J. Dean and M.R. Henzinger. Finding related pages in the world wide web. In Proceedings of the Eighth International World Wide Web Conference WWW-1999, Toronto, May 1999.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....if the seed node was actually a page we were interested in, the MAX algorithm would have discovered a list of pages related to the seed page. Finding pages related to a query web page is a standard feature of most modern search engines. This is an active research area with a growing literature [12, 7, 10]. The current techniques use content analysis, link analysis, or a combination of both. We propose the use of the MAX algorithm as a tool for discovering web pages, related to a query web page. The idea of using link analysis algorithms for fining related pages was fist suggested by Kleinberg ....

....or a combination of both. We propose the use of the MAX algorithm as a tool for discovering web pages, related to a query web page. The idea of using link analysis algorithms for fining related pages was fist suggested by Kleinberg [12] This idea was later enhanced by Dean and Henzinger [7]. Given a query page q Dean and Henzinger construct a vicinity graph around q as follows. Let B denote a step, that follows a link backwards, and let F denote a step that follows a link forward. Starting from the query page q they collect a set of pages that can be reached by following B, F , BF ....

J. Dean and M. R. Henzinger. Finding related pages in the world wide web. In Proceedings of the Eighth International World-Wide Web Conference (WWW9), 1999.

....[17, 18] For instance, Suel and Yuan [18] observe that on average, around three quarters of the links from a page are to other pages on the same domain host. The last observation about page similarity is the underlying basis of several algorithms for Web search and for finding related Web pages [6, 10]. Given these observations, we attempt to construct a partition that has the following properties: Property 1: Pages with similar adjacency lists are grouped together, as much as possible. Observation 1 predicts the existence of groups of pages with similar adjacency lists. If such pages are ....

J. Dean and M. R. Henzinger. Finding related pages in the World Wide Web. Computer Networks, 31(11--16):1467-- 1479, 1999.

....structure of the Web. Prominent examples are the HITS algorithm for determining good hub and authority pages [39] and the PageRank algorithm for approximating the probability that a page is visited by a random surfer on the Web [9] which have been successfully applied to various web mining tasks [13, 23, 7]. Prerequisite for this type of research is the availability of extended document indexing techniques that allow fast access to both outgoing and incoming hyperlinks of a page [6] An excellent overview of recent research in the area, which is nowadays referred to as web mining, can be found in ....

Jeffrey Dean and Monika R. Henzinger. Finding related pages in the World Wide Web. In A. Mendelzon, editor, Proceedings of the 8th International World Wide Web Conference (WWW-8), pages 389--401, Toronto, Canada, 1999.

....approach (used commonly in push type of systems e.g. 32] information is presented to the user according to his her preference specifications. This is clearly inappropriate for our problem. Web resource discovery aims to find Web pages relevant to users requests or interests (e.g. 9] [13][20] 21] 27] This approach uses techniques such as link analysis, link topologies, text classification methods to find relevant pages. The pages can also be grouped into authoritative pages, and hubs. However, relevant pages, which are often judged by keywords, are not sufficient for our ....

#Dean, J. & Henzinger, M.R. Finding related pages in the World Wide Web. WWW8, 1999.

....[12, 30] For instance, Suel and Yuan [30] observe that on average, around three quarters of the links from a page are to other pages on the same domain host. The last observation about page similarity is the underlying basis of several algorithms for Web search and for finding related Web pages [7, 25]. Given these observations, we attempt to construct a partition that has the following properties: Property 1: Pages with similar adjacency lists are grouped together, as much as possible. Observation 1 predicts the existence of groups of pages with similar adjacency lists. If such pages are ....

Jeffrey Dean and Monika R. Henzinger. Finding related pages in the World Wide Web. Computer Networks, 31(11--16):1467--1479, 1999. 25

....to by many good hubs is an even better authority. HITS repeatedly updates the hub and authority scores so that documents with high authority scores are expected to have relevant content, whereas documents with high hub scores are expected to contain links to relevant content. Dean and Henzinger [6] compared the performance of a variant of the HITS algorithm and that of the Co Citation algorithm in the task of finding related, or similar, web pages to a given URL. While the HITS based algorithm did outperform CoCitation, both algorithms where judged in their user trials to have performed ....

....outperform both Co Citation and HITS. This effect is even more apparent when one differentiates between recommendations that are rated very similar and those that are merely rated similar . The fact that HITS also under performs Co Citation is also at odds with the results of Dean Henzinger [6], whose Companion algorithm (a slightly modified version of HITS) had outperformed the Co Citation algorithm. To quantify the performance of the algorithms, we define the precision at r for a given algorithm to be the total number of recommendations receiving a positive score from the users ....

J. Dean and M. R. Henzinger. Finding related pages in the world wide web. In Proceedings of the Eighth International World Wide Web Conference, pages 389--401, Toronto, Canada, May 1999.

....Table 4. Model AveP P 10 P 20 Run ID d 0 0.1913 0.3000 0.26 0 LA2 1 d 1 0.196 0.3000 0.2745 LA2 2 d 3 0.1977 0.276 0.2713 LA2 3 s 3 0.176 0.246 0.2330 LA2 4 baseline 0.1211 0.1808 0.1745 Model AveP P 10 P 20 Run ID d 0 0.2754 0.2022 0.1478 SA2 1 d 1 0.2935 0.2289 0. 16 9 SA2 2 d 3 0.2905 0.2311 0.16 1 SA2 3 s 3 0.246 0.2133 0.16 3 SA2 4 baseline 0.21480.1756 0.136 Table 4. Evaluation results of II A2 In the tables, the model d n means that the relevant document rdoc[1] is used in duplicate and n pseudo relevant documents are used for query expansion. The model ....

....Model AveP P 10 P 20 Run ID d 0 0.1913 0.3000 0.26 0 LA2 1 d 1 0.196 0.3000 0.2745 LA2 2 d 3 0.1977 0.276 0.2713 LA2 3 s 3 0.176 0.246 0.2330 LA2 4 baseline 0.1211 0.1808 0.1745 Model AveP P 10 P 20 Run ID d 0 0.2754 0.2022 0.1478 SA2 1 d 1 0.2935 0.2289 0.16 9 SA2 2 d 3 0.2905 0. 2311 0.16 1 SA2 3 s 3 0.246 0.2133 0.16 3 SA2 4 baseline 0.21480.1756 0.136 Table 4. Evaluation results of II A2 In the tables, the model d n means that the relevant document rdoc[1] is used in duplicate and n pseudo relevant documents are used for query expansion. The model s n means ....

[Article contains additional citation context not shown here]

J. Dean and M. R. Henzinger. Finding related pages in the World Wide Web. In Proceedings of the 8th World-Wide Web Conference, 1999.

....that use link analysis to make semantic inferences is a correlation between the graph topology of the Web and the meaning of pages, or more precisely the conjecture that one can infer what a page is about by looking at its neighbors. Such a conjecture has been implied or stated in various forms [21, 13, 2, 6, 9, 8, 16, 23, 24]. 2.1 Similarity measures The basic idea of semantic mapping is to quantitatively measure the relationships between content, link, and semantic topology of the Web at a fine level of resolution by building maps of #s in a space where the coordinates are given by #c and # l , i.e. by similarity ....

J. Dean and M. Henzinger. Finding related pages in the World Wide Web. Computer Networks, 31(11--16):1467--1479, 1999.

....Association rules. 1 Introduction The problem of modelling and predicting a user s accesses on a Web site has attracted a lot of research interest. It has been used [20] to improve the Web performance through caching [2, 12] and prefetching [34, 22, 35, 29, 39, 40] recommend related pages [19, 38], improve search engines [11] and personalize browsing in a Web site [39] Nowadays, the improvement of Web performance is a very significant requirement. Since the Web s popularity resulted in heavy traffic in the Internet, the net effect of this growth was a significant increase in the user ....

J. Dean and M. Henzinger. Finding related pages in the World Wide Web. In Proceedings of the World Wide Web Conference (WWW'99), pages 1467-1479, 1999.

....and its predicate is the presence of the item in the local index, thus, a peer can participates in a rule only if it shares the corresponding item. Our underlying intuition, taken from extensive previous research in the Data Mining and Text Retrieval communities (see, e.g. 16] 8] 7] 20] [15]) is that, on average, peers that share items (in particular rare items) are more likely to satisfy each other s queries than random peers. More precisely, search using possession rules exploits presence of pairwise co location associations between items. Beyond the resolution of the search, ....

J. Dean and M. R. Henzinger. Finding related pages in the world wide web. WWW8 / Computer Networks, 31(11-16):1467--1479, 1999.

....to process large volume of Web log data to identify interesting Web access patterns. Such patterns not only lead to a better understanding of Web user behavior but also are valuable for various Web applications intended for characterizing the Web structure [11, 2] improving searching on the Web [10, 15, 17], and optimizing the Web performance via caching and prefetching [4, 12, 18, 21, 25] The dramatic increase of HTTP tra#c on the Internet has resulted in wide spread use of large proxy servers as critical Internet infrastructure components. Web access logs collected at these proxy servers provide ....

Je#rey Dean and Monika R. Henzinger. Finding related pages in the World Wide Web. Computer Networks (Amsterdam, Netherlands: 1999.

.... implemented Web systems, including search and meta search engines (e.g. Inf02b, SE95, SE97, DH97c, HD97, Inf02a, ZE98, ZE99, LG98b, LG98a, Goo02, Alt02, Ink02, Fas02, McB94, Lyc02b] focused crawlers (e.g. CGMP98, CvdBD99, BSHJ 99, Men97, MB00, Lie97, RM99] linkage analyzers (e.g. BFJ96, DH99, KRRT99, FLGC02, CDR 99, BH98b, BP98, DGK IBM00, Kle98] and intelligent Web agents (e.g. AFJM95, JFM97, Mla96, Lie95, Lie97, MB00, BS97, LaM96, LaM97, Lie97, PP97, Dav99a] as we will describe below. While in this dissertation we are concerned with extracting information from Web ....

....trace needs to be collected (which we do, and use in experiments we report later in Chapter 6) Intuitions about Web content locality are so fundamental that in many cases they are not mentioned, even though without them the systems would fail to be useful. When mentioned explicitly (as in [MB00, DH99, GKR98, BS97, Kle98, BP98, CDR Ami98] their influence is measured implicitly, if at all. This chapter is an attempt to rectify the situation we wish to measure the extent to which these ideas hold. This chapter primarily addresses two topics: it examines the textual similarity of pages ....

[Article contains additional citation context not shown here]

Je#rey Dean and Monika R. Henzinger. Finding related pages in the World Wide Web. In Proceedings of the Eighth International World Wide Web Conference, pages 389--401, Toronto, Canada, May 1999.

....has a corresponding data item, and its predicate is the presence of the item in the local index, thus, a peer can participates in a rule only if it shares the corresponding item. Our underlying intuition, taken from extensive previous research in the Data Mining and Text Retrieval communities ([17, 9, 10, 8, 5, 22, 16]) is that, on average, peers that share items (in particular rare items) are more likely to satisfy each other s queries than random peers. More precisely, search using possession rules exploits presence of pairwise co location associations between items. Beyond the resolution of the search, ....

J. Dean and M. R. Henzinger. Finding related pages in the world wide web. WWW8 / Computer Networks, 31(11-16):1467--1479, 1999.

....of well known non members. Given the set of pages on some topic, a community is defined as a set of Web pages that link (in either direction) to more pages in the community than to the pages of outside community. The flow based approach can be used to guide the crawling of related pages. In [29], the Companion algorithm is proposed to find the related pages of the seed pages presented by specializing the HITS algorithm 10 exploiting the weight of the links and the order of the links in a page. The Companion algorithm first builds a subgraph of the Web near the seed, and extracts the ....

Jeffrey Dean, and Monica R.Henzinger, Finding related pages in the world wide web. in proc. of 8th WWW conference, 1999.

....The total size of the logs accumulated in one year is about 500 GB. Then we will also report the application of link analysis to extract web communities. A web community is a collection of web pages created by individuals or any kind of associations that have a common interest on a specific topic[2, 1]. We proposed a technique to create a web community chart, that connects related web communities[5] The scale of exponentially growing WWW has posed new challenges to the research community. We will discuss some problems we face on large scale web mining in order to make usable applications of ....

....Result page of query expansion system 3 Cyber Community Mining Link analysis has been used to reveal the structure of the web, and to rank search results. Recently some researches also point out its potential to extract web communities, interconnected web pages whose content with similar interest[2, 1]. Our approach goes further to visualize those web communities in the form of a navigatable chart. 3.1 Hyperlinks Archive Our data set for experiments is an archive of Japanese web pages. The archive includes about 17 million pages in the jp domain, or ones in other domains but written in ....

J. Dean, M.R. Henzinger "Finding related pages in the World Wide Web" In Proc of the 8th WWW Conf., 1999

....of well known non members. Given the set of pages on some topic, a community is defined as a set of Web pages that link (in either direction) to more pages in the community than to the pages of outside community. The flow based approach can be used to guide the crawling of related pages. In [28], the Companion algorithm is proposed to find the related pages of a seed pages presented by specializing the HITS algorithm exploiting link weighting and order of links in a page. The Companion algorithm first builds a subgraph of the Web near the seed, and extracts the authorities and hubs in ....

Jeffrey Dean, and Monica R.Henzinger, Finding re- lated pages in the world wide web. in proc. of 8th WWW conference, 1999.

....graph 1 2 3 and choose these as synonyms. For this last step we use a similarity measure between vertices in graphs that was introduced in [BHD, Hey01] The problem of searching synonyms is similar to that of searching similar pages on the web; a problem that is dealt with in [Kle99] and [DH99]. In these references, similar pages are found by searching authoritative pages in a subgraph focused on the original page. Authoritative pages are pages that are similar to the vertex authority in the structure graph hub authority: We ran the same method on the dictionary graph and obtained ....

....in the de nition of cow, neither does the latter appear in the de nition of the former. Thus, the methods described above cannot nd this word. Larger neighborhood graphs could be obtained either as Kleinberg does in [Kle99] for searching similar pages on the Web, or as Dean and Henziger do in [DH99] for the same purpose. However, such subgraphs are not any longer focused on the original word. That implies that our variation of Kleinberg s algorithm forgets the original word and may produce irrelevant results. When we use the vicinity graph of Dean and Henziger, we obtain a few interesting ....

Jerey Dean and Monika Rauch Henzinger.Finding related pages in the world wide web.WWW8 / Computer Networks, 31(11-16):1467-1479, 1999.

....community chart from thousands of seed pages on a broad topic. To identify com Authority Authority Hub Hub Hub Figure 1: Typical graph structure of hubs and authorities munities and to deduce relationships between communities, we use a modified version of a related page algorithm, Companion [6], which provides related pages to a given page. First, we extend the seed set by applying the algorithm to each page, and adding related pages into the seed set. Then we again apply the algorithm to each page in the extended seed set, and investigate how each page derives other pages as related ....

....found communities by extracting complete bipartite graphs that consist of authorities and hubs. HITS can also be used to find pages related to a given seed page. Finding related pages is similar to finding a community including the seed. Our method is based on a related page algorithm, Companion [6] proposed by Dean et al. which takes a seed page as an input, then outputs related pages to the seed. Dean et al. specialized HITS [9] for finding related pages, and improved the precision by exploiting link weighting and the order of links in a page. Companion first builds a subgraph of the Web ....

[Article contains additional citation context not shown here]

Jeffrey Dean and Monika R. Henzinger. Finding related pages in the World Wide Web. In Proceedings of the 8th World-Wide Web Conference, 1999.

....of well known non members. Given the set of pages on some topic, a community is defined as a set of web pages that link (in either direction) to more pages in the community than to the pages of outside community. The flow based approach can be used to guide the crawling of related pages. In [4], an approach to find the related pages of a seed pages presented by specializing the HITS algorithm exploiting link weighting and order of links in a page. Companion first builds a subgraph of the Web near the seed, and extracts authorities and hubs in the graph using HITS. The authorities are ....

Jeffrey Dean, and Monica R.Henzinger, Finding related pages in the world wide web. 8th international WWW conference, 1999.

....of well known non members. Given the set of pages on some topic, a community is defined as a set of web pages that link (in either direction) to more pages in the community than to the pages of outside community. The flow based approach can be used to guide the crawling of related pages. In [5], an approach to find the related pages of a seed pages presented by specializing the HITS algorithm exploiting link weighting and order of links in a page. Companion first builds a subgraph of the Web near the seed, and extracts authorities and hubs in the graph using HITS. The authorities are ....

Jeffrey Dean, and Monica R.Henzinger, Finding related pages in the world wide web. in Proc. 8th WWW, 1999.

....Table 5. Related hosts for www.airfrance.com. content on their site (e.g. specific books) rewarding them for the traffic sent through. In addition to citation we could also use co citation to identify affiliated or related parts of the web. 6.2. Relatedness by Co citation Dean and Henzinger [10] showed that cocitation analysis on the web graph works well for finding related web pages. Their best algorithm achieved a precision 10 of 0.4. We extended the idea to the hostgraph. Our approach is as follows: Let B be a set of up to 100 hosts that point to a given host S with outlink count ....

J. Dean and M. Henzinger. Finding related pages in the world wide web. In Proc. 8th WWW Conference, 1999.

No context found.

J. Dean and M.R. Henzinger. Finding related pages in the world wide web. In Proceedings of the Eighth International World Wide Web Conference WWW-1999, Toronto, May 1999.

No context found.

Jerey Dean and Monika R. Henzinger. Finding related pages in the World Wide Web. Computer Networks (Amsterdam, Netherlands: 1999.

No context found.

Dean, J., Henzinger, M.R.: Finding related pages in the World Wide Web. In: Proceeding of the 8th International World Wide Web Conference (WWW). (1999) 1467--1479

No context found.

J. Dean and M. Henzinger. Finding related pages in 330 the World Wide Web. In 8th World Wide Web Conference, Toronto, Canada, May 1999.

No context found.

J. Dean and M. R. Henzinger. Finding related pages in the World Wide Web. Computer Networks (Amsterdam, Netherlands: 1999.

No context found.

J. Dean and M. R. Henzinger. Finding related pages in the World Wide Web. Computer Networks (Amsterdam, Netherlands: 1999.

No context found.

J. Dean and M. Henzinger. Finding related pages in the world wide web. In Proc. of WWW, 1999.

No context found.

J. Dean and M. R. Henzinger, "Finding related pages in the World Wide Web," in Proc. of WWW-8, the Eighth International World Wide Web Conference, Toronto, Canada, 1999, pp. 389--401.

No context found.

J. Dean and M. R. Henzinger. Finding related pages in the World Wide Web. Computer Networks (Amsterdam, Netherlands: 1999.

No context found.

J. Dean and M. Henzinger. Finding related pages in the World Wide Web. In Proc. 8th International World Wide Web Conference, Toronto, Canada, 1999.

No context found.

J. Dean and M. Henzinger. Finding related pages in the world wide web. In Proc. of WWW, 1999.

No context found.

J. Dean and M. Henzinger. Finding related pages in 330 the World Wide Web. In 8th World Wide Web Conference, Toronto, Canada, May 1999.

No context found.

J. Dean and M. R. Henzinger. Finding related pages in the World Wide Web. Computer Networks (Amsterdam, Netherlands: 1999.

No context found.

Jeffrey Dean and Monika Rauch Henzinger. Finding related pages in the world wide web. WWW8 / Computer Networks, 31(11-16):1467--1479, 1999.

No context found.

J. Dean and M. Henzinger. Finding Related Pages in the World Wide Web. Proceedings of WWW8, 1999.

No context found.

J. Dean and M. R. Henzinger. Finding related pages in the World Wide Web. In A. Mendelzon, editor, Proceedings of the 8th International World Wide Web Conference (WWW-8), pages 389--401, Toronto, Canada, 1999.

No context found.

J. Dean and M. Henzinger. Finding related pages in the World Wide Web. In Proc. of the 8th International World Wide Web Conference, May 1999.

No context found.

J. Dean and M. Henzinger. Finding related pages in the World Wide Web. In Proc. of the 8th International World Wide Web Conference, May 1999.

No context found.

J. Dean and M. Henzinger. Finding related pages in the World Wide Web. In Proc. 8th International World Wide Web Conference, Toronto, Canada, 1999.

No context found.

J. Dean and M. R. Henzinger. Finding related pages in the World Wide Web. In Proceedings of the 8th International World Wide Web Conference, 1999.

No context found.

Dean, J., and Henzinger, M.R., "Finding related pages in the World Wide Web," Computer Networks 31(11-16), 1467-1479, 1999. Available at http://citeseer.nj.nec.com/dean99finding.html

No context found.

J. Dean and M. R. Henzinger. Finding related pages in the World Wide Web. In Proc. 8th WWW Conference, 1999.

No context found.

J. Dean and M. R. Henzinger. Finding related pages in the world wide web. WWW8 / Computer Networks, 31(11-16):1467--1479, 1999.

No context found.

J. Dean and M. R. Henzinger. Finding related pages in the world wide web. WWW8 / Computer Networks, 31(1116) :1467--1479, 1999.

No context found.

Dean, J. Henzinger, M.: Finding related pages in the World Wide Web. Proceedings of the Eighth International World Wide Web Conference (1999) 389-401

No context found.

J. Dean and M. R. Henzinger. Finding related pages in the World Wide Web. In WWW '99.

No context found.

J. Dean and M. Henzinger. Finding Related Pages in the World Wide Web. Proceedings of WWW8, 1999.

No context found.

J. Dean and M. Henzinger. Finding related pages in the World Wide Web. In Proceedings of the World Wide Web Conference (WWW'99), pages 1467--1479, 1999.

No context found.

Jeffrey Dean, and Monica R.Henzinger, Finding related pages in the world wide web. 8th international WWW conference, 1999.

First 50 documents  Next 50

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