Established
information retrieval systems that are utilized in academic institutions (e.g. libraries,
colleges/universities), government agencies, and the massive variety of Internet
databases all consider “retrieval evaluation a critical and integral component”
(Baeza-Yates & Ribeiro-Neto,
2011) . This evaluation is a measurement of performance,
correlation of user search choices, and allows for comparison of other IR systems.
§ Optimum IR performance reflects search results that
are relevant to the user – corresponding to search selections
provided by the researcher.
§ Ranking
of documents retrieved should represent high relevance of the user’s subject
matter (topic researched); however, for different users, relevance may be
interpreted subjectively to meet their individual information needs.
§ Top-ranking
documents (via system algorithms) appear at the beginning of the research
results listing, allowing the user to quickly review the most pertinent
information retrieved, and subsequently compare results within other search
engines.
§ Benchmark
algorithms are utilized within information systems to obtain relevant search
retrievals and effective evaluations.
The massive amount of databases correlated to any particular user search
requires precision of algorithmic functions.
According to Baeza-Yates, “Fundamental property that characterizes
metadata is an aboutness relationship with some other resource . . . metadata
is about a specific resource” (Baeza-Yates & Ribeiro-Neto, 2011, p. 716) .
Aboutness
is the concept of what the user’s topic/subject matter pertains to in the
context of retrieving information. Chu
emphasizes that “information objects can be about many things, so
representation of aboutness is inherently incomplete” (Chu, 2010). In determining aboutness, subject analysis
boils down to answering basic questions: What is the information about? What is the information for? The capability of algorithms to retrieve
information (as representation of information is based on aboutness) that
correlates appropriately is the basis of providing relevant information to the
user.
Now, about that chocolate bar?
References
Baeza-Yates, R., & Ribeiro-Neto. (2011). Modern
Information Retrieval; the concepts and technology behind search; second
edition. New York: Pearson Education Limited.
Chu,
H. (2010). Information representation and retrieval in the digital age.
Medford, New Jersey: Information Today, Inc.