Results 1  10
of
20
Time series knowledge mining
, 2006
"... An important goal of knowledge discovery is the search for patterns in data that can help explain the underlying process that generated the data. The patterns are required to be new, useful, and understandable to humans. In this work we present a new method for the understandable description of loca ..."
Abstract

Cited by 20 (2 self)
 Add to MetaCart
An important goal of knowledge discovery is the search for patterns in data that can help explain the underlying process that generated the data. The patterns are required to be new, useful, and understandable to humans. In this work we present a new method for the understandable description of local temporal relationships in multivariate data, called Time Series Knowledge Mining (TSKM). We define the Time Series Knowledge Representation (TSKR) as a new language for expressing temporal knowledge. The patterns have a hierarchical structure, each level corresponds to a single temporal concept. On the lowest level, intervals are used to represent duration. Overlapping parts of intervals represent coincidence on the next level. Several such blocks of intervals are connected with a partial order relation on the highest level. Each pattern element consists of a semiotic triple to connect syntactic and semantic information with pragmatics. The patterns are very compact, but offer details for each element on demand. In comparison with related approaches, the TSKR is shown to have advantages in robustness, expressivity, and comprehensibility. Efficient algorithms for the discovery of the patterns are proposed. The search for coincidence as well as partial order can be formulated as variants of the well known frequent itemset problem. One of the best known algorithms for this problem is therefore adapted for our purposes. Human interaction is used during the mining to analyze and validate partial results as early as possible and guide further processing steps. The efficacy of the methods is demonstrated using several data sets. In an application to sports medicine the results were recognized as valid and useful by an expert of the field.
Optimal multiscale patterns in time series streams
 In SIGMOD
, 2006
"... We introduce a method to discover optimal local patterns, which concisely describe the main trends in a time series. Our approach examines the time series at multiple time scales (i.e., window sizes) and efficiently discovers the key patterns in each. We also introduce a criterion to select the best ..."
Abstract

Cited by 19 (2 self)
 Add to MetaCart
(Show Context)
We introduce a method to discover optimal local patterns, which concisely describe the main trends in a time series. Our approach examines the time series at multiple time scales (i.e., window sizes) and efficiently discovers the key patterns in each. We also introduce a criterion to select the best window sizes, which most concisely capture the key oscillatory as well as aperiodic trends. Our key insight lies in learning an optimal orthonormal transform from the data itself, as opposed to using a predetermined basis or approximating function (such as piecewise constant, shortwindow Fourier or wavelets), which essentially restricts us to a particular family of trends. Our method lifts that limitation, while lending itself to fast, incremental estimation in a streaming setting. Experimental evaluation shows that our method can capture meaningful patterns in a variety of settings. Our streaming approach requires order of magnitude less time and space, while still producing concise and informative patterns. 1.
Local correlation tracking in time series
 In ICDM
, 2006
"... We address the problem of capturing and tracking local correlations among time evolving time series. Our approach is based on comparing the local autocovariance matrices (via their spectral decompositions) of each series and generalizes the notion of linear crosscorrelation. In this way, it is pos ..."
Abstract

Cited by 18 (1 self)
 Add to MetaCart
(Show Context)
We address the problem of capturing and tracking local correlations among time evolving time series. Our approach is based on comparing the local autocovariance matrices (via their spectral decompositions) of each series and generalizes the notion of linear crosscorrelation. In this way, it is possible to concisely capture a wide variety of local patterns or trends. Our method produces a general similarity score, which evolves over time, and accurately reflects the changing relationships. Finally, it can also be estimated incrementally, in a streaming setting. We demonstrate its usefulness, robustness and efficiency on a wide range of real datasets. 1
Unsupervised simultaneous learning of gestures, actions and their associations for humanrobot interaction
 In IEEE IROS
, 2009
"... Abstract — HumanRobot Interaction using free hand gestures is gaining more importance as more untrained humans are operating robots in home and office environments. The robot needs to solve three problems to be operated by free hand gestures: gesture (command) detection, action generation (related ..."
Abstract

Cited by 18 (12 self)
 Add to MetaCart
(Show Context)
Abstract — HumanRobot Interaction using free hand gestures is gaining more importance as more untrained humans are operating robots in home and office environments. The robot needs to solve three problems to be operated by free hand gestures: gesture (command) detection, action generation (related to the domain of the task) and association between gestures and actions. In this paper we propose a novel technique that allows the robot to solve these three problems together learning the action space, the command space, and their relations by just watching another robot operated by a human operator. The main technical contribution of this paper is the introduction of a novel algorithm that allows the robot to segment and discover patterns in its perceived signals without any prior knowledge of the number of different patterns, their occurrences or lengths. The second contribution is using a GangerCausality based test to limit the search space for the delay between actions and commands utilizing their relations and taking into account the autonomy level of the robot. The paper also presents a feasibility study in which the learning robot was able to predict actor’s behavior with 95.2% accuracy after monitoring a single interaction between a novice operator and a WOZ operated robot representing the actor. I.
K.: Changepoint detection using krylov subspace learning
 In: Proceedings of the SIAM Internations Conference on Data Mining. (2007
"... We propose an efficient algorithm for principal component analysis (PCA) that is applicable when only the inner product with a given vector is needed. We show that Krylov subspace learning works well both in matrix compression and implicit calculation of the inner product by taking full advantage of ..."
Abstract

Cited by 14 (1 self)
 Add to MetaCart
We propose an efficient algorithm for principal component analysis (PCA) that is applicable when only the inner product with a given vector is needed. We show that Krylov subspace learning works well both in matrix compression and implicit calculation of the inner product by taking full advantage of the arbitrariness of the seed vector. We apply our algorithm to a PCAbased changepoint detection algorithm, and show that it results in about 50 times improvement in computational time. t
T.: Robust singular spectrum transform
"... Abstract. Change Point Discovery is a basic algorithm needed in many time series mining applications including rule discovery, motif discovery, casual analysis, etc. Several techniques for change point discovery have been suggested including wavelet analysis, cosine transforms, CUMSUM, and Singular ..."
Abstract

Cited by 13 (11 self)
 Add to MetaCart
(Show Context)
Abstract. Change Point Discovery is a basic algorithm needed in many time series mining applications including rule discovery, motif discovery, casual analysis, etc. Several techniques for change point discovery have been suggested including wavelet analysis, cosine transforms, CUMSUM, and Singular Spectrum Transform. Of these methods Singular Spectrum Transform (SST) have received much attention because of its generality and because it does not require adhoc adjustment for every time series. In this paper we show that traditional SST suffers from two major problems: the need to specify five parameters and the rapid reduction in the specificity with increased noise levels. In this paper we define the Robust Singular Spectrum Transform (RSST) that alleviates both of these problems and compare it to RSST using different synthetic and realworld data series. 1
On Comparing SSAbased Change Point Discovery Algorithms
"... Abstract — Change point discovery is an important problem in data mining and industrial systems. Different approaches have been proposed and some of the most promising approaches are based on singular spectrum analysis (SSA). These algorithms have the advantages of requiring no adhoc tuning for dif ..."
Abstract

Cited by 6 (6 self)
 Add to MetaCart
(Show Context)
Abstract — Change point discovery is an important problem in data mining and industrial systems. Different approaches have been proposed and some of the most promising approaches are based on singular spectrum analysis (SSA). These algorithms have the advantages of requiring no adhoc tuning for different types of signals and having a builtin noise attenuation mechanism. In this paper we try to unify these approaches and present a novel method for comparing change point discovery algorithms. We then use the proposed method to compare different SSA based change point discovery algorithms. Even though we focused on comparing only SSA based algorithms, the proposed metric applicable to any kind of change point discovery algorithm and have the advantages of requiring no localization steps, and being independent of any predefined thresholds (unlike traditional metrics). I.
Discovering Causal Change Relationships Between Processes in Complex Systems
"... Abstract — Complex systems involve the interaction between many processes that may or may not have causal relations to each other. In such systems, discovering causal relations can provide significant insights into the internals of the system and facilitate fault discovery and recovery procedures. I ..."
Abstract

Cited by 3 (3 self)
 Add to MetaCart
(Show Context)
Abstract — Complex systems involve the interaction between many processes that may or may not have causal relations to each other. In such systems, discovering causal relations can provide significant insights into the internals of the system and facilitate fault discovery and recovery procedures. In this paper, we provide a novel causality detection algorithm based on robust singular spectrum transform that combines features of autoregressive modeling and perturbation analysis. The proposed approach was evaluated using both synthetic and real data and was shown to provide superior performance to the standard linear Grangercausality test. It also provides a natural way to detect common causes that may give false positives in other causality tests. I.
Mining Causal Relationships in Multidimensional Time Series
"... Abstract. Time series are ubiquitous in all domains of human endeavor. They are generated, stored, and manipulated during any kind of activity. The goal of this chapter is to introduce a novel approach to mine multidimensional timeseries data for causal relationships. The main feature of the propos ..."
Abstract

Cited by 2 (2 self)
 Add to MetaCart
(Show Context)
Abstract. Time series are ubiquitous in all domains of human endeavor. They are generated, stored, and manipulated during any kind of activity. The goal of this chapter is to introduce a novel approach to mine multidimensional timeseries data for causal relationships. The main feature of the proposed system is supporting discovery of causal relations based on automatically discovered recurring patterns in the input time series. This is achieved by integrating a variety of data mining techniques. The main insight of the proposed system is that causal relations can be found more easily and robustly by analyzing meaningful events in the time series rather than by analyzing the time series numerical values directly. The RSST (Robust Singular Spectrum Transform) algorithm is used to find interesting points in every time series that is further analyzed by a constrained motif discovery algorithm (if needed) to learn basic events of the time series. The Grangercausality test is extended and applied to the multidimensional timeseries describing the occurrences of these basic events rather than to the raw timeseries data. The combined algorithm is evaluated using both synthetic and real world data. The real world application is to mine records of activities during a humanrobot interaction experiment in which a human subject is guiding a robot to navigate using free hand gesture. The results show that the combined system can provide causality graphs representing the underlying relations between the human’s actions and robot behavior that cannot be recovered using standard causal graph learning procedures.
Mining Abnormal Patterns from Heterogeneous TimeSeries with Irrelevant Features for Fault Event Detection
, 2008
"... We address the issue of detecting fault events in multivariate time series. We suppose the following realistic situation: A) the features to which multivariate time series correspond are heterogeneous; B) relative to a large number of normal examples, only a small number of examples of fault events ..."
Abstract

Cited by 2 (0 self)
 Add to MetaCart
We address the issue of detecting fault events in multivariate time series. We suppose the following realistic situation: A) the features to which multivariate time series correspond are heterogeneous; B) relative to a large number of normal examples, only a small number of examples of fault events are available in advance; and C) many features irrelevant to fault events are included. In such a situation, we require realtime, highaccuracy processing. We propose an algorithm to resolve the issue. Key ideas in it include: 1) transforming the timeseries for each feature into a sequence of anomaly scores, in order to map heterogeneous features to homogeneous features (an anomaly score indicates the degree of anomaly relative to an ordinal sequence) and then representing the pattern of a fault event in terms of anomaly score vectors; 2) selecting features specifying a fault event by means of iterative optimization using both normal and fault anomaly score vectors. We then monitor the degree of abnormal with regard to test anomaly score vectors by matching with the abnormal patterns. We demonstrate the effectiveness of our proposed algorithm through an application to an actual automobile fault diagnosis data set.