Shape recovery: A visual method for evaluation of information retrieval experiments

Rorvig, Mark E.; Fitzpatrick, Steven J.

doi:10.1002/1097-4571(2000)9999:9999<::aid-asi1032>3.3.co;2-#

Cited by 5 publications

(7 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Visualization Information Retrieval Interface conferences (VIRI) from ACM‐SIGIR (Asssociation of Computing Machinery‐Special Interest Group on Information Retrieval) generated interesting work on visual methods of evaluation (Havre, Hetzler, Perrine, Jurrus, & Miller, 2001; Korfhage, Lin, & Dubin, 1995; Rorvig, 1999a; Rorvig & Hemmje, 1996; Rorvig & Fitzpatrick, 1998; Rorvig et al, 1998; White & McCain, 1997; Zhang & Nguyen, 2005). Shape recovery (Rorvig & Fitzpatrick, 2000) is used to visually examine four treatments. For each treatment, the coefficients are used to rank terms and the top 200 are selected to construct document surrogates.…”

Section: Methodsmentioning

confidence: 99%

“…The first coefficient is based on the chi‐square statistic and the second on a discriminative function. The study uses a visualization technique, the shape recovery method (Ma, 2002; Rorvig & Fitzpatrick, 2000; Zhang & Nguyen, 2005), to discover variations among experimental treatments.…”

Section: Introductionmentioning

confidence: 99%

“…The angular distance, measured from the origin point of the document space, can also be used to derive similarity. A series of experiments with several metrics found that the cosine of the pairwise document angle consistently provided acceptable results for document clustering (Rorvig, 1999a; Rorvig & Fitzpatrick, 1998; Rorvig & Fitzpatrick, 2000).…”

Section: Introductionmentioning

confidence: 99%

“…Maintaining the rank of each data point, the procedure calculates all their coordinates. The methodology was called shape recovery , but some distortion is introduced by the scaling process (Rorvig & Fitzpatrick, 2000). In general, documents are expected to satisfy the hypothesis that relevant documents have greater cohesiveness and are closer to each other than to nonrelevant documents, which disperse into a periphery (Van Rijsbergen, 1989; Voorhees, 1985).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Using the shape recovery method to evaluate indexing techniques

Oyarce

2008

J. Am. Soc. Inf. Sci.

View full text Add to dashboard Cite

Text representation, central to information processing, must be descriptive and discriminative. Although some of the many techniques to construct document representations may outperform others for certain tasks, no one is consistently better than others. Representations are still problematic. Evaluation techniques are needed to penetrate foundational questions about term behavior in representation. A study that applies the shape recovery analysis method is reported here as an evaluative tool to compare different indexing schemes. Three weight coefficients are used to rank indexing terms and are compared to the documents' full text. Two of the weight coefficients are novel and the third relies on the chi-squared distribution. Multidimensional scaling reduces the dimensional space of the document surrogates into a two-dimensional Cartesian space. Ten concentric circles evenly separated at 10% intervals of relevant data points starting at the centroid are used to construct a precision-recall curve. ANOVA is used for a straightforward computation of the 4 ؋ 11 matrix of test data to see whether the four treatments yield the same P-R result. A post hoc HSD Tukey multiple comparisons test among pairwise treatments is also used to discover homogeneous groups. The findings show the value of the methodology to study term weighting schemes, and their descriptiveness and discriminative power, as well as the potential strength of the novel coefficients introduced.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Using the shape recovery method to evaluate indexing techniques

Oyarce

2008

J. Am. Soc. Inf. Sci.

View full text Add to dashboard Cite

show abstract

“…The bull's eye has been used on many occasions to ease understanding of the search results (e.g. Rorvig and Fitzpatrick 2000, Sutcliffe et al 2000).…”

Section: Methodsmentioning

confidence: 99%

A Search Interface for an SDI: Implementation and Evaluation of Metadata Visualization Strategies

Aditya¹,

Kraak²

2007

Transactions in GIS

View full text Add to dashboard Cite

Search interfaces for data discovery are essential in the context of a Geospatial Data Infrastructure. This paper discusses the development of a search interface, aimed at effective searches and an integration of a catalogue function and thematic mapping. For this purpose, the Resources Description Framework (RDF) is used to store the summaries of metadata of geospatial contents. These summaries are indexed based on their topics. Their topic categories are expressed using the RDF schema. The RDF queries and the transformation of results play a prominent role in the visualization of search results as a table, a thumbnail, and a bull's‐eye display with possibilities to project footprints of results on top of a basic map and thematic layers. To highlight their matches to the search terms, the Hausdorff distance and the closeness degree were applied respectively to determine the geographic and thematic relevance. Feedback from participants testing the prototype indicated the benefits of the table and thumbnail displays as well as the possibility to combine these displays with metadata and thematic mapping for data discovery. The feedback also confirmed some design challenges in the depictions of relevance and symbolization of metadata.

show abstract