M. Perkowitz and O. Etzioni. Towards adaptive Web sites: conceptual framework and case study. Computer Networks (Amsterdam, Netherlands: 1999.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....approach is that learning is unsupervised, i.e. no labeled data is required. More recently, systems such as PHOAKS [50] and our own # ##[25] 24] have sought to use cooperative information retrieval techniques for personalization. End to End personalization is predicated on adaptive Web sites[39], 38] which change the information returned in response to a user request based on the user. Very primitive forms of this can be seen in sites that ask the users to provide some basic information such as address, phone number, and keywords indicating interest, and then tailor their information ....

....indicating interest, and then tailor their information content (and especially ads) based on zip code, area code, demographic pro le, etc. However, in general the appearance of a particular page, including links on it, can also be changed when Web sites are adaptive. For example, Etzioniet al... [39] de ne operations such as promotions demotions, highlighting and linking that could be done on static pages to create content tailored for a speci c user dynamically. While interesting, their formalism seems to demand a robust clustering technique in order to work successfully. Moreover, factors ....

M. Perkowitz and O. Etzioni. Towards adaptiveweb sites: Conceptual framework and case study. Arti cial Intelligence, 118:245-275, 2000.

....an input parameter, however Perkowitz developed a method to automatically determine the value of K. They ran the K means algorithm multiple times, starting with a large value and gradually decreasing it. They were able to efficiently determine a good value for K during the clustering of web pages [10]. Our algorithm only needs to cluster strongly connected words, but the K means algorithm divides whole words into K clusters without removing weak relations. COBWEB is a incremental conceptual clustering algorithm. Each cluster records the probability of each attribute and value, and the ....

....News Dude, in our approach we model a continuum of long term to short term interests. Syskill Webert [9] use a predefined profile, which significantly increases the classification accuracy on previously unseen web pages. They emphasize the importance of a user profile. Perkowitz and Etzioni [10] introduced SCML, a concept learning algorithm that only extracts some concepts in a set of data. 3. USER INTEREST HIERARCHY A user interest hierarchy (UIH) organizes a user s general to specific interests. Towards the root of a UIH, more general (longer term) interests are represented by larger ....

Perkowitz, M., and Etzioni, O. Towards adaptive Web sites: Conceptual framework and case study, Artificial Intelligence 118, 245-275, 2000.

....15] suggest analyzing user access patterns to help design better web pages, sites, and browsers. Perkowitz et al. 15] proposed the design of adaptive web sites by promoting and demoting pages, highlighting links, adding hotlinks, and clustering related pages. Two years later, Perkowitz et al. [16] presented the PageGater algorithm, which analyzes user access logs in order to identify candidate link sets to be included in index pages. Their algorithm performs the following steps: a) process the access log into visits, b) nd clusters of linked pages and create an adjacency matrix, c) nd ....

Mike Perkowitz and Oren Etzioni. Towards Adaptive Web Sites: Conceptual Framework and Case Study. In Proceedings of the Eigth International WWW Conference, Toronto, Canada, May 1999. http://www8.org/w8-papers/2bcustomizing /towards/towards.html.

....Pirolli et al. 33] propose to create aggregation of Web pages according to their importance or their content. Perkowitz et al. 30] propose the design of adaptive Web sites by promoting and demoting pages, highlighting hyperlinks, adding hyperlinks and clustering related pages. Perkowitz et al. [31] present an algorithm that analyses user access logs in order to identify candidate hyperlink clusters to be included in index pages. The performance of the algorithm is measured according to the quality of the clusters; specifically, they assess the quality of a cluster by answering the following ....

Mike Perkowitz and Oren Etzioni. Towards adaptive Web sites: Conceptual framework and case study. Artijcial Inteligence, 118(1-2):245-275, 2000.

....Pirolli et al. 4] propose to create aggregation of Web pages according to their importance or their content. Perkowitz et al. 5] propose the design of adaptive Web sites by promoting and demoting pages, highlighting hyperlinks, adding hyperlinks and clustering related pages. Perkowitz et al. [6] present an algorithm that analyzes user access logs in order to identify candidate hyperlink clusters to be included in index pages. The performance of the algorithm is measured according to the quality of the clusters; speci cally, they assess the quality of a cluster by answering the following ....

M. Perkowitz and O. Etzioni (1999), Towards adaptive Web sites: Conceptual framework and case study, Arti cial Intelligence, vol. 118, no. 1-2, pp. 245-275.

....an input parameter, however Perkowitz developed a method to automatically determine the value of K. They ran the K means algorithra multiple times, starting with a large value and gradually decreasing it. They were able to efficiently determine a good value for K during the clustering of web pages [10]. Our algorithm only needs to cluster strongly connected words, but the K means algorithm divides whole words into K clusters without removing weak relations. COBWEB is a incremental conceptual clustering algorithm. Each cluster records the probability of each attribute and value, and the ....

....News Dude, in our approach we model a continuum of long term to short term interests. Syskill Webert [9] use a predefined profile, which significantly increases the classification accuracy on previously unseen web pages. They emphasize the importance of a user profile. Perkowitz and Etzioni [10] introduced SCML, a concept leaming algorithm that only extracts some concepts in a set of data. 3. USER INTEREST HIERARCHY A user interest hierarchy (UIH) organizes a user s general to specific interests. Towards the root of a UIH, more general (longer term) interests are represented by larger ....

Perkowitz, M., and Etzioni, O. Towards adaptive Web sites: Conceptual framework and case study, Artificial Intelligence 118,245-275, 2000.

....The classification of a document to a unique cluster, using classic clustering methods is not satisfying for the web domain. In order to find a partition of all web documents we want to find groups of pages that belong together. That means that pages may be placed into more than one cluster. In [18] new clustering algorithms, called cluster mining algorithms are presented. These are specifically designed to satisfy the requirements of the web domain. A cluster mining algorithm is an unsupervised algorithm for identifying a small set of possibly overlapping clusters with limited coverage. ....

....a quadratic matrix whose cells are filled with the values of co occurrence frequencies. If two pages are already linked, their corresponding cell is set to zero. The view of the matrix as a graph aims at applying graph algorithms to the task of creating clusters of related pages. Paliouras et al. [18] presented a modification of PageGather algorithm, proposing the normalization of the weights assigned to graph edges. This additional step removes biases for the attributes that appear very often in all users. Additionally, the modified cluster mining algorithm does not restrict its search to ....

[Article contains additional citation context not shown here]

Perkowitz M, Etzioni O. "Towards adaptive Web sites: Conceptual framework and case study". Artificial Intelligence 2000; 118: 245-275

....is frequently accessed to a higher level in the page hierarchy, so that visitors are able to find that particular information quicker. Extensive research has been done in adapting websites to facilitate an individual user or a group of users, using both content based and access based adaptation [19]. Until now, the main research issue has been techniques such as information retrieval, classification, page clustering and semantics matching for contentbased adaptation; and data mining, association rule discovery, page clustering, techniques for access based adaptation [13] 18] 20] ....

.... such as information retrieval, classification, page clustering and semantics matching for contentbased adaptation; and data mining, association rule discovery, page clustering, techniques for access based adaptation [13] 18] 20] 25] For a more extensive list of references, see e.g. [19]. In the Hypermedia community, some attempts have been made to formulate adaptation models [6] 7] often integrated within the DEXTER reference model [13] None of these models targets websites specifically though, building on standard hypermedia systems functionality, or they are very ....

[Article contains additional citation context not shown here]

Perkowitz, M., and Etzioni, O. Towards Adaptive Web Sites: Conceptual Framework and Case Study, in Artificial Intelligence 118 (1-2) (2000)

....and context dependent properties of the content to be delivered. What we provide to the application developer is a general model of context and how it can be used to manipulate information, its structure and presentation. Our aim is therefore not on adaptive web sites per se as described in [PE99] but rather on context models and mechanisms that could be used to support the development of such sites. We conclude this section by commenting on the wide variety of commercial content management systems that have been developed to support both the development and operation of complex web ....

Mike Perkowitz and Oren Etzioni. Towards adaptive Web sites: conceptual framework and case study. Computer Networks (Amsterdam, Netherlands: 1999), 31(11--16):1245--1258, 1999.

....23] suggest analyzing user access patterns to help design better web pages, sites, and browsers. Perkowitz et al. 23] propose the design of adaptive web sites by promoting and demoting pages, highlighting links, adding hyperlinks, and clustering related pages. Two years later, Perkowitz et al. [24] presented an algorithm that analyses user access logs in order to identify candidate link sets to be included in index pages; their algorithm performs the following steps: 1. process the access log into visits, 2. nd clusters of linked pages and create an adjacency matrix, 3. nd maximal ....

Mike Perkowitz and Oren Etzioni. Towards Adaptive Web Sites: Conceptual Framework and Case Study. In Proceedings of the Eigth International WWW Conference, Toronto, Canada, May 1999. http://www8.org/w8-papers/2b-customizing/towards/towards.html.

....robustness, as well as incremental performance and capacity scalability with an expanding user base. Possible approaches for coping with these requirements are the so called end to end solutions, where origin servers adapt their content on the fly, taking into account consoli dated user profiles [35] or the terminal device and network connection involved in a user session [28] For an end to end approach to work properly, however, adaptation software has to be inserted at each origin server. Consequently, software updates have to be prop agated to all origin servers whenever new terminal ....

M. Perkowitz and O. Etzioni. Towards adaptive Web sites: Conceptual framework and case study. Artificial Intelligence, 118:245-275, 2000.

....work, we evaluated a variety of visitor models, including Na ve Bayes mixture models and mixtures of Markov models, concluding that a mixture of Markov models performed best for the task. In this paper, we argue that RMMs can perform substantially better for this same task. Perkowitz and Etzioni [25] also address the shortcut problem, but they use a simpler prediction method: for each pair of pages P, Q on the site, their system records how often Q is viewed by following some chain of links via P . When page P is requested after these statistics have been computed, the system adds the top m ....

....more user data in order to make predictions. Like the traditional Markov approaches, Perkowitz and Etzioni s system can t predict good shortcuts for pages which weren t visited in the training data. In addition to their work on the shortcut problem, Perkowitz and Etzioni developed IndexFinder [25], which uses page meta data to cluster web pages into conceptually similar groups, and subsequently builds coherent index or hub pages of links. The meta data is similar in spirit to the values used to instantiate RMM relations, although IndexFinder does not segregate pages into relations, or ....

M. Perkowitz and O. Etzioni. Towards adaptive web sites: Conceptual framework and case study. Artificial Intelligence Journal, 118(1--2), 2000.

....the web site to satisfy the need of users better by providing better site map, change layout of the pages and the placement of links etc. Perkowitz and Etzioni have proposed the concept of adaptive web site that dynamically optimize the structure of web pages based on the users access pattern[8]. The analysis of web log to understand the characteristics of web users has been one of the major topics in web mining. Some data mining techniques has been applied on web logs to predict future user behavior and to derive marketing intelligence[11] 6] 7] Currently many e commerce applications ....

M. Perkowitz, O. Etzioni. "Towards Adaptive Web Sites: Conceptual Framework and Case Study", In Proc. of WWW8, May 1999.

.... stations would have the computing and storage resources to host such proxies [32] End to end solutions represent a possible alternative, with origin servers adapting con tent on the fly according to the terminal device and network connection involved in a user session [32] or user profiling [39]. Under the end to end approach, however, adaptation software has to be inserted at each origin server. Consequently, software updates have to be propagated to all origin servers whenever new terminal devices and content encoding pro tocols emerge. Furthermore, on the fly adaptation of content ....

M. Perkowitz and O. Etzioni. Towards adaptive Web sites: Conceptual framework and case study. Artificial Intelligence, 118:245-275, 2000.

....domains of interest. Web Watcher [1] tries to predict what links will be followed from the current page, while Avanti [2] dynamically customizes the content of the Web page based on the user s needs and tastes, and predicts the user s next step and the eventual goal. Perkowitz and Etzioni s [3] adaptive Web server uses feedback from all the users to modify the site s content. The second direction seeks to improve the client satisfaction with the Web server by increasing its performance. A multicast solution [4] for distributing commonly requested pages, which also accommodates service ....

M. Perkowitz, O. Etzioni. Towards Adaptive Web Sites: Conceptual Framework and Case Study. Artificial Intelligence 118 (2000), pp. 245-275.

....( 5] There are known two ways of addressing this problem. One is the customization of the Web site (we will not refer to this) The other, more recent, approach is the optimization of the site s structure to make it easier to use for all visitors. This is the trend followed by the authors in [5, 6, 7]. More precisely, they investigate the data accumulated in Web server access logs and identify a number of cohesive, possibly overlapping clusters of pages that they conclude, based on users access patterns detected in logs data, that are related to a particular topic. For some of these clusters ....

....a particular topic. For some of these clusters are synthesized index pages which contain a link for every page in that cluster. The methods used in the above mentioned papers are AI traditional clustering techniques adapted to the speci c of the problem. In this paper, we proceed similarly as in [5, 6, 7]. Namely, we want to create orientation pages with links to the related by content pages of the site. But we propose a di erent manner to partition the site: using Web mining methods instead of AI clustering techniques. Also, this partitioning will be made with more accuracy, as we shall see. The ....

Perkowitz M., Etzioni O., Towards adaptive Web sites: Conceptual framework and case study, Arti cial Intelligence 118 (

.... attempt to enhance the experience and reduce the latency of the average user, a number of authors have suggested ways of improving the design of websites, such as promoting and demoting pages, highlighting links, and clustering related pages in an adaptive fashion depending on user access patterns [4, 7]. In this paper we consider the strategy of adding hotlinks , i.e. shortcuts from web pages at or near the home page of a site to popular pages a number of levels down in the (generally directed) network of pages. The idea of hotlinks was suggested by Perkowitz and Etzioni [7] and studied ....

....patterns [4, 7] In this paper we consider the strategy of adding hotlinks , i.e. shortcuts from web pages at or near the home page of a site to popular pages a number of levels down in the (generally directed) network of pages. The idea of hotlinks was suggested by Perkowitz and Etzioni [7] and studied earlier by Bose et al. 2] for the special case of a website represented by a complete binary tree. Experimental results showing the validity of this approach are given in [3] We model a website as a rooted directed tree T = V; E) where V is a collection of webpages connected by a ....

M. Perkowitz and O. Etzioni, \Towards Adaptive Web Sites: Conceptual Framework and Case Study", Proceedings of Eight WWW Conference, 1999.

....databases, a few possible applications are outlined. The activity of analyzing click stream data obtained from web servers is rapidly becoming an essential activity for e businesses [7] The purpose may be to do business analysis, user behavior analysis [3] or to produce adaptive web sites [9, 10]. Because bulks of click stream data are continuously being accumulated for such analyses, it is becoming increasingly important for IT departments to control the growth in data volumes. Perhaps, the IT department wants to control the growth in data volumes by removing data more than 1 years old. ....

M. Perkowitz and O. Etzioni. Towards Adaptive Web Sites: Conceptual Framework and Case Study. In Proceedings of the 8th International World Wide Web Conference/Computer Networks, 31(1116) :1245--1258, 1999.

....contains in itself some useful information. If, for example, some links are frequently activated in the same order on a given server, we can deduce that the links are indeed next links, like the ones you sometimes find at the bottom of web pages. Adaptive sites use this type of information. [2,3] (c) They act as implicit sources of information when the author didn t actually specify it. This is the case in the two previous examples. d) They act as an explicit source of information, declared by the author in the following example. Our syndicate C.F.D.T. is opposed to the government ....

Perkowitz M. and Etzioni O. "Towards Adaptive Web Sites: Conceptual Framework and Case Study." Proceedings of WWW8. 1999.

....Perkowitz and Etzioni [1997] challenged the AI community to build adaptive web sites: sites that automatically improve their organization and presentation by learning from visitor access patterns. In the spirit of this challenge, many research projects have been proposed and implemented [Perkowitz and Etzioni, 2000; Fink et al. 1996; Yan et al. 1996; J uhne et al. 1998; Joachims et al. 1997; Pazzani and Billsus, 1999; Sarukkai, 2000] Many of these projects, like much of the Web today, assume the visitor is browsing with a large color display and fast network connection. In addition, these works ....

....viewed by others and avoid becoming lost. However, tour guides assume that every page along the trail is important, and typically are limited to only suggesting which link on a page to follow next (as opposed to creating shortcuts between pages) SurfLen [Fu et al. 2000] and PageGather [Perkowitz and Etzioni, 2000] suggest pages to visit based on page requests co occurrent in past sessions 2 . These algorithms suggest the top m pages that are most likely to co occur with the visitor s current session, either by presenting a list of links (SurfLen) or by constructing a new index page containing the links ....

M. Perkowitz and O. Etzioni. Towards adaptive web sites: Conceptual framework and case study. Art. Int. J., 118(1--2), 2000.

.... full economic potential of the Web [13, 15] Knowledge of user access patterns is useful in numerous applications: supporting Web site design decisions such as content and structure justi cations [6, 23] optimizing systems by enhancing caching schemes and load balancing, making Websites adaptive [16], supporting business intelligence and marketing decisions [3] testing user interfaces, monitoring for security purposes, and more importantly, in Web Personalization applications such as recommendation systems [18] and target advertising. Commercial products such as Personify TM [25] ....

Perkowitz M., and O. Etzioni. 2000. Toward adaptive Web sites: Conceptual framework and case study. Articial Intelligence 118, p.p 245275, 2000.

....Mixture model columns are annotated with the number of clusters. Errorbars denote 95 confidence intervals. expected savings computation is substantially faster than and positively complements PROTEUS s more general approach. 5 Related work Two closely related lines of research are IndexFinder [15] and Digestor [3] IndexFinder creates singular transformations that appeal to all visitors at the site, in particular, generating new index pages hubs of links to other pages on the site. These pages are evaluated based strictly on the navigational usage patterns of past visitors the ....

M. Perkowitz and O. Etzioni. Towards adaptive web sites: Conceptual framework and case study. Art. Int. J., 118(1--2), 2000.

....getting what they want. The most advanced personalization sites do not require the user to perform the personalization manually by selecting desired categories of information, instead it performs the personalization based on the users actions. These sites are often referred to as adaptive sites [11, 12]. However, to be able to perform the personalization automatically, the web system has to analyze the user s previous behavior carefully to determine their habits and preferences. Thus, it is very important to find powerful and effective ways to analyze large amounts of clickstream data, ....

M. Perkowitz and O. Etzioni. Towards Adaptive Web Sites: Conceptual Framework and Case Study. In Proceedings of the Eighth International World Wide Web Conference, 1999.

....Avanti [2] project customizes dynamically the content of the Web page based on user s needs and tastes and predicts both the user s next step and their eventual goal. The previous projects adapt the site s presentation for each individual user according to their preferences. Perkowitz and Etzioni [3] proposed another solution for adaptive Web servers. The system allows the server to learn what all the users that access the site want and to automatically adapt the site s content. Thus, it makes the Web site easier to use for everyone, including the clients who are accessing the site for the ....

M. Perkowitz, O. Etzioni, "Towards Adaptive Web Sites: Conceptual Framework and Case Study", Artificial Intelligence 118 (2000), pg. 245-275.

....In the following we give a survey on 15 existing approaches to customization. Considering them with respect to phases, most of these approaches consider customization from an implementation point of view. There are a lot of research prototypes [13] 14] 17] 19] 27] 37] 38] 42] [44] and also some commercial systems such as [40] 43] In contrast to that, few approaches for customization focus on analysis and design such as [9] 20] 49] We encountered only one modeling approach, supporting customization throughout all phases, namely [45] The survey presented is ....

....components (M) an off line component for the usage mining task in terms of identifying user activity patterns and an on line component performing dynamic adaption of hypertext and presentation. They focus mostly on the issues of clustering and recommendation algorithms (M) Perkowitz and Etzioni [44] introduce the so called IndexFinger page synthesis program to increase the quality of the web site for users. The user context (C.1) in terms of the interaction history (C.3) is analyzed by usage mining techniques to determine access patterns. These access patterns are identified by the ....

M. Perkowitz, and O. Etzioni: "Towards Adaptive Web Sites: Conceptual Framework and Case Study", Proc. of the The Eighth International World Wide Web Conference, (WWW8), Toronto, Canada, May 1999.

....the patterns with which users browse the site. Furthermore, it is rare that one site structure is appropriate for all classes of users. The work by Perkowitz and Etzioni [102, 103, 104] pioneered the eld of adaptive web sites; an extended description of their work is presented in this issue [105]. Such sites restructure themselves depending on usage patterns. The site can be adapted for classes of users or individual users. The key challenges involved are to infer from the browsing patterns the interesting structures of the site that may be useful for a class of users. Enforcing ....

M. Perkowitz and O. Etzioni. Towards adaptive web sites: Conceptual framework and case study. Articial Intelligence, this issue, 2000.

....[65, 66] consider the costs and benefits of alerting users in different ways versus deferring messages to later, by considering a user s context and the time dependent utility of different messaging actions. 3. 3 User Modeling with Relational Markov Models As part of our work on Adaptive Websites [119, 120, 5] we have developed several learning algorithms for acquiring predictive models of user behavior from observations. Our initial work [4] evaluated the suitability of previous techniques, including Naive Bayes mixture models, first order, second order, and propositional Markov models. While a ....

M. Perkowitz and O. Etzioni. Towards adaptive web sites: Conceptual framework and case study. Artificial Intelligence, To appear, 2000.

....and the use of timeouts to minimize cost of prediction errors. Our work on adaptive interfaces focussed initially on website design [Perkowitz and Etzioni, 1997] and led to algorithms for datamining web logs to discover aggregate patterns, which powered the automatic creation of index pages [Perkowitz and Etzioni, 2000] . Later we concentrated on mining individual behavior patterns, generating personalized sites for display on small, wireless devices [Anderson et al. 2001b] Our emphasis was on information goal seeking behavior, common to wireless internet use, rather than general browsing or surfing. The ....

M. Perkowitz and O. Etzioni. Towards adaptive web sites: Conceptual framework and case study. Artificial Intelligence, 118:245--276, 2000.

....Next, to demonstrate the potential power of adaptive web sites, we consider the index page synthesis case study in more depth. We introduce the IndexFinder page synthesis system and describe its application to a live Web site. More detail about our algorithms and experiments can be found in [6]. 1.1 VARIETIES OF ADAPTIVE WEB SITES Work on adaptive web sites has taken a variety of approaches, from providing automated tour guides, to making personalized recommendations, to charting the footprints left by visitors to a site. All approaches, however, must address the following ....

....where it is manufactured, and so on. Descriptions are represented as attribute value pairs (e.g. filetype=image, price=cheap) Data about how often pages occur together in user visits is extracted from the web server logs; IndexFinder then applies a novel statistical cluster mining technique [6] to the data and produces candidate clusters as output. In statistical cluster mining, objects (e.g. web pages) are grouped together based solely on a similarity measure (e.g. how often they co occur in user visits) the algorithm makes no guarantee that the clusters are pure or complete with ....

[Article contains additional citation context not shown here]

M. Perkowitz and O. Etzioni. Towards adaptive web sites: Conceptual framework and case study. Artificial Intelligence, To appear, 2000. 9

No context found.

M. Perkowitz and O. Etzioni. Towards adaptive Web sites: conceptual framework and case study. Computer Networks (Amsterdam, Netherlands: 1999.

No context found.

M. Perkowitz and O. Etzioni. Towards adaptive Web sites: conceptual framework and case study. Computer Networks (Amsterdam, Netherlands: 1999), 31(11--16):1245--1258, 1999

No context found.

M. Perkowitz and O. Etzioni. Towards adaptive Web sites: conceptual framework and case study. Computer Networks (Amsterdam, Netherlands: 1999.

No context found.

Mike Perkowitz, Oren Etzioni, "Towards Adaptive Web Sites: Conceptual Framework and Case Study", Artificial Intelligence, Vol. 118, No. 1--2, pp. 245--275, 2000. 39

No context found.

Mike Perkowitz, Oren Etzioni, "Towards Adaptive Web Sites: Conceptual Framework and Case Study", Computer Networks, Vol. 31, No. 11--16 (Proc. of WWW8), pp. 1245--1258, May 1999.

No context found.

M. Perkowitz and O. Etzioni. Towards adaptive web sites: Conceptual framework and case study. Artificial Intelligence, 118:245--275, 2000.

No context found.

Perkowitz, M. and Etzioni, O. (1999) Towards adaptive web sites: conceptual framework and case study. Computer Networks (and ISDN Systems), 31. Proc. 8th Int. World Wide Web Conf., Toronto, Canada, 11--14 May, pp. 1245--1258. Elsevier Science, Netherlands.

No context found.

M. Perkowitz and O. Etzioni. Towards Adaptive Web Sites: Conceptual Framework and Case Study. In Proceedings of the 8th International World Wide Web Conference/Computer Networks, 31(1116) :1245--1258, 1999.

No context found.

Perkowitz, M., Etzioni, O.: Towards adaptive Web sites: conceptual framework and case study. WWW8. (1999)

No context found.

Mike Perkowitz and Oren Etzioni. Towards adaptive web sites: Conceptual framework and case study. Artificial Intelligence, 118(1-2), 2000.

No context found.

M. Perkowitz, O. Etzioni, Towards Adaptive Web Sites: Conceptual Framework and Case Study, in Proc. of WWW8, 1999

No context found.

Mike Perkowitz, Oren Etzioni, "Towards Adaptive Web Sites: Conceptual Framework and Case Study", Artificial Intelligence, Vol. 118, No. 1--2, pp. 245--275, 2000. 39

No context found.

Mike Perkowitz, Oren Etzioni, "Towards Adaptive Web Sites: Conceptual Framework and Case Study", Computer Networks, Vol. 31, No. 11--16 (Proc. of WWW8), pp. 1245--1258, May 1999.

No context found.

Perkowitz M., Etzioni O.: Towards Adaptive Web Sites: Conceptual Framework and Case Study. Computer Networks 31, Proceedings of the 8th International WWW Conference (1999)

No context found.

Perkowitz, M., and Etzioni, O., Towards adaptive web sites: conceptual framework and case study, Proc. 8th WWW Conf., 1999.

No context found.

M. Perkovitz, O. Etzioni, Towards adaptive web sites: Conceptual framework and case study, Artificial Intelligence 118, (2000) 245275.

No context found.

Mike Perkowitz and Oren Etzioni. Towards adaptive Web sites: conceptual framework and case study. Computer Networks (Amsterdam, Netherlands: 1999), 31(11--16):1245--1258, 1999.

No context found.

Mike Perkowitz and Oren Etzioni. Towards adaptive web sites: Conceptual framework and case study. Artificial Intelligence, 118:245--275, 2000.

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