B. Padmanabhan and A. Tuzhilin. Pattern discovery in temporal databases: a temporal logic approach. Proc. ACMKDD, 351-354, 1996.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....defined to be a collection of events that occur relatively close to each other in a given partial order. A time window is moved across the input sequence and all episodes that occur in some user specified percentage of windows are reported. The model was further generalized by Padmanabhan et al. [17] to suit temporal logic patterns. 5.1.2 Periodic Patterns Full cyclic pattern was first studied in [16] The input data to [16] is a set of transactions, each of which consists a set of items. In addition, each transaction is tagged with an execution time. The goal is to find association rules ....

B. Padmanabhan and A. Tuzhilin. Pattern discovery in temporal databases: a temporal logic approach. Proc. ACMKDD, 351-354, 1996.

....The general model is presented in Section 3. Section 4 outlines the major steps of our algorithm. Section 5 presents experimental results. The conclusion is drawn in Section 6. 2 Related Work Most previous work on mining sequence data fell into two categories: discovering sequential patterns [1, 2, 4, 7, 8, 9, 10, 14] and mining periodic patterns[3, 6, 11, 12] The primary difference between them is that the models of sequential pattern purely take into account the number of occurrences of the pattern while the frameworks for periodic patterns focus on characterizing cyclic behaviors. Due to space limitations, ....

B. Padmanabhan and A. Tuzhilin. Pattern discovery in temporal databases: a temporal logic approach. Proc. SIGKDD, 351-354, 1996.

....temporal patterns. For instance, patterns like event A occurs during the time event B happens or event A s occurrence time overlaps with that of event B and both of these events happen before event C appears cannot be expressed as simple sequential orders. On the other hand, temporal logic [7] is suggested to be used for expressing temporal patterns defined over categorical data. Temporal operators such as since, until and next are used. We may have patterns like event A always occurs until event B appears . Simple ordering of events is considered. In this paper, we consider ....

B. Padmanabhan and A. Tuzhilin. Pattern discovery in temporal databases: A temporal logic approach. In 2nd International Conf. on Knowledge Discovery and Data Mining, pages 351--354, August 1996.

.... in [6] and [63] The search for sequences of events that occur in a particular order and within a particular time interval is described in [40] and [39] A logic for expressing temporal patterns defined over categorical data as a means for discovering patterns in sequences is described 6 in [48]. Recent approaches for the discovery of patterns in sequences are described in [67] 25] and [68] The problem of mining for patterns in time series has received a considerable amount of attention recently. An approach that queries the Fourier series representation of a sequence is described ....

B. Padmanabhan and A. Tuzhilin. Pattern discovery in temporal databases: a temporal logic approach. In Proceedings of the Second International Conference on Knowledge Discovery and Data Mining (KDD'96), pages 351--354, Portland, Oregon, August 1996.

....The experimental data originates from a fault management database of a mobile communication network. The problem of discovering recurrent combinations of alarms from such databases has been considered in [24, 27, 31, 39] Closely related data mining problems have been considered, e.g. in [6, 22, 41, 46, 50, 55, 47, 59]. The dataset consists of a sequence of 46662 alarms omitted by the network elements such as base stations and transmission devices during a period of one month. The time granularity of the data is one second. The average frequency of alarms is approximately 1500 alarms day, or 1 alarm minute, but ....

B. Padmanabhan and A. Tuzhilin. Pattern discovery in temporal databases: A temporal logic approach. In Proceedings of the Second International Conference on Knowledge Discovery and Data Mining (KDD'96), pages 351--354, 1996.

....by using incremental recognition of episodes; see [38] for details, and [35] for extensions with logical variables etc. The results are good: the algorithms are efficient, and using them one can find easily comprehensible results about the combinations of event types that occur together. See also [42] for a temporal logic approach to this area. 4.3 Finding keys or functional dependencies The key finding problem is: given a relation r, find all minimal keys of r. It is a special case of the problem of finding the functional dependencies that hold in a given relation. Applications of the key ....

B. Padmanabhan and A. Tuzhilin. Pattern discovery in temporal databases: A temporal logic approach. In Proceedings of the Second International Conference on Knowledge Discovery and Data Mining (KDD'96), pages 351--354, 1996.

....of association rules [1] or general dependencies [23, 39] might find interesting applications in chess databases for discovering typical piece patterns, such as In many cases when white castles queen sides, he will sooner or later play h4. Other techniques are able to discover temporal patterns [37] or interesting deviations from the norm [22] For an excellent collection of papers on various KDD techniques consult [10] There are also chess specific KDD tasks that deserve a deeper investigation, such as the discovery of playing strategies from endgame databases, which we briefly discuss in ....

B. Padmanabhan and A. Tuzhilin. Pattern discovery in temporal databases: A temporal logic approach. In E. Simoudis, J. Han, and U.M. Fayyad, editors, Proceedings of the 2nd International Conference on Knowledge Discovery and Data Mining, pages 351--354, Portland, OR, 1996. AAAI Press.

....[Mannila and Raiha, 1994; Pfahringer and Kramer, 1995] might find interesting applications in chess databases for discovering typical piece patterns, such as In many cases when white castles queen sides, he will sooner or later play h4. Other techniques are able to discover temporal patterns [Padmanabhan and Tuzhilin, 1996] or interesting deviations from the norm [Kl osgen, 1995] For an excellent collection of papers on various KDD techniques consult [Fayyad et al. 1995] There are also chess specific KDD tasks that deserve a deeper investigation, such as the discovery of playing strategies from endgame ....

B. Padmanabhan and A. Tuzhilin. Pattern discovery in temporal databases: A temporal logic approach. In E. Simoudis, J. Han, and U.M. Fayyad, editors, Proceedings of the 2nd International Conference on Knowledge Discovery and Data Mining, pages 351--354, Portland, OR, 1996. AAAI Press.

No context found.

Padmanabhan, B. and A. Tuzhilin. 1996. Pattern Discovery in Temporal Databases: A Temporal Logic Approach. Proceedings of the Second International Conference on Knowledge Discovery in Databases (KDD-96), 351-354.

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