A Bayesian approach to user stopping rules for information retrieval systems

Kraft, Donald H.; Waller, W.G.

doi:10.1016/0306-4573(81)90049-2

Cited by 8 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Library and information science researchers have found the concept of satisficing useful in understanding human information seeking behaviour [21]. Early examples of the use of bounded rationality and associated stopping rules included the work of Kraft and Waller [22], who drew on Bayesian probability theory when hypothesizing that levels of satiation or disgust would determine rules for an optimal stopping point in a database search. They noted the difficulty a user faces in predetermining their level of satiation or disgust and concluded a need existed for a dynamic model capable of 'reflecting the changes in the user's information requirements' during the progress of the search.…”

Section: Rational Choice Decision Making Dominant In Lismentioning

confidence: 99%

“…However, the mathematic modelling and the experimental contexts used in the development of rational decision theory [27] are very different from the real world settings in which people confront choices during the process of seeking and using information. The earlier studies such as those by Kraft and Waller [22] and Morehead and Rouse [23] typically excluded the possibility of participants receiving and acting on feedback during the decision making process, a constraint that necessarily limits the applicability of these theories to real world situations [28]. More recently, Agosto [26] and Zach [2] both concluded that satisficing as defined by Simon did not fully account for the stopping behaviour of participants in their studies.…”

Section: Rational Choice Decision Making Dominant In Lismentioning

confidence: 99%

See 1 more Smart Citation

Judgements during information seeking: a naturalistic approach to understanding the assessment of enough information

Berryman

2007

Journal of Information Science

View full text Add to dashboard Cite

In this article, theories of human judgement and decision making are reviewed and their use by library and information science researchers examined. A different perspective on judgement and decision making is offered by the field of naturalistic decision making (NDM) and the implications of this approach are considered for an expanded understanding of how judgements and decisions are made during information seeking. This discussion is illustrated by a case from a recent empirical investigation into how judgements of enough information are made in the workplace. The article concludes with a critical evaluation of the NDM approach. It is argued that NDM, a recent development in decision theory, offers a new perspective from which to investigate judgements and decisions during information seeking.

show abstract

Section: Rational Choice Decision Making Dominant In Lismentioning

confidence: 99%

Section: Rational Choice Decision Making Dominant In Lismentioning

confidence: 99%

Judgements during information seeking: a naturalistic approach to understanding the assessment of enough information

Berryman

2007

Journal of Information Science

View full text Add to dashboard Cite

show abstract

“…Most information retrieval systems provide either an ordered list of documents, which the user can examine until they decide to stop retrieving documents (Kraft & Waller, 1981), or a set of documents to be examined, with another set of documents being excluded from presentation to the user (Losee, 1998;Salton & McGill, 1983).…”

Section: Information Retrieval Performancementioning

confidence: 99%

Are two document clusters better than one? The Cluster Performance Question for information retrieval

Losee

Church

2004

J. Am. Soc. Inf. Sci.

View full text Add to dashboard Cite

When do information retrieval systems using two document clusters provide better retrieval performance than systems using no clustering? We answer this question for one set of assumptions and suggest how this may be studied with other assumptions. The "Cluster Hypothesis" asks an empirical question about the relationships between documents and user-supplied relevance judgments, while the "Cluster Performance Question" proposed here focuses on the when and why of information retrieval or digital library performance for clustered and unclustered text databases. This may be generalized to study the relative performance of m versus n clusters.

show abstract

When is “enough” enough? Modeling the information‐seeking and stopping behavior of senior arts administrators

Zach

2004

J. Am. Soc. Inf. Sci.

View full text Add to dashboard Cite

Among managers, those who are responsible for nonprofit organizations in general and arts organizations in particular have been an understudied group. These managers have much in common with their for-profit counterparts, but their environment also differs in significant ways. The goal of this exploratory research effort was to identify how senior administrators in fine arts museums and symphony orchestras go about identifying and acquiring the information they want to complete a range of management tasks. Deciding when and where to look for information, obtaining the "right" information at the time it is needed, evaluating its credibility and utility, and determining when "enough" information has been collected are challenges facing this group of information users every day. A multiple-case studies design involving a replication strategy was selected to structure the research process. Data were collected from 12 arts administrators using a pretested interview protocol that included the Critical Incident Technique. Patterns in the data were identified, and the data were further reviewed for disconfirming evidence. The study resulted in a list of the types and sources of information that arts administrators use as well as a list of the factors or "stopping criteria" that influence them to end their informationseeking process. A model describing the way in which arts administrators go about acquiring the information they want was also developed. The main findings of the study are (a) arts administrators do not consider information seeking to be a discrete management task, (b) they rely heavily on direct personal experience to fill their information-seeking needs, and (c) they are "satisficers" when it comes to seeking information. Based on Simon's alternative to rational choice theory, satisficers are people who are willing to pursue a "good enough" option rather than the best possible option (maximizers) (Simon, 1956). Since arts administrators have not been studied in the context of LIS research before, understanding more about where they go for information, what factors influence the level of effort they are willing to invest in seeking information, and how they decide when they have "enough" information provides insights into the information-seeking behavior of a new user group. Furthermore, although this research effort is focused on specific users in a specific field, the results from this study may be compared to what we already know about other user groups to confirm and expand existing models of information-seeking behavior.

show abstract

A Bayesian approach to user stopping rules for information retrieval systems

Cited by 8 publications

References 11 publications

Judgements during information seeking: a naturalistic approach to understanding the assessment of enough information

Judgements during information seeking: a naturalistic approach to understanding the assessment of enough information

Are two document clusters better than one? The Cluster Performance Question for information retrieval

When is “enough” enough? Modeling the information‐seeking and stopping behavior of senior arts administrators

Contact Info

Product

Resources

About