D. Mladeni#. Personal WebWatcher: Implementation and design. Technical Report IJS-DP-7472, Department of Intelligent Systems, Joz, ef Stefan Institute, 1996.  Home/Search   Document Not in Database   Summary   Related Articles   Check

.... how is it used to build the pro le Is the system adaptive in that the pro le changes over time, hopefully adjusting to a user s actual interests Examples for learning algorithms are probabilistic algorithms, genetic algorithms [29] and algorithms working in the vector space model [50] [31] contains a detailed bibliography for all of these approaches in the context of text learning. 2 representation of user pro les: How are the interests of a user stored Common representations include Boolean or weighted keyword vectors, semantic nets, ngrams, and keyword vectors for a small ....

....Whenever possible, references to comparisons of products in one of the categories are given. 39] is a compilation of freely available information ltering systems (some of which will not be discussed here) and [20] is an early approach to categorizing Usenet news ltering systems. 30] and [31] contain a more recent discussion on some of the systems, and [9] gives an overview on other intelligent information brokering systems. Finally, 40] 3 contains a thorough discussion of trends in text ltering, in particular with respect to personalized approaches. The word agent is rarely ....

[Article contains additional citation context not shown here]

D. Mladenic. Text-learning and intelligent agents. Technical Report IJS-DP-7948, J. Stefan Institute, Department for Intelligent Systems, Ljubljana, 1998. REFERENCES 28

....tabular form. Whenever possible, references to comparisons of products in one of the categories are given. 39] is a compilation of freely available information ltering systems (some of which will not be discussed here) and [20] is an early approach to categorizing Usenet news ltering systems. [30] and [31] contain a more recent discussion on some of the systems, and [9] gives an overview on other intelligent information brokering systems. Finally, 40] 3 contains a thorough discussion of trends in text ltering, in particular with respect to personalized approaches. The word agent is ....

....is an adaptive version of WebWatcher. Architecture: User pro les are rather a goal for one browsing session, and 2.2 Filtering and Rating 15 this goal is stored at the server s side. The same is true for the collaborative implicit feedback (which links were chosen) Personal WebWatcher [30] augments WebWatcher with adaptive behavior towards one user. It is thus a recommendation service, too. The suggestions are restricted to links that already exist on a page, and if the system considers them interesting, these links are highlighted. Information source: To build and update pro les, ....

D. Mladenic. Personal WebWatcher: design and implementation. Technical Report IJS-DP-7472, J. Stefan Institute, Department for Intelligent Systems, Ljubljana, 1998.

.... how is it used to build the profile Is the system adaptive in that the profile changes over time, hopefully adjusting to a user s actual interests Examples for learning algorithms are probabilistic algorithms, genetic algorithms [33] and algorithms working in the vector space model [48] [35] contains a detailed bibliography for all of these approaches in the context of text learning. representation of user profiles: How are the interests of a user stored Common representations include Boolean or weighted keyword vectors, semantic nets, n grams, and keyword vectors for a small ....

....Whenever possible, references to comparisons of products in one of the categories are given. 40] is a compilation of freely available information filtering systems (some of which will not be discussed here) and [23] is an early approach to categorizing Usenet news filtering systems. 34] and [35] contain a more recent discussion on some of the systems, and [9] gives an overview on other intelligent information brokering systems. Finally, 41] contains a thorough discussion of trends in text filtering, in particular with respect to personalized approaches. 2.1. APPLICATIONS OF ....

[Article contains additional citation context not shown here]

D. Mladeni'c. Text-learning and intelligent agents. Technical Report IJSDP -7948, J. Stefan Institute, Department for Intelligent Systems, Ljubljana, 1998.

....form. Whenever possible, references to comparisons of products in one of the categories are given. 40] is a compilation of freely available information filtering systems (some of which will not be discussed here) and [23] is an early approach to categorizing Usenet news filtering systems. [34] and [35] contain a more recent discussion on some of the systems, and [9] gives an overview on other intelligent information brokering systems. Finally, 41] contains a thorough discussion of trends in text filtering, in particular with respect to personalized approaches. 2.1. APPLICATIONS OF ....

....adaptive version of WebWatcher. Architecture: User profiles are rather a goal for one browsing session, and this goal is stored at the server s side. The same is true for the collaborative implicit feedback (which links were chosen) 2.1. APPLICATIONS OF PERSONALIZATION 36 Personal WebWatcher [34] augments WebWatcher with adaptive behavior towards one user. It is thus a recommendation service, too. The suggestions are restricted to links that already exist on a page, and if the system considers them interesting, these links are highlighted. Information source: To build and update ....

[Article contains additional citation context not shown here]

D. Mladeni'c. Personal WebWatcher: design and implementation. Technical Report IJS-DP-7472, J. Stefan Institute, Department for Intelligent Systems, Ljubljana, 1998.

....Web documents (not Yahoo Web documents that are used in learning) accessible from the hierarchy domain. The final result of learning is a set of specialized classifiers. 4 Experimental results Our experiments are performed using our recently developed machine learning system Learning Machine [4] that supports usage of different machine learning techniques on large data sets with especially designed modules for learning on text and collecting data from the Web. In our experiments we observe the influence of different number of features (vector size in feature vector document ....

Grobelnik, M., Mladeni'c, D., Learning Machine: design and implementation, Technical Report IJS-DP7824, Department for Intelligent Systems, J.Stefan Institute, 1998.

No context found.

D. Mladeni#. Personal WebWatcher: Implementation and design. Technical Report IJS-DP-7472, Department of Intelligent Systems, Joz, ef Stefan Institute, 1996.

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