T. Oates, M. Schmill, and P. R. Cohen. Parallel and Distributed Search for Structure in Multivariate Time Series. In Machine Learning: ECML-97, volume 1224 of Lecture Notes in Computer Science : Lecture Notes in Artificial Intelligence, pages 191--198, New York, NY, 1997. Springer-Verlag. 9th European Conference on Machine Learning.  Home/Search   Document Details and Download   Summary   Related Articles   Check

.... search and pruning, the idea of using an optimal refinement operator has been used by a number of learning systems, e.g. in [4] Optimistic estimate pruning is closely related to general AI search procedures such as A and has already been used e.g. in the temporal pattern discovery system MSDD [10]. Minimal support pruning, of course, is a central element e.g. of any association rule algorithm. In terms of the sampling ideas used here, the idea of using Chernoff bounds to determine error limits on sampled frequencies was e.g. proposed by [1] for use with association discovery algorithms. ....

T. Oates, M. Schmill, and P. Cohen. Parallel and distributed search for structure in multivariate time series. In M. van Someren and G. Widmer, eds, Machine Learning: ECML-97, Berlin, New York, 1997. Springer Verlag.

.... 1996b) learning probabilistic planning operators (Oates Cohen 1996a) and acquiring rules for correlating and predicting asynchronous events (Oates, Jensen, Cohen 1998) Previous work explored distributing msdd s search space over multiple machines as a mechanism for reducing its running time (Oates, Schmill, Cohen 1997). In this paper, we describe three methods for reducing the size of the search space that msdd considers and empirically evaluate their utility. The remainder of this section discusses the core msdd algorithm. Section 2 describes the three search space reduction methods and section 3 summarizes ....

Oates, T.; Schmill, M. D.; and Cohen, P. R. 1997. Parallel and distributed search for structure in multivariate time series. In Proceedings of the Ninth European Conference on Machine Learning.

....problem. Examples of temporally structured data include time series of economic indicators, distributed network status reports, and continuous streams such as flight recorder data. We have developed a family of algorithms for finding structure in multivariate, discrete valued time series data (Oates Cohen 1996b; Oates, Schmill, Cohen 1996; Oates et al. 1995) In this paper, we introduce a new member of that family for handling event based data, and offer an empirical characterization of a time series based algorithm. 1 Introduction Dependency detection is an approach to finding patterns in time ....

....of temporally structured data include time series of economic indicators, distributed network status reports, and continuous streams such as flight recorder data. We have developed a family of algorithms for finding structure in multivariate, discrete valued time series data (Oates Cohen 1996b; Oates, Schmill, Cohen 1996; Oates et al. 1995) In this paper, we introduce a new member of that family for handling event based data, and offer an empirical characterization of a time series based algorithm. 1 Introduction Dependency detection is an approach to finding patterns in time series or event data based on ....

[Article contains additional citation context not shown here]

Oates, T.; Schmill, M. D.; and Cohen, P. R. 1996. Parallel and distributed search for structure in multivariate time series. Technical Report 96-23, University of Massachusetts at Amherst, Computer Science Department.

....of one or more fields unspecified, which is denoted by assigned those fields the wildcard value . Therefore, the space of all possible PIEs is given by P = Theta f i=1 (V i [ fg) Note that E ae P. 1 Medd is based on our earlier work with a similar algorithm named msdd (Oates Cohen 1996; Oates, Schmill, Cohen 1997; Oates et al. 1995) 2 We rigorously define close temporal proximity in Section 2.3. Consider an extremely simple event structure containing two fields status and element such that V status = fup, downg and V element = fhost, routerg. Then E and P are as follows: E = ae (up host) ....

Oates, T.; Schmill, M. D.; and Cohen, P. R. 1997. Parallel and distributed search for structure in multivariate time series. To appear in Proceedings of the Ninth European Conference on Machine Learning.

....fields the wildcard value . Therefore, the space of all possible PIEs is given by P = Theta f i=1 (V i [ fg) Note that E ae P. Consider a simple event structure containing two fields status and element such that 1 medd is based on our earlier work with a similar algorithm named msdd (Oates Cohen 1996; Oates, Schmill, Cohen 1996; Oates et al. 1995) V status = fup, downg and V element = fhost, routerg. Then E and P are as follows: E = up host) up router) down host) down router) P = 8 : up host) up router) up ) down host) down router) down ) host) router) ....

....value . Therefore, the space of all possible PIEs is given by P = Theta f i=1 (V i [ fg) Note that E ae P. Consider a simple event structure containing two fields status and element such that 1 medd is based on our earlier work with a similar algorithm named msdd (Oates Cohen 1996; Oates, Schmill, Cohen 1996; Oates et al. 1995) V status = fup, downg and V element = fhost, routerg. Then E and P are as follows: E = up host) up router) down host) down router) P = 8 : up host) up router) up ) down host) down router) down ) host) router) 9 = A PIE p 2 P is ....

[Article contains additional citation context not shown here]

Oates, T.; Schmill, M. D.; and Cohen, P. R. 1996. Parallel and distributed search for structure in multivariate time series. Technical Report 96-23, University of Massachusetts at Amherst, Computer Science Department. Long version of conference paper with same title.

No context found.

T. Oates, M. Schmill, and P. R. Cohen. Parallel and Distributed Search for Structure in Multivariate Time Series. In Machine Learning: ECML-97, volume 1224 of Lecture Notes in Computer Science : Lecture Notes in Artificial Intelligence, pages 191--198, New York, NY, 1997. Springer-Verlag. 9th European Conference on Machine Learning.

No context found.

T. Oates, M. Schmill, and P. Cohen. Parallel and distributed search for structure in multivariate time series. In M. van Someren and G. Widmer, eds, Machine Learning: ECML-97, Berlin, New York, 1997. Springer Verlag.

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