Retrieval Software Research Articles

First steps in choosing information retrieval packages

Microcomputers offer a new scope and a new challenge to libraries and information units. For many, computerisation of procedures is at last within budgetary restrictions and the possibility of providing new services through the use of the microcomputer is a strong incentive to buy. However, the implementation of a useful working computer system requires careful research and planning which is often left to the purchaser because of the high cost of consultancy relative to the cost of a micro‐computer. Self help is the only realistic option for many libraries. This article is intended to give useful background information and to help structure ideas before the business of selecting a suitable information retrieval (IR) package is begun. This is an introductory article intending to outline basic concepts.

SIRE: Information storage and searching

SIRE is an information retrieval package that enables a user to design, build, search and maintain text databases. It is easy to learn and use, and it has many powerful features. As we delved into our study, we began to compare SIRE to driving a car with automatic transmission — that is, all the gears and functions are there, but so much happens without intervention of the driver that it can appear to be less complex than it really is. In some ways SIRE's ease‐of‐use and streamlined features may lead users to be unaware of its power and sophistication; it is well suited for the non‐professional searcher. For anyone wanting to search PC‐based text files, SIRE is a software package they will definitely want to consider.

The CD‐ROM Services of SilverPlatter(tm) Information, Inc.

The SilverPlatter system is a complete, stand‐alone system, consisting of an IBM (or compatible) personal computer, a CD‐ROM drive, software, and one or more databases. The search and retrieval software is logical, powerful, and simple. Very large databases, such as ERIC and PsycLIT, will soon be available on the system for “local” installation in schools, libraries, and (even) psychiatrists' offices.

An overview of information retrieval software for microcomputers in France

The potential uses of microcomputers seem unlimited and information retrieval is now an important area of application. This paper presents the teachings and the conclusions of a survey undertaken by the DBMIST (Direction des Bibliothèques, des Musées et de l'Information Scientifique et Technique) in the French Department of National Education. Forty systems have been studied but only ten of them appear to be interesting. A range of criteria allows us to classify packages in two main categories: the first suitable for personal systems or very small information centres recording up to approximately 5,000 documents; and the second appropriate for average documentation units.

4–1–1: file retrieval package

4–1–1 is well‐suited for situations where numerous documents stored on a hard disk need to be organized and categorized for expedient retrieval. It is excellent for handling such files as those for correspondence, personnel and budget management. 4–1–1's interactive search capabilities and document security features will be particularly attractive to those who handle confidential material. By maintaining 4–1–1's catalog conscientiously, it is possible to eliminate, or at least diminish the number of paper copies of records kept in vertical files.

ACNUC--a portable retrieval system for nucleic acid sequence databases: logical and physical designs and usage.

ACNUC is a database structure and retrieval software for use with either the GenBank or EMBL nucleic acid sequence data collections. The nucleotide and textual data furnished by both collections are each restructured into a database that allows sequence retrieval on a multi-criterion basis. The main selection criteria are: species (or higher order taxon), keyword, reference, journal, author, and organelle; all logical combinations of these criteria can be used. Direct access to sequence regions that code for a specific product (protein, tRNA or rRNA) is provided. A versatile extraction procedure copies selected sequences, or fragments of them, from the database to user files suitable to be analysed by user-supplied application programs. A detailed help mechanism is provided to aid the user at any time during the retrieval session. All software has been written in FORTRAN 77 which guarantees a high degree of transportability to minicomputers or mainframes.

Information Retrieval Systems for Microcomputers

The processing and storage capacity of microcomputers can now support sophisticated and powerful information retrieval systems, previously available only on larger mainframes. The characteristics of information retrieval software are discussed as are the performance capabilities of two software programs, SIRE and ZyINDEX.

A Computer-Implemented Data Base for Coalbed Methane Resource Characterization and Research

Summary The USBM has obtained gas content data on more than 1,500 coal samples from approximately 250 coalbeds in 17 states. A computer implemented coalbed methane data base has been established to respond more efficiently to health, safety, and mine-planning related data requests from the coal industry and to the resource characterization needs of those interested in using coal bed methane. The system also will be used for coalbed methane research, including multivariate statistical analysis of the physical, chemical, and geologic province factors influencing the methane generation and storage capacity of coal, with the ultimate goal of developing an indirect method of estimating the gas content of coalbeds. Introduction The USBM began measuring the gas content of exploration coal cores by the direct method in 1972. Since that time, more than 1,500 coal samples have been tested. As data volume increased and additional analytical procedures were included in the research program, the need for more efficient responses to industry data requests and for in-house data analysis became apparent. For these reasons, a coalbed methane data base has been established and is available for use. Direct Method Gas Content Testing The direct method test procedure was developed by the USBM as an aid in estimating the amount of methane that would be released in underground coal mines. Our initial research results were used to construct a graph (Fig. 1) that related direct method test values to actual measured methane emissions from nearby mines. The correlation was good for large deep mines that had been in operation for several years and that had a sustained coal production of at least several thousand tons per day. The gas content data production of at least several thousand tons per day. The gas content data also can be used in conjunction with basic coal geology mapping to delineate geographic areas of probable high (or low) volumes of in-situ gas. Detailed studies of this type have been documented. This information is useful for both mine planning and resource recovery and utilization applications. A list of estimated in-situ gas volumes for selected coalbed areas is given in Table 1. Test Procedures The direct method test procedure is performed on fresh exploration coal cores collected at the drillsite. After the core's physical characteristics are described, the sample is sealed in a desorption canister. Periodically, the accumulated gas is bled into a water-filled graduated cylinder (Fig. 2) to measure the volume of gas released. The total gas content of a sample is composed of three parts: lost gas-that portion lost before the coal is sealed in the canister (estimated by a portion lost before the coal is sealed in the canister (estimated by a graphical procedure); desorbed gas-that portion liberated from the sample while sealed in the canister (measured directly by the water displacement method); and residual gas-that portion remaining in the coal sample at the end of the desorption period that will not freely desorb (determined by crushing the sample in a sealed ball mill and measuring the released gas by the water displacement method). Database System The software of the database system is a menu-driven program (Fig. 3) which selects data from the main data bank for database processing (reports for requestors) or as a file suitable for statistical analysis. The data base is on our PDP 11/70 TM computer. Database processing is executed with a commercially available storage, sorting, retrieval, and report-writing software package (Datatrieve 11 TM). Information requests are processed with the system, and the data in the desired output format are printed and forwarded to the requestor. Information such as coalbed name, location (state and county), depth of sample, gas content, and coal rank is available from the system. A representative printout is shown in Fig. 4. Data analysis is executed by extracting data from the main data bank and transferring it to an output file (Fig. 3) with a FORTRAN-compatible format. These data can be read as input to statistical analysis programs. The database system is being used in the first attempt to evaluate the influences and interactions of all measured physical, chemical, and geologic province parameters on measured in-situ gas contents. JPT p. 846

The use of a commercial microcomputer database management system as the basis for bibliographic information retrieval

A bibliographic retrieval package developed on the database management system (DBMS), dBASE II, under the CP/M operating system is described, together with its use by the author. The background to its development and a discussion of the DBMS approach is included. The system is intended for use by non-trained staff and its acceptance of a wide range of document types makes it ideal for use in special libraries. The article demonstrates how general purpose software may be utilized by information units.

DBASE II — Library use of a microcomputer database management system

The nature of database management software is considered in the light of library requirements, and is compared with information retrieval software. Specific features of one system, dBASE II, are considered with respect to a particular library application, a collection consisting of library supplier trade literature maintained by the Library of the College of Librarianship Wales. Database management software seems appropriate in such a case as ad hoc retrieval is not a requirement, as the database management system permits rapid system development, and facilitates the creation of printed lists derived from the system's databases.

Information storage and retrieval systems on mainframes and minicomputers: a comparison of text retrieval packages available in the UK

Information provided by the vendors of nine text based package systems is presented in a standard format. Features which are now generally accepted as ‘necessary’ are identified, and the areas of difference between the various package systems are discussed. Criteria are proposed for package selection by prospective users. The packages are ADP/3RIP, ASSASSIN 6, BASIS, CAIRS, DECO, DOCU/MASTER, INFOText, SEARCH and STATUS.

Developing custom software and merging with an existing retrieval system: a case study

The stages of effort undertaken by the National Epilepsy Library and Resource Center of the Epilepsy Foundation of America to dovetail an off‐the‐shelf database retrieval package with a custom correspondence response package are presented. In a Wang VS‐80 minicomputer environment, a concept for a package to streamline the process of answering 25,000 queries from the general public, using an in‐place resources database for referral requirements, was developed. A methodical and thorough system design phase emphasizing operator friendliness and top‐down design coupled with appropriate experience with the Wang operating system was essential for success. Involving staff who possessed different operational perspectives in the system testing stage is recommended. Suggestions on how to avoid oversights in all these critical stages are presented. The original flexibility designed into the software has produced a package that can be enhanced in order to remain responsive to the Foundation's needs in the years to come.

OpScan-MIMS. A data network system.

The expansion of a unique system for medical, epidemiological, and laboratory information management is described. The system utilizes an automated, optically scanned data entry mode (OpScan), and a generalized, interactive storage and retrieval software package (MIMS). Aggregate medical, epidemiological, and laboratory information are monitored, with the permission of the locality, via telephone terminals at four sexually transmitted disease clinics. OpScan-MIMS combines an efficient, rapid, and cost-effective method for data entry with a user-friendly computer system that requires no specialized training in computer language.

Microcomputers for information storage and retrieval

Information storage and retrieval (ISR) is a concept associated with large quantities of information stored in relatively sizable (and therefore, costly) computers and which requires relatively extensive software to accommodate the whole process. Generally, the connection between software and hardware is extremely strong; i.e., a particular Information Storage and Retrieval package can only be used on the computer for which it was written. This is the case, for example, with IBM's ‘Stairs’, Siemens’ ‘Golem’ and Univac's ‘Unidas’ package.

Software for microcomputers in libraries and information units

Microcomputers are increasingly being used in libraries and information units and this paper provides an overview of the software available for these applications. A description of a computer system's software is given as well as a checklist of points to consider when acquiring software. The available software is described under the headings of basic software, word processing software, database management systems, inhouse information retrieval software, software for assisting with the searching of external online information retrieval systems, inhouse library housekeeping software, and financial and management software. Many references are given as well as the names and addresses of organisations supplying software.

Pitfalls in the use of Retrieval Software

Highlights some of the problems involved in the use of audit retrieval software. Specifically, the following points have been addressed. Defective data may distort the results obtained by using retrieval software. To avoid such distortion, the auditor should employ an integrity oriented retrieval strategy. Guidelines for developing integrity oriented retrievals were provided. The auditor's perspective in parallel processing was defined. Guidelines for performing audits based on parallel processing were established.

‘POLIS’ IN PARLIAMENT: COMPUTER‐BASED INFORMATION RETRIEVAL IN THE HOUSE OF COMMONS LIBRARY

The Library provides information, documentation and research services to Members of the House of Commons. The organization of the Library and the history of POLIS (the Parliamentary On‐Line Information System) are described. The system became operational in 1980, and was set up by and is operated by Scicon Computer Services Ltd. Data entry is performed by the Library's Indexing Unit. Details of the telecommunications, software, hardware and database are given. UNIDAS retrieval software is used, and subject indexing is based on a thesaurus compiled by the Library. Most Library staff are being trained to use the retrieval facilities. Other offices of the House of Commons and of the House of Lords also have terminals linked to the system. Approved non‐Parliamentary users may also subscribe to POLIS via the public‐switched telephone network, Euronet or PSS. Other uses of computers by the Library are described.

Comparison of cairs and status information storage and retrieval software over a 3 month period

Following a period of about a year in which “desk” assessments were made of information storage and retrieval packages to narrow the options down, the software packages CAIRS and STATUS were evaluated operationally in the Ministry of Agriculture, Fisheries and Food, during a the period between February and July 1981. The packages, loaded on to separate computers, were used to build up databases and databanks based on existing technical information/library services. As far as possible, direct comparison were made from implementation through to retrievals. A range of types of services from bibliographic (library) through to factual text and numeric data ranging across several Divisions of the Ministry were covered. Input and retrieval was remotely via GPO lines and private wire networks. The paper presents the major conclusions with particular emphasis on the databank component of the trial.

The British Library automated information service

The British Library Automated Information Service (BLAISE) is a new name in the library and information world. It is in essence an on‐line library‐housekeeping and information service which is being created by the British Library based on established information retrieval software. The Library will use the ELHILL IHC software package which is run by the National Library of Medicine in the USA. BLAISE will provide on‐line access to the MEDLARS and the MARC files with access to the former in the spring of 1977 on an IBM 370 computer at a bureau in Harlow, Essex.

Associative/parallel processors for searching very large textual data bases

This paper describes an approach to solving a major problem in the information processing sciences— that of searching very large (5-50 billion characters) data bases of unstructured free-text for random queries within a reasonable time and at an affordable price. The need by information specialists and knowledge workers for large, fast low-cost text and document retrieval systems is growing rapidly. Conventional approaches to the problem have usually depended upon expensive, general purpose computers, upon special pre-preprocessing of the textual data (e.g. file inverting, indexing, abstracting, etc.), and upon elaborate, costly software. The resulting retrieval systems often cost hundreds of dollars per query and the full scanning of an uninverted, unstructured billion byte textual data base could take hours of computer services. However, in spite of these restrictions, such full text search systems have proved useful and even indispensible for many applications. Computer technology of the late 1960's and the 1970's, in both hardware and software (e.g., minicomputers, low-cost, high density disk storage, “chip” electronics, natural language query systems, etc.), have made i t practical to build special purpose, low-cost text retrieval systems. Such a system has been built, tested, and is now in a production stage. The system called the Associative File Processor (AFP), utilizes a conventional minicomputer (DEC's PDP-11/45) for control, off-the-shelf high density disks for storage, a special purpose parallel search module as a text term detector, and query and retrieval software. The AFP is currently being field tested at two sites. Full text, parallel searches on un-preprocessed textual data bases are being performed at the effective matching rates of 4 billion bytes per second (8K byte key memory times 500 Kbyte/second data stream). Estimated costs are 10 to 25 cents per query for a one billion byte data base. The costs per query and the time for searching increase in a linear fashion as data base increases. A basic architecture for the AFP is described and an implemented version is discussed. A more powerful term detector module is also under development. This system is designed around a finite state automaton algorithm.