Belew, R.K.: Finding out about : a cognitive perspective on search engine technology and the WWW (2001)
0.04
0.03997163 = product of:
0.1399007 = sum of:
0.06297635 = weight(_text_:wide in 3346) [ClassicSimilarity], result of:
0.06297635 = score(doc=3346,freq=12.0), product of:
0.1312982 = queryWeight, product of:
4.4307585 = idf(docFreq=1430, maxDocs=44218)
0.029633347 = queryNorm
0.47964367 = fieldWeight in 3346, product of:
3.4641016 = tf(freq=12.0), with freq of:
12.0 = termFreq=12.0
4.4307585 = idf(docFreq=1430, maxDocs=44218)
0.03125 = fieldNorm(doc=3346)
0.039451245 = weight(_text_:web in 3346) [ClassicSimilarity], result of:
0.039451245 = score(doc=3346,freq=16.0), product of:
0.09670874 = queryWeight, product of:
3.2635105 = idf(docFreq=4597, maxDocs=44218)
0.029633347 = queryNorm
0.4079388 = fieldWeight in 3346, product of:
4.0 = tf(freq=16.0), with freq of:
16.0 = termFreq=16.0
3.2635105 = idf(docFreq=4597, maxDocs=44218)
0.03125 = fieldNorm(doc=3346)
0.010677892 = weight(_text_:information in 3346) [ClassicSimilarity], result of:
0.010677892 = score(doc=3346,freq=14.0), product of:
0.052020688 = queryWeight, product of:
1.7554779 = idf(docFreq=20772, maxDocs=44218)
0.029633347 = queryNorm
0.20526241 = fieldWeight in 3346, product of:
3.7416575 = tf(freq=14.0), with freq of:
14.0 = termFreq=14.0
1.7554779 = idf(docFreq=20772, maxDocs=44218)
0.03125 = fieldNorm(doc=3346)
0.026795205 = weight(_text_:retrieval in 3346) [ClassicSimilarity], result of:
0.026795205 = score(doc=3346,freq=10.0), product of:
0.08963835 = queryWeight, product of:
3.024915 = idf(docFreq=5836, maxDocs=44218)
0.029633347 = queryNorm
0.29892567 = fieldWeight in 3346, product of:
3.1622777 = tf(freq=10.0), with freq of:
10.0 = termFreq=10.0
3.024915 = idf(docFreq=5836, maxDocs=44218)
0.03125 = fieldNorm(doc=3346)
0.2857143 = coord(4/14)
- Abstract
- The World Wide Web is rapidly filling with more text than anyone could have imagined even a short time ago, but the task of isolating relevant parts of this vast information has become just that much more daunting. Richard Belew brings a cognitive perspective to the study of information retrieval as a discipline within computer science. He introduces the idea of Finding Out About (FDA) as the process of actively seeking out information relevant to a topic of interest and describes its many facets - ranging from creating a good characterization of what the user seeks, to what documents actually mean, to methods of inferring semantic clues about each document, to the problem of evaluating whether our search engines are performing as we have intended. Finding Out About explains how to build the tools that are useful for searching collections of text and other media. In the process it takes a close look at the properties of textual documents that do not become clear until very large collections of them are brought together and shows that the construction of effective search engines requires knowledge of the statistical and mathematical properties of linguistic phenomena, as well as an appreciation for the cognitive foundation we bring to the task as language users. The unique approach of this book is its even handling of the phenomena of both numbers and words, making it accessible to a wide audience. The textbook is usable in both undergraduate and graduate classes on information retrieval, library science, and computational linguistics. The text is accompanied by a CD-ROM that contains a hypertext version of the book, including additional topics and notes not present in the printed edition. In addition, the CD contains the full text of C.J. "Keith" van Rijsbergen's famous textbook, Information Retrieval (now out of print). Many active links from Belew's to van Rijsbergen's hypertexts help to unite the material. Several test corpora and indexing tools are provided, to support the design of your own search engine. Additional exercises using these corpora and code are available to instructors. Also supporting this book is a Web site that will include recent additions to the book, as well as links to sites of new topics and methods.
- LCSH
- World Wide Web / Computer programs
Web search engines
- RSWK
- Suchmaschine / World Wide Web / Information Retrieval
- Subject
- Suchmaschine / World Wide Web / Information Retrieval
World Wide Web / Computer programs
Web search engines
Bartell, B.T.; Cottrell, G.W.; Belew, R.K.: Optimizing similarity using multi-query relevance feedback (1998)
0.01
0.0052989167 = product of:
0.037092414 = sum of:
0.0071344664 = weight(_text_:information in 1152) [ClassicSimilarity], result of:
0.0071344664 = score(doc=1152,freq=4.0), product of:
0.052020688 = queryWeight, product of:
1.7554779 = idf(docFreq=20772, maxDocs=44218)
0.029633347 = queryNorm
0.13714671 = fieldWeight in 1152, product of:
2.0 = tf(freq=4.0), with freq of:
4.0 = termFreq=4.0
1.7554779 = idf(docFreq=20772, maxDocs=44218)
0.0390625 = fieldNorm(doc=1152)
0.029957948 = weight(_text_:retrieval in 1152) [ClassicSimilarity], result of:
0.029957948 = score(doc=1152,freq=8.0), product of:
0.08963835 = queryWeight, product of:
3.024915 = idf(docFreq=5836, maxDocs=44218)
0.029633347 = queryNorm
0.33420905 = fieldWeight in 1152, product of:
2.828427 = tf(freq=8.0), with freq of:
8.0 = termFreq=8.0
3.024915 = idf(docFreq=5836, maxDocs=44218)
0.0390625 = fieldNorm(doc=1152)
0.14285715 = coord(2/14)
- Abstract
- We propose a novel method for automatically adjusting paprameters in ranked-output text retrieval systems to improve retrieval performance. A renaked-output text retrieval system implements a ranking function which orders documents, placing documents estimated to be more relevant to the user's query before less relevant ones. The systems adjusts its parameters to maximize the match between the systems's document ordering and a target ordering. The target ordering is typically given by user feedback on a set of sample queries, but is more generally any document preference relation. We demonstrate the utility of the approach by using it to estimate a similarity measure (scoring the relevance of documents to queries) in a vector space model of information retrieval. Experimental results using several collections indicate that the approach automatically finds a simimilarity measure which performs equivalently to or better that all 'classic' similarity measures studied. It also performs within 1% of an estimated optimal measure (found by exhaustive sampling of the similarity measures). The method is compared to two alternative methods: a perceptron learning rule motivated by Wong and Yao's (1990) Query Formulation method, and a Least Squared learning rule, motivated by Fuhr and Buckley's (1991) Probabilisitc Learning approach. Though both alternatives have useful characteristics, we demonstrate empirically that neither can be used to estimate the parameters of the optimal similarity measure
- Source
- Journal of the American Society for Information Science. 49(1998) no.8, S.742-761
