SEARCH · Search PubMed
Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
End-user searching of National Library of Medicine (NLM) online data bases during eleven years has been investigated through transaction logs, questionnaires, and follow-up interviews. From 1976 through 1984, pathologists and pharmacists performed 8,313 searches. Highlights of our studies are compared with a review of other end-user research. Volume of searching is directly related to the convenient placement of the terminal in the work place. Slightly fewer than half of all potential searchers actually search for themselves. Practices of pharmacists and pathologists do not differ in important ways. Nonmediated searchers feel they need answers more promptly than do those who obtain mediated searches. End-users perform very simple searches, mostly using only the AND operator. Problems with techniques are fewer and more easily solved than those with the vocabulary and content of the system. The major problems, with the most powerful capabilities of MEDLINE--subheadings and explosions--sometimes cause substantial loss of references, but in relatively few searches. One-on-one teaching is most popular, with trial-and-error the most frequent procedure used in actual learning.
This article reports on five separate studies designed for the National Library of Medicine (NLM) to develop and test methodologies for evaluating the products of large databases. The methodologies were tested on literatures of the medical behavioral sciences (MBS). One of these studies examined how well NLM covered MBS monographic literature using CATLINE and OCLC. Another examined MBS journal and serial literature coverage in MEDLINE and other MBS-related databases available through DIALOG. These two studies used 1010 items derived from the reference lists of sixty-one journals, and tested for gaps and overlaps in coverage in the various databases. A third study examined the quality of the indexing NLM provides to MBS literatures and developed a measure of indexing as a system component. The final two studies explored how well MEDLINE retrieved documents on topics submitted by MBS professionals and how online searchers viewed MEDLINE (and other systems and databases) in handling MBS topics. The five studies yielded both broad research outcomes and specific recommendations to NLM.
The national Library of Medicine celebrated its sesquicentennial in 1986. The Library has played an important role in medical communication in this country for most of the 150 years since its founding. In the last 20 years, especially, the NLM's responsibilities have been expanded to include a variety of nontraditional library activities. Many of these new activities are based on computer and communications technology. Research now going on in the Library's Lister Hill Center shows promise of greatly altering the way future scientists and health practitioners have access to biomedical information.
This article describes the Indexing Aid Project for conducting research in the areas of knowledge representation and indexing for information retrieval in order to develop interactive knowledge-based systems for computer-assisted indexing of the periodical medical literature. The system uses an experimental frame-based knowledge representation language, FrameKit, implemented in Franz Lisp. The initial prototype is designed to interact with trained MEDLINE indexers who will be prompted to enter subject terms as slot values in filling in document-specific frame data structures that are derived from the knowledge-base frames. In addition, the automatic application of rules associated with the knowledge-base frames produces a set of Medical Subject Heading (MeSH) keyword indices to the document. Important features of the system are representation of explicit relationships through slots which express the relations; slot values, restrictions, and rules made available by inheritance through "is-a" hierarchies; slot values denoted by functions that retrieve values from other slots; and restrictions on slot values displayable during data entry.
This article discusses the development of a vertically oriented CD-ROM database product in the medical subdiscipline of oncology. Called OncoDisc, the CD-ROM is mastered by ISG. It contains three major information collections: (1) PDQ (a system of a series of files and a set of relationships; the user accesses the data through those relationships); (2) Cancerlit (the research literature that underlies the treatment information contained in PDQ), and (3) full-text articles. SearchLITE, the retrieval system, is written in C language and has been implemented on the the DEC VAX family and the IBM PC/XT and PC/AT. The disc provides a personal library of oncology information for immediate local use by the health professional; it requires no subscription to an online service, no telecommunications, and no online search charges.
Professionals in the information field need a variety of up-to-date skills. Currently there are many options for learning and practicing these skills. Academic departments, professional societies, consultants, and various commercial organizations, all of whom offer courses, conferences, workshops, seminars, institutes and the like, compete with each other for training dollars. Other options for trainees include in-house training, how-to books and manuals, databases on CD-ROM, CAI and emulators, and videotapes. Major changes in training are taking place as a result of the rapidly increasing number of direct end users of information services and products. All potential trainees, whether end users or information professionals, must carefully assess the relative value of the training options available.
Search results for nine topics in the Medical Behavioral Sciences are reanalyzed to compare the overall performance of descriptor and citation search strategies in identifying relevant and novel documents. Overlap percentages between an aggregate "descriptor-based" database (MEDLINE, EXCERPTA MEDICA, PSYCINFO) and an aggregate "citation-based" database (SCISEARCH, SOCIAL SCISEARCH) ranged from 1% to 26%, with a median overlap of 8% relevant retrievals found using both search strategies. For seven topics in which both descriptor and citation strategies produced reasonably substantial retrievals, two patterns of search performance and novelty distribution were observed: 1) Where descriptor and citation retrieval showed little overlap, novelty retrieval percentages differed by 17-23% between the two strategies; 2) Topics with a relatively high percentage retrieval overlap showed little difference (1-4%) in descriptor and citation novelty retrieval percentages. These results reflect the varying partial congruence of two literature networks and represent two different types of subject relevance.
This article reports on exploratory experiments in evaluating and improving a thesaurus through studying its effect on retrieval. A formula called DISTANCE was developed to measure the conceptual distance between queries and documents encoded as sets of thesaurus terms. DISTANCE references MeSH (Medical Subject Headings) and assesses the degree of match between a MeSH-encoded query and document. The performance of DISTANCE on MeSH is compared to the performance of people in the assessment of conceptual distance between queries and documents, and is found to simulate with surprising accuracy the human performance. The power of the computer simulation stems both from the tendency of people to rely heavily on broader-than (BT) relations in making decisions about conceptual distance and from the thousands of accurate BT relations in MeSH. One source for discrepancy between the algorithms' measurement of closeness between query and document and people's measurement of closeness between query and document is occasional inconsistency in the BT relations. Our experiments with adding non-BT relations to MeSH showed how these non-BT non-BT relations to MeSH showed how these non-BT relations could improve document ranking, if DISTANCE were also appropriately revised to treat these relations differently from BT relations.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
There have been a number of major evaluations of the performance of retrieval systems against large full text and surrogate (bibliographic) databases. These evaluations have concentrated on the experimental determination of the Precision Ratio, the fraction of retrieved items that are relevant to an information request, and the Recall Ratio, the fraction of the total number of relevant items that were actually retrieved. While these measures have met with general acceptance, they have also generated much controversy. The purpose of this article is to review the results of several of the largest evaluations and to propose a simple model for the performance of such systems that may help explain the relationship between these measures and user behavior.
In order to investigate the current state of the art in clinical cancer research, a survey of comparative cancer clinical trials was conducted using a MEDLINE literature search of the 6-month interval from July to December of 1985. Data were obtained primarily from the published abstracts. The major observations from the study are as follows. First, trials in cancer research are published in an extraordinarily wide range of journals covering most of the major medical specialties in addition to the many cancer specialty journals. Second, randomization is now used quite extensively. It is the predominant method of control in chemotherapy trials and in trials reported in cancer journals. However, its use is much less common in other specialties, especially surgery, so efforts to popularize randomization in these specialties would be beneficial. Third, sample sizes are highly variable, and the median sample size (96 patients) is too small to reliably detect and evaluate moderate treatment advances. Finally, our survey reveals that published trials show an inordinately large proportion of breakthroughs in treatment compared with the generally accepted view that only slow progress is being made in developing effective cancer treatments. Our view is that this result reflects the strong tendency to publish only studies with positive results. This hypothesis is supported by the fact that in nonrandom trials there tend to be more positive conclusions and an absence of a strong association between conclusions and sample size.
MEDLINE is presented as a prototype for on-line bibliographic search systems. Creation of the data base, indexing language, and file organization are reviewed. On accessing the files, search logic is illustrated with a sample MEDLINE search. NLM's development of a document delivery system to complement its bibliographic retrieval system is discussed.
Passage retrieval (already operational for lawyers) has advantages in output form over reference retrieval and is economically feasible. Previous experiments in passage retrieval for scientists have demonstrated recall and false retrieval rates as good or better than those of present reference retrieval services. The present experiment involved a greater variety of forms of retrieval question. In addition, search words were selected independently by two different people for each retrieval question. The search words selected, in combination with the computer procedures used for passage retrieval, produced average recall ratios of 72 and 67%, respectively, for the two selectors. The false retrieval rates were (except for one predictably difficult question) respectively 13 and 10 falsely retrieved sentences per answer-paper retrieved.
Explore the source record for details and available documents.
The National Library of Medicine has offered TOXLINE, an online interactive bibliographic database of biomedical (toxicology) information since 1972. Files from 11 secondary sources comprise the TOXLINE database. The sources supplied bibliographic records in different formats and data structures. Data from each supplier's format had to be converted into a format suitable for TOXLINE. Three different, successive retrieval systems were used for the TOXLINE database which required reformatting of the data. Algorithms for generating terms for inverted file search methods were tested. Special characters peculiar to the scientific literature were evaluated during search term generation. Developing search term algorithms for chemical names in the scientific literature required techniques different from those used for nonscientific literature. Problems with replication of bibliographic records from multiple secondary sources are described. Some observations about online interactive databases since TOXLINE was first offered are noted.