“…complex vs. simple (exhaustivity tuning dimension), and broad vs. narrow (extent tuning dimension). This opens a way to create experimental settings that are more sensitive to situational factors, the issue that has been raised in the Boolean/best-match comparisons [ 11,20].…”
Section: Conclusion and Discussionmentioning
confidence: 99%
“…comparisons, have suggested that studying the overall superiority of one model over the other may be a naive approach [11,20]. Boolean queries seem to perform better in some situations, and best match queries in other situations.…”
Traditional methods for the system-oriented evaluation of Boolean IR systems suffer from validity and reliability problems. Laboratory-based research neglects the searcher and studies suboptimal queries. Research on operational systems falls to make a distinction between searcher performance and system performance. This approach is neither capable of measuring performance at standard points of operation (e.g. across R0.0-RI.0).A new laboratory-based evaluation method for Boolean IR systems is proposed. It is based on a controlled formulation of inclusive query plans, on an automatic conversion of query plans into elementary queries, and on combining elementary queries into optimal queries at standard points of operation. Major results of a large case experiment are reported. The validity, reliability, and efficiency of the method are considered in the light of empirical and analytical test data.
“…complex vs. simple (exhaustivity tuning dimension), and broad vs. narrow (extent tuning dimension). This opens a way to create experimental settings that are more sensitive to situational factors, the issue that has been raised in the Boolean/best-match comparisons [ 11,20].…”
Section: Conclusion and Discussionmentioning
confidence: 99%
“…comparisons, have suggested that studying the overall superiority of one model over the other may be a naive approach [11,20]. Boolean queries seem to perform better in some situations, and best match queries in other situations.…”
Traditional methods for the system-oriented evaluation of Boolean IR systems suffer from validity and reliability problems. Laboratory-based research neglects the searcher and studies suboptimal queries. Research on operational systems falls to make a distinction between searcher performance and system performance. This approach is neither capable of measuring performance at standard points of operation (e.g. across R0.0-RI.0).A new laboratory-based evaluation method for Boolean IR systems is proposed. It is based on a controlled formulation of inclusive query plans, on an automatic conversion of query plans into elementary queries, and on combining elementary queries into optimal queries at standard points of operation. Major results of a large case experiment are reported. The validity, reliability, and efficiency of the method are considered in the light of empirical and analytical test data.
“…Equation (2) is an approximation to the original distance and computes faster than Equation (1). In the test data used in this work, less than 2% of strings contain a repeated 2-gram, while repeated 3-grams are even more rare.…”
Section: Error Counting Methodsmentioning
confidence: 99%
“…However, neither mechanism guarantees superior performance for every query [1]. They encourage the use of unstructured 'natural language' queries by example text (for example, find all documents with a title like 'modeling and reuse of information over time') and return lists of texts ranked according to their similarity with the query.…”
Similarity searching in text databases with multiple field types is still an open problem. We focus our attention on the "COmmunity Research and Development Information Service" (CORDIS) database of the European Union and we evaluate the effectiveness of many text retrieval methods in terms of precision, recall and ranking quality. Our experiments indicate that different field types should be handled by different retrieval methods.
“…) Of all the arguments put forward above, only one (that Boolean search is difficult for novice users) seems valid. A study by Paris and Tibbo () also questioned the criticism of Boolean retrieval:These results do not in any way prove the superiority of partial match techniques or exact match techniques, but they do suggest that different queries demand different techniques. Further study and analysis are needed to determine which elements of a query make it best suited for partial match or exact match retrieval.…”
This paper considers classical bibliographic databases based on the Boolean retrieval model (such as MEDLINE and PsycInfo). This model is challenged by modern search engines and information retrieval (IR) researchers, who often consider Boolean retrieval a less efficient approach. The paper examines this claim and argues for the continued value of Boolean systems, and suggests two further considerations: (a) the important role of human expertise in searching (expert searchers and "information literate" users) and (b) the role of library and information science and knowledge organization (KO) in the design and use of classical databases. An underlying issue is the kind of retrieval system for which one should aim. Warner's (2010) differentiation between the computer science traditions and an older libraryoriented tradition seems important; the former aim to transform queries automatically into (ranked) sets of relevant documents, whereas the latter aims to increase the "selection power" of users. The Boolean retrieval model is valuable in providing users with the power to make informed searches and have full control over what is found and what is not. These issues may have significant implications for the maintenance of information science and KO as research fields as well as for the information profession as a profession in its own right.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
