Pirolli, P., Schank, P., Hearst, M., and Diehl, Scatter/Gather Browsing Communicates the Topic Structure of a Very Large Text Collection, in Proceedings of CHI'96 (Vancouver BC, April 1996), ACM Press, 213-220.  Home/Search   Document Not in Database   Summary   ACM   TOC   Related Articles   Check

....explore and comprehend collections of items. SenseMaker [1] focuses on supporting users in the contextual evolution of their interest in a topic. It attempts to make it easy to evolve a collection, e.g. expanding it through queryby example or limiting it by applying a filter. Scatter Gather [11] supports the browsing of large text collections, allowing users to iteratively reveal topic structure and locate desirable documents. WebBook and WebForager [6] allow users to define, visualize, and manipulate groups of related web pages. More relevant to the concerns of this paper are ....

Pirolli, P., Schank, P., Hearst, M., and Diehl, Scatter/Gather Browsing Communicates the Topic Structure of a Very Large Text Collection, in Proceedings of CHI'96 (Vancouver BC, April 1996), ACM Press, 213-220.

....displayed on the screen. The windows themselves constitute clusters of documents having common significant words. Under the words are displayed the titles of the first three documents. The user can broaden the information (scatter) or reduce information (gather) by means of appropriate commands (Pirolli,1996) (Pitkow Pirolli,1996) An interesting project in the field of information display is the WebBook. In this project an attempt was made to present pages of information in a metaphor of a printed book. The user can page a 3 dimensional graphic object resembling a book and arrive quickly at the ....

....The windows themselves constitute clusters of documents having common significant words. Under the words are displayed the titles of the first three documents. The user can broaden the information (scatter) or reduce information (gather) by means of appropriate commands (Pirolli,1996) Pitkow Pirolli,1996). An interesting project in the field of information display is the WebBook. In this project an attempt was made to present pages of information in a metaphor of a printed book. The user can page a 3 dimensional graphic object resembling a book and arrive quickly at the page which interests him ....

Pirolli, P. et al, (1996). Scatter/ gather browsing communicates the topic structure of a very large text collection. CHI '96 Proceedings, New York: ACM, 213-220.

....collections in which there are a number of documents having common significant words. Under the words are displayed the titles of the first three documents. The user is given the possibility of broadening the information (scatter) or to reduce information (gather) by means of appropriate commands [12] [13] This system has the advantage in everything relating to browsing a large unfamiliar database of documents. In the Butterfly user interface, results are displayed in a three dimensional manner where at the head of the butterfly items are displayed which identify the document, on the left ....

P. Pirolli et al. Scatter/gather browsing communicates the topic 15 structure of a very large text collection. CHI '96 Proceedings, New York: ACM 213-220. (1996)

....which made it possible for the user to introduce a series of desired words, following which the system produces a small set of documents which fulfill the search criteria. The results of the study showed an advantage to the Scatter Gather method for tasks which look for a specific document [21]. Another advantage of using this approach was found when there was a need to leaf through large, unknown databases. Expansion through producing groups of documents which have a common subject was shown by Pitkow and Pirolli. According to their method, if two different documents (a and b) are ....

Pirolli, P. et al., Scatter/gather browsing communicates the topic structure of a very large text collection. Proceedings of CHI '96, ACM, (1996), 213-220.

.... [Manning, Sch utze 1999] Several scientists have proposed information retrieval algorithms for the Web which are based on analysis of hyperlink structures [Botafogo et al. 1992] Carriere, Kazman 1997] Chakrabarti et al. 1998a] Chakrabarti et al. 1998b] Frisse 1988] Kleinberg 1998] [Pirolli et al. 1996a] Rivlin et al. 1994] A simple means of measuring the quality of a Web page proposed by [Carriere, Kazman 1997] is to count the number of pages which have pointers to the page is used in the WebQuery system and the Rankdex search engine 91 . Another search engines which use link infomation ....

Pirolli, P., Schank, P., Hearst, M., Diehl, C., \Scatter/Gather browsing communicates the topic structure of a very large text collection", Proc. ACM SIGCHI, CHI'96 Vancouver, B.C., Canada, ACM Press, NY (Apr. 13-18, 1996): www.acm.org/sigchi/chi96/proceedings/papers/Pirolli/pp txt.htm

....the formulation of database like queries over published XML document types [Mace, 1998] Finally, from the HCI and graphics research community have come a whole range of interactive visualisation techniques. Some of these are addressed at established problems in IR (e.g. scatter gather browser [Pirolli, 1996] and 3D representations of document spaces such as Cone Trees [Robertson, 1991] and populated information terrains [Benford, 1994] databases (e.g. visual query languages for relational [Haw, 1994; Papantonakis, 1995; Benzi, 1998] and object databases [Chavda, 1997; Murray, 1998] and ....

P. Pirolli, P. Schank, M. Hearst and D. C (1996). Scatter/gather browsing communicates the topic structure of a very large text collection. Proceedings of CHI'96, Vancouver, ACM Press. pp. 213--220.

....collections of information resources. SenseMaker [2] focuses on supporting users in the contextual evolution of their interest in a topic. They attempt to make it easy to evolve a collection, e.g. expanding it by query by example operations or limiting it by applying a filter. Scatter Gather [11] supports the browsing of large collections of text, allowing users to iteratively reveal topic structure and locate desirable documents. Card, Robertson, and York [3] describe the WebBook, which uses a book metaphor to group a collection of related web pages for viewing and interaction, and the ....

Pirolli, P., Schank, P., Hearst, M., and Diehl, Scatter/Gather Browsing Communicates the Topic Structure of a Very Large Text Collection, in Proceedings of CHI'96 (Vancouver BC, April 1996), ACM Press, 213-220.

....collections of information resources. SenseMaker [2] focuses on supporting users in the contextual evolution of their interest in a topic. It attempts to make it easy to evolve a collection, e.g. expanding it by query by example operations or limiting it by applying a filter. Scatter Gather [25] supports the browsing of large collections of text, allowing users to iteratively reveal topic structure and locate desirable documents. Hightower et al. [12] addressed the observation that users often return to previously visited pages. They used Pad [3] to implement PadPrints, browser ....

Pirolli, P., Schank, P., Hearst, M., and Diehl, Scatter/Gather Browsing Communicates the Topic Structure of a Very Large Text Collection, in Proceedings of CHI'96 (Vancouver BC, April 1996), ACM Press, 213-220.

....relies on description logics for consistency maintenance. It is also worth noting that others support the notion of topic sets as a valuable asset for search in constrained domains. One company s product line [18] consists of topic sets for various search engine languages. Additionally others [12] support the notion of a hierarchy for site exploration. Evidence phrases will undoubtedly reduce the no matching document problem and thus increase potentially 3 For the official description logic home page, see http: dl.kr.org dl . relevant retrievals. The information retrieval[14] ....

P. Pirolli, P. Schank, M. Hearst, and C. Diehl. Scatter /Gather Browsing Communicates the Topic Structure of a Very Large Text Collection. In CHI 1996 Conference Proceedings, Vancouver, BC, Canada, April, 1996.

....such as Scatter Gather [CKPT92] HP96] have been used to enable entire collections and query retrieval results to be browsed more easily. Work in this area has shown that document clustering is often an effective way to give the user a better sense of the topics present in a set of documents [PSHD96] The success of such systems often hinges on the effectiveness of the clustering methods employed. There is a long history of empirical work in document clustering, an excellent survey of which is given by Willett [Wil88] Indeed, the description of Scatter Gather is very specific about the ....

Peter Pirolli, Patricia Schank, Marti Hearst, and Christine Diehl. Scatter/gather browsing communicates the topic structure of a very large text collection. In Proceedings of CHI, 1996.

....dimension for the view. This organizing dimension can be iteratively changed by the user. Bundling has already been used successfully in innovative interfaces for exploring relational databases [4, 13] where it is known as grouping or aggregation) as well as for browsing databases of full text [12] (where it is known as clustering) The concept of duplicate detection has also long been important in the relational database field. As shown in Table 1, the SenseMaker approach differs from most relational database and Information Retrieval (IR) clustering interfaces in the extent to which it ....

Pirolli, P., Schank, P., Hearst, M., and Diehl, C. Scatter/ Gather browsing communicates the topic structure of a very large text collection, in Proceedings of CHI '96 (Vancouver Canada, May 1996), ACM Press, 213-20.

....a Web locality as a complex abstract space in which WWW pages of different functional categories or types are arranged. These functional categories might be defined by a user s specific set of interests, or the categories might be extracted from the collection itself through inductive technologies [13]. An example category might be organizational home page. Typical members of the category would describe an organization and have links to many other Web pages, providing relevant information about the organization, its divisions or departments, summaries of its purpose, and so on. We specified a ....

Pirolli, P., P. Schank, M. Hearst, and C. Diehl. Scatter/Gather browsing communicates the topic structure of a very large text collection. in Conference on Human Factors in Computing Systems, CHI-96. (1996, Association for Computing Machinery. pp.

....and a set of heuristics that select productions in ways that achieve adaptive information foraging behavior. Since ACT IF is a behaving production system, we may compare traces of its behavior against those of human information foragers. We present a production system model of data collected in Pirolli et al. 1996) in a study of a information system for very large collections of full text documents, called Scatter Gather. The production system model operates by heuristics that instantiate the information diet and information patch models. This model may be considered as an extensive revision of the ACT R ....

....SG window some clusters have been selected Scatter Gather the window DO DISPLAY TITLES Goal is to process SG window some clusters have been selected Display the titles in the window Figure 10. Production rules used in the ACT IF model of the Scatter Gather protocols obtained in Pirolli et al. 1996). Information Foraging 40 Assessment of Information Scent by Spreading Activation Spreading activation provides the mechanism modeling people s assessment of information scent. In this section, we describe the spreading activation model of information assessment and describe how spreading ....

[Article contains additional citation context not shown here]

Pirolli, P., Schank, P., Hearst, M., & Diehl, C. (1996). Scatter/Gather browsing communicates the topic structure of a very large text collection.

.... clustering algorithm employed may also affect the topics seen, however most clustering algorithms will yield roughly Scatter Gather may be applied to an entire corpora, in which case static off line computations may be exploited to speed dynamic on line clustering [5] We have recently shown [18] that this use of Scatter Gather successfully conveys some of the content and structure of the corpus. However, that study also showed that Scatter Gather was less effective than a standard similarity search when the subjects were provided with a query. That is, subjects constrained to only ....

....documents. In fact, the top ranked cluster almost always contains at least 50 of the relevant documents retrieved, and usually a much larger percentage. The third, fourth, and fifth ranked clusters usually contain 10 or fewer. 6 5 Topic 201 has been thrown out by the TREC sponsors. 6 See [18] for an earlier discussion of this behavior in a somewhat different setting. Figure 2: Scatter Gather results on auto, car, vehicle and electric with a cutoff of 250. Figure 3: Scatter Gather results on auto, car, vehicle and safety with a cutoff of 250. bucket mean expected t value 1 20 .802 ....

Peter Pirolli, Patricia Schank, Marti A. Hearst, and Christine Diehl. Scatter/gather browsing communicates the topic structure of a very large text collection. In Proceedings of the ACM SIGCHI Conference on Human Factors in Computing Systems, Vancouver, WA, May 1996. ACM.

....and refinement. This is not to say that query reformulation is unimportant, but rather that we chose not to emphasize that for these experiments. Furthermore, we have evidence that the Scatter Gather interface helps users determine alternative terms by which to augment their queries, see [20]. Figure 3 shows the entire interface when set in TileBar mode. We will use query 216 as a running example with which to illustrate the components of the interface: What research is ongoing to reduce the effects of osteoporosis in existing patients as well as prevent the disease occurring in ....

Peter Pirolli, Patricia Schank, Marti A. Hearst, and Christine Diehl. Scatter/gather browsing communicates the topic structure of a very large text collection. In Proceedings of the ACM SIGCHI Conference on Human Factors in Computing Systems, Vancouver, WA, May 1996. ACM. (to appear).

No context found.

Pirolli P., Schank P., Hearst M.A., Diehl C. (1996): Scatter/Gather Browsing Communicates the Topic Structure of a Very Large Text Collection, in the Proceedings of ACM CHI Conference on Human Factors in Computing Systems: 213-220.

No context found.

P. Pirolli, P. Schank, M. Hearst, and C. Diehl. Scatter/Gather Browsing Communicates the Topic Structure of a Very Large Text Collection. In Human Factors in Computing Systems CHI '96, pages 213-220.

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