R. Thompson and W. Croft. Support for browsing in an intelligent text retrieval system. International Journal of Man-Machine Studies, pages 639-668, 1989.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....interesting and useful, and we have no evidence to the contrary, but we have not yet developed a formal evaluation that can measure this. 4. RELATED WORK There have been a large number of systems built for the purpose of browsing the information within text collections. Ex amples include IsR [17], Kohonen Maps [19] Themescapes and Galaxies[18] and Galaxy of News[14] These systems select significant words and phrases and display them in such a way as to allow the user to graphically gist what topics are contained in a system. All these systems are term centered rather than document ....

R. Thompson and W. Croft. Support for browsing in an intelligent text retrieval system. International Journal of Man-Machine Studies, pages 639-668, 1989.

....task, and traverse the associated links in the component repository. In general, people who look for information prefer browsing to searching because they do not need to commit resources at first and can incrementally develop their requirements after evaluating the information along the way [22]. Mili et al. 17] claim that browsing is the predominant pattern of component repository usage because many software developers often cannot formulate clearly defined queries. Instead, they rely on browsing to get acquainted with available components in the repository. Browsing, however, is not ....

Thompson, R.H. and W.B. Croft, Support for Browsing in an Intelligent Text Retrieval Systerr International Journal of Man-Machine Studies. 30(6):639-668, 1989.

....of indexing are not negligible. This is particularly true in certain types of search when there is no good goal definition. The amount of search required is much greater than the previous systems support. Browsing is an attractive alternative. It has been used extensively in information retrieval [29,48], knowledge bases [39] and hypertext [14,47,49] and even relational databases [42] 17 R. Thompson et al. 48] offer support for browsing in information retrieval. They make the statement with reference to Bates paper [2] Bates (1986) points out the advantage of browsing by showing how it ....

....The amount of search required is much greater than the previous systems support. Browsing is an attractive alternative. It has been used extensively in information retrieval [29,48] knowledge bases [39] and hypertext [14,47,49] and even relational databases [42] 17 R. Thompson et al. [48] offer support for browsing in information retrieval. They make the statement with reference to Bates paper [2] Bates (1986) points out the advantage of browsing by showing how it takes advantage of two cognitive capabilities. The first one is the greater ability to recognize what is wanted over ....

[Article contains additional citation context not shown here]

R.H. Thompson and W. B. Croft. Support for Browsing In An Intelligent Text Retrieval System. Int. J. Man-Machine Studies Vol. 30 pp 639-668 (1989)

....Another example of structured documents are web documents, where the structure can be both internal (HTML or XML) and external (documents hierarchically connected by hyper links) In this work, we are concerned with hierarchically inter linked web documents. Many authors such as [Chi97] Fri88] [TC89] and [CvR96] argue that to allow for an e ective retrieval of structured web documents, the information retrieval (IR) model must exploit not only the content of documents, but also the structural knowledge associated with the documents, that is, the hyper links (or links) that connect documents. ....

R.H. Thomson and W.B. Croft. Support for browsing in an intelligent text retrieval system. International Journal of Man-Machine Studies, 30(2):639-668, 1989.

.... on a measure of each document s similarity to the original query [9] Some work with general queries and general web pages [8] while others are tailored to more focused tasks [10] A second class of tools provide graphical means for accessing data based, for example, on inter document similarity [1, 12, 4], relationships to fixed attributes [11, 6] and query term distribution patterns [5] A third approach is to cluster the collection of documents. For example, hierarchical agglomerative clustering (HAC) 13] finds cluster 1 A slightly condensed version of this paper was published in KDD 98 [7] ....

R. H. Thompson and B. W. Croft. Support for browsing in an intelligent text retrieval system. International Journal of Man-Machine Studies, 30(6):639--668, 1989.

....1 Introduction An important service offered by an information access system is the organization of retrieval results. Conventional systems rank results based on an automatic assessment of relevance to the query [20] Alternatives include graphical displays of interdocument similarity (e.g. [1, 22, 7]) relationship to fixed attributes (e.g. 21, 14] and query term distribution patterns (e.g. 12] In this paper we will discuss and evaluate the use of Scatter Gather [4, 5] as a tool for navigating retrieval results. The Scatter Gather browsing paradigm clusters documents into ....

R. H. Thompson and B. W. Croft. Support for browsing in an intelligent text retrieval system. International Journal of Man [sic] -Machine Studies, 30(6):639--668, 1989.

....and reliability of searching software (or other kinds of) libraries has considered a wide variety of techniques for improving browsing systems. A survey of techniques specific to software reuse may be found in [31] and [17] Analogous techniques for information retrieval systems are summarized in [38]. 1 In the most common approach to improving browsing, the browsing system remains purely passive. The speed and reliability of browsing are improved by increasing the effectiveness of the browsing operations, either by introducing new, more powerful operations or by organizing the library so ....

....theory postulates that people naturally think about categories of things not in terms of formal attributes but in terms of examples . Overall, browsing is seen as a more natural and effective process when the user is uncertain of the target description. In summarizing the advantages of browsing, [38] cites [2] Bates (1986) points out the advantage of browsing by showing how it takes advantage of two cognitive capabilities. The first one is the greater ability to recognize what is wanted over being able to describe it. The second capability is being able to skim or perceive at a ....

R. H. Thompson and W. B. Croft. Support for Browsing In An Intelligent Text Retrieval System. Int. J. Man-Machine Studies Vol. 30 pp 639--668 (1989)

....aimed at improving the speed and 1 reliability of searching software (or other kinds of) libraries has considered a wide variety of techniques. A survey of techniques specific to software reuse may be found in [34] and [19] Analogous techniques for information retrieval systems are summarized in [43]. The speed and reliability of search are improved by increasing the effectiveness of the search operators, either by introducing new, more powerful operators or by organizing the library so that the existing operators are more effective. Our approach to improving search is complementary to this. ....

....theory postulates that people naturally think about categories of things not in terms of formal attributes but in terms of examples . Overall, browsing is seen as a more natural and effective process when the user is uncertain of the target description. In summarizing the advantages of browsing, [43] cites [2] Bates (1986) points out the advantage of browsing by showing how it takes advantage of two cognitive capabilities. The first one is the greater ability to recognize what is wanted over being able to describe it. The second capability is being able to skim or perceive at a glance. ....

[Article contains additional citation context not shown here]

R. H. Thompson and W. B. Croft. (1989) Support for Browsing In An Intelligent Text Retrieval System. International Journal of Man-Machine Studies Vol. 30 pp 639--668

....Hypermedia publication Knowledge acquisition Object network 1 INTRODUCTION The increasing volume and complexity of hypermedia publications require multiple means of accessing the information contained in them. These include powerful and flexible browsing, navigation, and retrieval mechanisms [1]. A comprehensible and appealing presentation of the search results is equally important. It is our belief that a combination of improved search techniques and dynamic generation of graphic and text presentation of the retrieved information will be required to achieve better electronic ....

R.H. Thompsonand W.B. Croft, `Support for Browsing in an Intelligent Text Retrieval System', International Journal of Man--Machine Studies, 30, 639--668, (1989).

....and the contents of the database. Many enhancements of the basic approach have been studied such as using relevance feedback, using a thesaurus, extended Boolean retrieval (Salton, 1989) As pointed out by many researchers (Conklin, 1987; Landauer, Dumais, Gomez Furnas, 1982; Oddy, 1977; Thompson Croft, 1989), an information retrieval system should also provide browsing mechanisms for users who do not know precisely what they want or how to get it. Browsing is based on freely exploring a structure such as a tree or a graph in order to find useful items. Browsing in some form of data space has long ....

....followed to find related nodes. Typically, the network which supports the browsing has to be manually constructed which is a major bottleneck particularly for large databases. Some possibilities are being investigated for automatically building the network based on some document similarity measure (Thompson Croft, 1989) or automatic computing of a hierarchy of document clusters for browsing in a restricted set of documents similar to some selected document (Crouch, Crouch Andreas, 1989) Other browsing tools are also used but are more limited in scope such as browsing a thesaurus for helping the user in the ....

Thompson, R. H. & Croft, B. (1989). Support for Browsing in an Intelligent Text Retrieval System. International Journal of Man-Machine Studies, 30, 639-668.

.... used for software repositories can be divided into three categories: enumerated classification, faceted, and free text indexing [Frakes, Gandel 1990] Hypertext systems have also been used, although some form of retrieval is necessary to make hypertext effective in large repositories [Halasz 1988; Thompson, Croft 1989] Enumerated classification is a well known retrieval method used by the Dewy Decimal system and the Computing Reviews Classification System [ACM 1992] In this method, information is placed in categories that are usually structured in a hierarchy of sub categories, much like the Unix file ....

Thompson, R. H., Croft, W. B. "Support for Browsing in an Intelligent Text Retrieval System," International Journal of Man-Machine Studies, 30, pp. 639-668.

....browsing: contextual, behavioral, motivational, cognitive, and resource based. The work reported here fits best with what they call the library, information science, and information retrieval traditions: browsing for unplanned discovery and as a problem solving technique. Thompson and Croft [12] address the benefits and requirements of browsing, and point out that browsers need a rich set of links, a flexible user interface, and search capabilities. They have a graphical system in which document and concept neighborhoods can be browsed. Crouch et al. 5] discuss a similar system focused ....

R. H. Thompson and W. B. Croft. "Support for browsing in an intelligent text retrieval system," International Journal of Man-Machine Studies, Volume 30, 1989, pp. 639-668.

....[5, 92] Efforts to provide cluster based organizations for large collections have become a focus of user interface research. In this setting, the clusters of a collection are presented to the user, perhaps based on an initial query, for subsequent manual browsing and honing of initial results [85]. These techniques have been incorporated into user interfaces for search engines on the web. For example, many internet search engines provide this feature as an alternative method for examining the ranked list of web pages that result from a search. Kohonen feature maps [43] have been adapted to ....

R.H. Thompson and B.W. Croft, "Support for Browsing in an Intelligent Text Retrieval System," International Journal of Man - Machine Studies, 30(6): 639-668, 1989.

....help system offering alternative phrasings and related terms can be quite useful during query construction. For instance, knowledge concerning the existence of more specific or less specific phrasings for the original query can be a reason for changing this original query. As was mentioned in [1], the results of a search can only be as good as the description of the items being sought. Therefore we need adequate support during the query construction phase. In terms of Meron Kuhns[2] we need to minimize the semantic noise in the information need representation. As a starting point for ....

R.H. Thompson and W.B. Croft. Support for browsing in an intelligent text retrieval system. International Journal of Man-Machine Studies, 30(2):639--668, 1989.

....help the user get an overview of the collection, or, alternatively, find interesting documents faster, as once one document has been found it is easy to find others similar to it in the visualized space. There are four major techniques to visualize inter document similarities: document networks [8, 31], spring embeddings [3, 30] document clustering [1, 13, 18, 38] and self organizing maps [16, 20, 22] Of the four major techniques, only document clustering appears to be both fast enough and intuitive enough to require little training or adjustment time from the user [13, 37] Numerous document ....

R. H. Thompson and W. B. Croft, Support for browsing in an intelligent text retrieval system. International Journal of Man-Machine Studies, vol. 30(6), 1989, pp 639-668.

.... on a measure of each document s similarity to the original query [9] Some work with general queries and general web pages [8] while others are tailored to more focused tasks [10] A second class of tools provide graphical means for accessing data based, for example, on inter document similarity [1, 12, 4], relationships to fixed attributes [11, 6] and query term distribution patterns [5] A third approach is to cluster the collection of documents. For example, hierarchical agglomerative clustering (HAC) 13] finds cluster centroids and clusters based on similarity to those centroids. A recent ....

R. H. Thompson and B. W. Croft. Support for browsing in an intelligent text retrieval system. International Journal of Man-Machine Studies, 30(6):639--668, 1989.

....with the results of our approaches in the evaluation section. Other related work in the IR literature includes: ffl the Scatter Gather cluster based approach to corpus navigation[3, 4] ffl the studies on clustering algorithms and their applications in the context of query driven retrieval [7, 6, 9, 8]. Our detection methods are inspired by the Scatter and Gather paper[3] including the choice of the basic groupaverage clustering (GAC) algorithm. However clustering algorithms per se are not the major focus of this study, nor are the applications or evaluations in a query driven retrieval ....

R.H. Tohompson and B.W. Croft. Support for browsing in an intelligent text retrieval system. In International Journal of Man-Machine Studies, pages 30(6)639--668, 1989.

....methods. Retrieval methods used for software repositories can be divided into three categories: enumerated classification, faceted, and free text indexing [12] Hypertext systems have also been used, although some form of retrieval is necessary to make hypertext effective in large repositories [16, 29]. Enumerated classification is a well known retrieval method used by the Dewy Decimal system and the Computing Reviews Classification System [1] The appeal of a classification scheme is the ability to iteratively divide an information space into smaller pieces that reduces the amount of ....

R. H. Thompson, W. B. Croft, "Support for Browsing in an Intelligent Text Retrieval System," International Journal of Man-Machine Studies, 30, 1989, pp. 639668.

....visualization technique is unique to VQE and helps users to capture database schemata and guides them to formulate syntactically valid query expressions. Many techniques for visualization and interactive operations have been proposed in the studies of information retrieval systems. I 3 R [ Thompson and Croft, 1989 ] is a distributed expert system based on a blackboard model. It configures a space by calculating the similarity among documents. BEAD [ Chalmers and Chitson, 1992 ] also configures a document space by calculation based on the analogy of potential fields in physics. These two systems don t ....

R. H. Thompson and W. B. Croft. Support for browsing in an intelligent text retrieval system. Int. Journal of Man-Machine Studies, 30:639 -- 668, 1989.

....for objectives of eliminating the imprecision problem connected with the application of a unique type of criterion and giving more liberty to the user in his retrieval. By the way, these were also the objectives of the IR3 system, in which multiple proximity criteria between documents were used [21]. 1.2 Connectionist models used in documentary systems have advantages and deficiencies The use of a connectionist model for the design of a documentary database interrogation system has a certain number of important general advantages. Due to the fact that it allows the adoption of unique ....

R. H. Thompson and W. B. Croft. Support for browsing in an intelligent text retrieval system. IJMMS, 30:639--668, 1989.

....mythology while Ares is the God of War in Greek mythology could be of great help. It is up to user to select the query which has to be submitted for evaluation. He or she makes the choice to look for slides depicting one of the gods, both of the gods, or deities in general. As was mentioned in [23], the results of a search can only be as good as the description of the items being sought. Therefore we need adequate support during the query construction phase, which is depicted on the left hand side of Figure 1. In terms of Meron Kuhns[16] we need to minimize the semantic noise in the ....

R.H. Thompson and W.B. Croft. Support for browsing in an intelligent text retrieval system. International Journal of Man-Machine Studies, 30(2):639--668, 1989.

....(www.aaai.org) All rights reserved. pages (Mauldin 1995) while others are tailored to more focused tasks (Shakes, Langheinrich, Etzioni 1997) A second class of tools provide graphical means for accessing data based, for example, on inter document similarity (Chalmers Chitson 1992; Thompson Croft 1989; Fowler, Fowler, Wilson 1991) relationships to fixed attributes (Spoerri 1993; Korfhage 1991) and query term distribution patterns (Hearst 1995) A third approach is to cluster the collection of documents. For example, hierarchical agglomerative clustering (HAC) Willet 1988) finds cluster ....

Thompson, R. H., and Croft, B. W. 1989. Support for browsing in an intelligent text retrieval system. International Journal of ManMachine Studies 30(6):639--668.

No context found.

Thompson R.H., Croft W.B. (1989): Support for browsing in an intelligent text retrieval system, in the International Journal of Man Machine Studies, 30: 639-668.

No context found.

R. H. Thompson and W. B. Croft. Support for browsing in an intelligent text retrieval system. International Journal of Man-Machine Studies, 30:639--668, 1989.

No context found.

Thompson, R.H. and Croft, W.B. "Support for Browsing in an Intelligent Text Retrieval System" International Journal of Man-Machine Studies, 30, pp. 639668, 1989

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC