms assigned then the terms are translated into a form so that operations can be conducted by a machine. The coding activity changes the tags into some form which a machine can operate upon to carry out the procedure for selecting information records.

The major objective of this study was to examine the effect of coding on the retrieval process. The secondary objective was to examine the applicability of a specific code—a prime number coding scheme. A testing variable (the varying of the number of indexing terms allowed per document) was introduced since depth of indexing is a reoccurring question of importance to documentalists.

The experimental model was a stimulated retrieval system as developed by William Goffman in a paper called “The Indirect Method of Information Retrieval,” published in 1968. The prime number coding scheme was based on an idea proposed by R. Creighton Buck in a paper called “Studied in Information Storage and Retrieval: On the use of Godel Indices in Coding.” published in 1961.

The experiment consisted in running retrieval tests using the Indirect Method search strategy (with and without a code). The tests were repeated 19 times, varying the number of allowable index terms per document.

The basic finding of this study is that coding interacts with the search strategy employed and that coding itself can affect retrieval results.

The conclusions were (1) The introduction of a code affected retrieval results. No matter how simple or “pure” a coding system is, it introduces a variable. (2) When the Indirect Method is used, indexing depth does not seem to be a significant factor. (3) The prime code is a definite space-saver. (4) A prime code is “machine-feasible” if a relatively limited thesaurus is used in conjunction with a strong search strategy.

Of course, an excellent coding scheme cannot save an ineffective indexing scheme. On the other hand, an excellent, intellectually solid indexing scheme can be seriously minimized by a poorly conceived coding technique.

MPACT Scores for Maria Ana Cleveland

A = 0
C = 0
A+C = 0
T = 0
G = 0
W = 0
TD = 0
TA = 0
calculated 2008-04-09 13:12:38

Advisors and Advisees Graph