N. Fuhr, S. Hartmanna, G. Lustig, M. Schwantner, and K. Tzeras. Air/x - a rule-based multistage Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Learning approaches for Detecting and Tracking News.. - Yang, Carbonell, Brown, .. (1999) (25 citations) (Correct)
....for event representation and a k means clustering method for document classification. Indirectly related work includes document clustering methods applied to retrieval and corpus navigation problems[25, 26, 12, 29, 24, 8, 34] and supervised learning algorithms applied to text categorization[13, 7, 31, 27]. Those results provide a rich background to our research, but do not directly address the problems of event detection and event tracking in temporal text and audio streams. 2 Event Analysis Before exploring the solution space, let us observe the properties of events in news stories, which may ....
N. Fuhr, S. Hartmanna, G. Lustig, M. Schwantner, and K. Tzeras. Air/x - a rule-based multistage
A Comparative Study on Feature Selection in Text Categorization - Yang, Pedersen (1997) (191 citations) (Correct)
....good indexing and summarization of document content. Documents categorization is one solution to this problem. A growing number of statistical classification methods and machine learning techniques have been applied to text categorization in recent years, including multivariate regression models[8, 27], nearest neighbor classification[4, 23] Bayes probabilistic approaches[20, 13] decision trees[13] neural networks[21] symbolic rule learning[1, 16, 3] and inductive learning algorithms[3, 12] A major characteristic, or difficulty, of text categorization problems is the high dimensionality of ....
....and hence does not treat words separately. Similarly, kNN treats a document as an single point in a vector space. The context sensitivity is in distinction to context free methods based on explicit independence assumptions such as naive Bayes classifiers[13] and some other regression methods[8]) A context sensitive classifier makes better use of the information provided by features than a context free classifier do, thus enabling a better observation on feature selection. 5) The two classifiers differ statistically. LLSF is based on a linear parametric model; kNN is a non parametric ....
N. Fuhr, S. Hartmanna, G. Lustig, M. Schwantner, and K. Tzeras. Air/x - a rule-based multistage
A Comparison of Two Learning Algorithms for Text Categorization - Lewis, Ringuette (1994) (94 citations) (Correct)
No context found.
Fuhr, Norbert, Hartmann, Stephan, Lustig, Gerhard, Schwantner, Michael, Tzeras, Konstadinos, and Knorz, Gerhard. AIR/X---a rule-based multistage
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