R. Fujino, H. Arimura, and S. Arikawa. Discovering unordered and ordered phrase association patterns for text mining. In Proc. of the 4th Pacific-Asia Conference on Knowledge Discovery and Data Mining, volume 1805 of Lecture Notes in Artificial Intelligence. Springer-Verlag, Apr. 2000.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....each phrase and only phrases having the value above a certain threshold are retained. Several methods take mixed approach, for example [Lin98] uses linguistic techniques to generate CHAPTER 4. PHRASE CONSTRUCTION 39 phrases and a statistical mutual information measure to weed out coincidences. FAA00] applies Apriori like frequent pattern mining algorithm to find both ordered and unordered phrases. Even though statistical methods do not take into account syntactic and or semantic features, they are much cheaper and produce phrases of the quality comparable to those discovered using ....

R. Fujino, H. Arimura, and S. Arikawa. Discovering unordered and ordered phrase association patterns for text mining. In Proc. 4th Pacific-Asia Conf. Knowledge Discovery and Data Mining (PAKDD'00), pages 281--293, Kyoto, Japan, 2000.

No context found.

R. Fujino, H. Arimura, and S. Arikawa. Discovering unordered and ordered phrase association patterns for text mining. In Proc. of the 4th Pacific-Asia Conference on Knowledge Discovery and Data Mining, volume 1805 of Lecture Notes in Artificial Intelligence. Springer-Verlag, Apr. 2000.

No context found.

R. Fujino, H. Arimura, S. Arikawa, Discovering unordered and ordered phrase association patterns for text mining. Proc. PAKDD2000, LNAI 1805, 281-293, 2000.

....the proposed algorithms. Optimized pattern mining is to nd those patterns that optimize a given statistical measure and attracting much attention in both data mining and machine learning communities [13, 18] We have devised fast and robust mining algorithm for nding simple text patterns [8, 12] and it was shown that the optimized pattern discovery is e ective in text mining in ill de ned environment. Thus, it is our future problem to develop optimized pattern discovery algorithm for tree structured data by extending our framework. In this paper, we consider the mining problem from ....

R. Fujino, H. Arimura, S. Arikawa, Discovering unordered and ordered phrase association patterns for text mining. In Proc. PAKDD

....subsequence patterns. It is challenging to apply our approach to find the best pattern in the sense of pattern languages introduced by Angluin [1] where the related consistency problems are shown to be very hard [6] Fujino et al. showed an another approach to find the best proximity pattern [3]. It may be interesting to combine these 5 approaches into one. We are now in the process of installing our algorithm into the core of the decision tree generator in the BONSAI system [7] ....

R. Fujino, H. Arimura, and S. Arikawa. Discovering unordered and ordered phrase association patterns for text mining. In Proc. of the 4th Pacific-Asia Conference on Knowledge Discovery and Data Mining, volume 1805 of Lecture Notes in Artificial Intelligence. Springer-Verlag, Apr. 2000.

.... Moreover, it is also challenging to apply our approach to find the best pattern in the sense of pattern languages introduced by Angulin [2] where the related consistency problems are shown to be very hard [13, 14, 17] Arimura et al. showed an another approach to find the best proximity pattern [3, 4, 10]. It may be interesting to combine these approaches into one. We plan to install our algorithm into the core of the decision tree generator in the BONSAI system [20] Acknowledgements The authors would like to thank Prof. Albert Apostolico and Prof. Hiroki Arimura for fruitful discussion. ....

R. Fujino, H. Arimura, and S. Arikawa. Discovering unordered and ordered phrase association patterns for text mining. In Proc. of the 4th Pacific-Asia Conference on Knowledge Discovery and Data Mining, volume 1805 of Lecture Notes in Artificial Intelligence. Springer-Verlag, Apr. 2000.

....more suitable than the ordered version. Besides this, we also have to deal with huge text databases that cannot fit into main memory. For the purpose, we developed another pattern discovery algorithm, called Levelwise Scan, for mining unordered phrase patterns from large disk resident text data [7]. Based on the design principle of the Apriori algorithm of Agrawal [1] the Levelwise Scan algorithm quickly discovers most frequent unordered patterns with d phrases and proximity k in time O(n 2 N(log n) d ) and space O(n log n R) on nearly random texts using a random sample of size n, ....

.... [1] the Levelwise Scan algorithm quickly discovers most frequent unordered patterns with d phrases and proximity k in time O(n 2 N(log n) d ) and space O(n log n R) on nearly random texts using a random sample of size n, where N is the total size of input text and R is the output size [7]. To cope with the problem of the huge feature space of phrase patterns, the algorithm combines the techniques of random sampling, the generalized suffix tree, and the pattern matching automaton. By computer experiments on large text data, the Levelwise Scan algorithm quickly finds patterns for ....

[Article contains additional citation context not shown here]

R. Fujino, H. Arimura, S. Arikawa, Discovering unordered and ordered phrase association patterns for text mining. Proc. PAKDD2000, LNAI 1805, 281--293, 2000.

.... Moreover, it is also challenging to apply our approach to find the best pattern in the sense of pattern languages introduced by Angulin [2] where the related consistency problems are shown to be very hard [13, 14, 17] Arimura et al. showed an another approach to find best proximity pattern [3, 4, 10]. It may be interesting to combine these approaches into one. In future work, we are plan to install our algorithm into the core of the decision tree generator in the BONSAI system [20] Acknowledgements The authors would like to thank Prof. Albert Apostolico and Prof. Hiroki Arimura for ....

R. Fujino, H. Arimura, and S. Arikawa. Discovering unordered and ordered phrase association patterns for text mining. In Proc. of the 4th Pacific-Asia Conference on Knowledge Discovery and Data Mining, Lecture Notes in Artificial Intelligence. Springer-Verlag, Apr. 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