The value of structured abstracts in information retrieval from MEDLINE

Objective: To determine whether information retrieval might be enhanced by making each component of a structured abstract a separately searchable field. Design: Parallel searches on records arranged so that each abstract could either be searched in its entirety or as a collection of structured fields. Subjects: Records about cardiovascular disease downloaded from MEDLINE and tested against clinical questions derived from a concurrent survey of CD-ROM use in three health science libraries. Intervention: The investigators identified terms used to structure abstracts on MEDLINE and imported the abstracts into an Idealist database prior to searching using keywords from clinical questions. Measurements and main results: recall and precision rates are given for each question in the ‘structured’ and the ‘unstructured’ set of records. Conclusions: Use of structured, rather than unstructured, abstracts improves precision at the ex-pense of recall and places heavier demands upon the searcher’s skill in choosing the correct fields within the abstract to search. Further investigation requires more accurate simulation of MEDLINE software that utilizes features such as MeSH terms, explode, facilities and delimiters.     

Indexing consistency in MEDLINE

The quality of indexing of periodicals in a bibliographic data base cannot be measured directly, as there is no one "correct" way to index an item. However, consistency can be used to measure the reliability of indexing. To measure consistency in MEDLINE, 760 twice-indexed articles from 42 periodical issues were identified in the data base, and their indexing compared. Consistency, expressed as a percentage, was measured using Hooper's equation. Overall, checktags had the highest consistency. Medical Subject Headings (MeSH) and subheadings were applied more consistently to central concepts than to peripheral points. When subheadings were added to a main heading, consistency was lowered. "Floating" subheadings were more consistent than were attached subheadings. Indexing consistency was not affected by journal indexing priority, language, or length of the article. Terms from MeSH Tree Structure categories A, B, and D appeared more often than expected in the high-consistency articles; whereas terms from categories E, F, H, and N appeared more often than expected in the low-consistency articles. MEDLINE, with its excellent controlled vocabulary, exemplary quality control, and highly trained indexers, probably represents the state of the art in manually indexed data bases.     

Analysis of Academic Libraries' Facebook Posts: Text and Data Analytics

This research analyzed a dataset of academic libraries' posts on Facebook. It applied a text and data analytics approach to a dataset collected from the Facebook posts of academic libraries at the top 100 English-speaking universities, as listed by the 2014 Shanghai World University Rankings. The dataset is from a two-year posting history of 18,333 unique posts, 113,621 likes, and 3401 comments. Less than a quarter of the libraries had more than 2000 post-related likes, and only seven received more than 100 comments on their postings. Content analysis identified the most prevalent single word (unigrams), bigrams (two-word sequences), and trigrams (three-word sequences) in high and low engagement content. Semantic analysis identified the semantic categories for posts with high and low engagement. The findings can assist academic libraries in their social media strategies for engagement, marketing, and visibility.