A natural language interface to a graph-based bibliographic information retrieval system

This paper describes a system for dealing with a certain kind of textual information. The system has been in operation for about one and a half years. It also indicates the nature of a new and greatly expanded system presently under development.The first system consists essentially of three parts, of which the most important is a thesaurus processor. The assumption is made that document content depends only on word content and that word relationships are defined by an hierarchical structure. The function of the thesaurus processor is to provide a simple language for developing and changing the thesaurus, whenever change is necessary. The remaining parts of the first system are a document processor for updating document files, and a search processor for batch requests which must scan the entire document file. Some statistics are given on performance of the system.The second system includes a thesaurus processor of expanded capability. It also includes a newly developed search language, which can be used to scan records for complex patterns of events. Input-output processing of records and of search results is left to the user of the system.

Download Full-text

Cognitive Engineering Issues in the Design of a Knowledge-Based Information Retrieval System

Proceedings of the Human Factors Society Annual Meeting ◽

10.1177/154193128502900411 ◽

1985 ◽

Vol 29 (4) ◽

pp. 362-366 ◽

Cited By ~ 1

Author(s):

Philip J. Smith ◽

Deborah A. Krawczak ◽

Steven J. Shute ◽

Mark Chignell

Keyword(s):

Information Retrieval ◽

Environmental Pollution ◽

Retrieval System ◽

Information Retrieval System ◽

Cognitive Engineering ◽

Computer Interface ◽

Human Factors Engineering ◽

Human Computer Interface ◽

Bibliographic Information ◽

Knowledge Based

EP-X (Environmental Pollution expert) is a prototype expert system that acts as an expert search intermediary for a bibliographic information retrieval system. It searches for documents in the Chemical Abstracts database in the domain of environmental pollution, providing an intelligent human-computer interface that accommodates a wide variety of user backgrounds and interests. This paper discusses 1) the general problem of bibliographic information retrival, 2) the value of combining Artificial Intelligence (AI) and Human Factors Engineering (HFE) techniques in the design of an intelligent human-computer interface. 3) the general architecture and functioning of EP-X, and 4) examples of specific issues that arise when attempting to apply both AI and HFE techniques to the design of an interactive computer system.

Download Full-text

TEMU KEMBALI INFORMASI BIBLIOGRAFI DENGAN BAHASA ALAMI PADA FIELD JUDUL DAN SUFJEK (Studi Efektivitas Katalog Induk Terpasang Perpustakaan UGM)

Berkala Ilmu Perpustakaan dan Informasi ◽

10.22146/bip.8260 ◽

2015 ◽

Vol 3 (1) ◽

Author(s):

Janu Saptari ◽

Purwono Purwono

Keyword(s):

Information Retrieval ◽

Natural Language ◽

Retrieval System ◽

Research Result ◽

Relevant Information ◽

Information Retrieval System ◽

Data Bases ◽

Searching Strategy ◽

Index Searching

This Research aim to know the retrieval effectiveness on Online Union Catalog of UGM Library, index pattern at data bases, knowing which's more effective of searching at entri oftitle and subjec and also to know the cause ofeffectiveness difference retrieval at the both entri. Along with growth and information accretion, the main problem have shifted from way of accessing information become to chosen the relevant information with its requirement. Information retrieval is not possible to be done in the manual system, because very big information corps and non-stoped growing larger. The information retrieval system is very needed to to assist the consumer in finding information. One of the information retrieval system is online union catalog. With the online union catalog, by users easily can look for the book title, pickings of research and other documentation of library where from and any time. Getting a number of relevant document with the requirement represent the crux in the searching activity, and effectiveness from online union catalog is this represent the key. Effectiveness from a system influenced by a lot of component which each other related/relevant like: quality of input metadata in data bases, index, searching strategy, ability of system application and keyword election. A system told effective if the system can find more amount document/ appropriate information of request by precission /high accuracy. Research done by testing search at online union catalog to use the natural language keyword. Keyword inclusion done at entri of title and entri subjek. Keyword of taken as sampel come from one of title matakuliah of each; every faculty in UGM. From each; every the topic then translated /formulated into 4 natural language keyword. Data obtained is grouped by relevant level. From the acquirement data; then analysed with the test ofnonparametrik Mann Whitney. From processing data got by conclusion that ratio of retrieval at title entries of equal to 66,6% and subject entries of equal to 58,3%. Data got by mean at title entries of equal to 85,9 document with the detail: very relevant 42,4%, less be relevant 24,2% irrelevant and also 33,5%. While at subject entries found by a mean data of equal to 62,5 document with the detail: very relevant 31,9%, less be relevant 26,45 and irrelevant 30,6%. Pursuant to this research result is suggested to increase performance from Online Union Catalog of UGM library. Is things required to improved by technique ofsearching and ability of searching system, the importance of taking care ofquality ofdata input, similarity ofmetadata and always the existence of renewal /updating ofcatalogue data from all member library. Keywords: effectiveness encounter ofretrieval information retrieval system online union catalog.

Download Full-text