Design, query, and evaluate information retrieval systems.
Meadow, Boyce, & Kraft (2006) defined information retrieval as “finding some desired information in a store of information or a database,” but they did not go on to define the precise meaning of a database because there is no “generally accepted, precise” definition (p. 2). To pause for a moment, let’s consider some of the databases a library uses. The most immediate ones that come to mind are the library catalogs which have records containing information about where a specific item is located in the library (or on an electronic record, if we are talking about e-books). Historically, libraries used card catalogs to locate information online, but now they are moving to electronic databases containing records relating to the library’s holdings (or to the holdings of several libraries, if we are talking about library consortia). Then you have online library databases which hold full-text copies of various articles from research journals, trade journal articles, reviews of relevant works, magazine articles, and sometimes whole books. The computerized circulation systems used for control of item circulation are also information retrieval systems (although they are exclusively for internal use only, that is, only library employees can use them to locate records on specific patrons). The paramount consideration in developing and designing an information retrieval system is the information user, “who uses an [information retrieval system] to attempt to satisfy [their information need]” (p. 5). A record, as defined by Meadow, Boyce, & Kraft (2006), is a “set of information about an entity” When creating a record in a database, it is important to know what attributes to include, and what values the attribute has, such as the name or title of the item, the author of the work, what subjects the item in the record has, and where the item is located in the library collections (p. 54). Some information may have closed attributes, defined in part by a controlled vocabulary or a set range of values which can be selected from, while yet others will have undefined values which will need to be entered manually. A user will query the formed database and its attributes and values. Querying is defined as “a statement or series of statements made to an information retrieval system for the purpose of specifying what information is to be retrieved and in what form” (p. 151). These queries can employ the use of the Boolean indicators AND, OR, and NOT, controlled vocabularies such as Library of Congress Subject Headings, and field searching (or “restrict[ing] your search to a specific field”) (Bell, 2012, pp. 19-26). Searches can be evaluated quantitatively for recall, precision, and taking the prior two concepts into account, the ultimate effectiveness of the search. Searches can also be evaluated qualitatively, for such qualities as usability and whether the library patrons consider the database a useful tool. I am incorporating three pieces of evidence. The first piece of evidence was a group project for INFO 202 in which I created a statement of purpose defining the user group and how they are to use the database through a set of rules to follow, a set of database records and a brief reflective conclusion. The second evidence I am submitting is another project for INFO 202 in which I and two other group members compiled a set of 30 articles and gleaned controlled vocabulary terms, listed those records in a database program, and evaluated our successes (or failures) in searching this database using tools of precision, recall, and effectiveness. The third piece of evidence is a searching exercise I completed with a group member in LIBR 244 for about 6 search queries.
I completed two courses, the first being the mandatory INFO 202, which is required of all students entering the SJSU iSchool program, and INFO 244, which dealt with searching online resources, search engines, and databases exclusively. In INFO 202, I learned about attributes, created a set of attributes (and collected these attributes in a controlled vocabulary), created a set of databases, and evaluated these databases for recall, precision, and effectiveness through a set of searches). In INFO 244, I conducted a series of exercises to learn the search strategies of lawn-mowing (or applying filters in successive searches to lower the number of results to increasingly more relevant search results), pearl growing (or taking a small number of results and making them grow into a larger number of results). I then used the information learned in this course and created an informational training packet to teach a specific user group (e.g. grandparents raising grandchildren) how to search the MedLine Plus database.
References
Bell, S. S. (2012). Librarian's Guide to Online Searching (3rd Edition). Santa Barbara, CA, USA: Libraries Unlimited. Retrieved from http://www.ebrary.com
Meadow, C. T., Boyce, B. R., & Kraft, D. H. (2006). Text Information Retrieval Systems. Oxford: Emerald Book Serials and Monographs.