Abstract
A binary approach to data storage and retrieval is introduced. It views the data base as a two-dimensional matrix that relates entities to all possible values the attributes of these entities may take. As such, it provides a unified solution to the two conflicting types of data base transactions-operational and managerial. An analytical investigation of the feasibility of binary storage and a compression method for reducing meaningless areas of the matrix are presented. Storage efficiencies of binary and conventional inverted file methods are compared and evaluated. An analysis of retrieval considerations associated with the binary matrix is given, particularly the issue of going from high to low orders of compression. Results of these analyses indicate that the binary data base's efficiency increases with increases in query complexity. Future research directions are sited and discussed.
Original language | English |
---|---|
Pages (from-to) | 233-254 |
Number of pages | 22 |
Journal | Information Processing and Management |
Volume | 21 |
Issue number | 3 |
DOIs | |
State | Published - 1985 |
Keywords
- Bayes theorem
- Data base
- compression techniques
- data dictionary
- data storage and retrieval
- inverted files
- managerial transactions
- operational transactions