New Type Of Pattern Research Articles

Surprising and novel multivariate sequential patterns using odds ratio for temporal evolution in healthcare

BackgroundPattern mining techniques are helpful tools when extracting new knowledge in real practice, but the overwhelming number of patterns is still a limiting factor in the health-care domain. Current efforts concerning the definition of measures of interest for patterns are focused on reducing the number of patterns and quantifying their relevance (utility/usefulness). However, although the temporal dimension plays a key role in medical records, few efforts have been made to extract temporal knowledge about the patient’s evolution from multivariate sequential patterns.MethodsIn this paper, we propose a method to extract a new type of patterns in the clinical domain called Jumping Diagnostic Odds Ratio Sequential Patterns (JDORSP). The aim of this method is to employ the odds ratio to identify a concise set of sequential patterns that represent a patient’s state with a statistically significant protection factor (i.e., a pattern associated with patients that survive) and those extensions whose evolution suddenly changes the patient’s clinical state, thus making the sequential patterns a statistically significant risk factor (i.e., a pattern associated with patients that do not survive), or vice versa.ResultsThe results of our experiments highlight that our method reduces the number of sequential patterns obtained with state-of-the-art pattern reduction methods by over 95%. Only by achieving this drastic reduction can medical experts carry out a comprehensive clinical evaluation of the patterns that might be considered medical knowledge regarding the temporal evolution of the patients. We have evaluated the surprisingness and relevance of the sequential patterns with clinicians, and the most interesting fact is the high surprisingness of the extensions of the patterns that become a protection factor, that is, the patients that recover after several days of being at high risk of dying.ConclusionsOur proposed method with which to extract JDORSP generates a set of interpretable multivariate sequential patterns with new knowledge regarding the temporal evolution of the patients. The number of patterns is greatly reduced when compared to those generated by other methods and measures of interest. An additional advantage of this method is that it does not require any parameters or thresholds, and that the reduced number of patterns allows a manual evaluation.

Detection of common subtrees with identical label distribution

Frequent pattern mining is a relevant method to analyse structured data, like sequences, trees or graphs. It consists in identifying characteristic substructures of a dataset. This paper deals with a new type of patterns for tree data: common subtrees with identical label distribution. Their detection is far from obvious since the underlying isomorphism problem is graph isomorphism complete. An elaborated search algorithm is developed and analysed from both theoretical and numerical perspectives. Based on this, the enumeration of patterns is performed through a new lossless compression scheme for trees, called DAG-RW, whose complexity is investigated as well. The method shows very good properties, both in terms of computation times and analysis of real datasets from the literature. Compared to other substructures like topological subtrees and labelled subtrees for which the isomorphism problem is linear, the patterns found provide a more parsimonious representation of the data.

Identifying novelties and anomalies for incremental learning in streaming time series forecasting

Time series data can be defined as a chronological sequence of observations on a variable of interest. A streaming time series is a time series that arrives continuously at high speed and has a data distribution that may change over time. Streaming time series data usually comes from electronic devices such as sensors and many of the applications dealing with streaming data in Industry 4.0 require real-time responses. Performing real-time forecasting offers the possibility to consider new types of patterns in the incoming streaming data, which is not possible when working with batch models. This paper presents a new approach to detect novelties and anomalies in real-time using a nearest-neighbors based forecasting algorithm. The algorithm works with an offline base model that is updated as stream data arrives following an incremental learning approach. It detects unknown patterns called novelties and anomalies. Novelties are included in the model in an online way and anomalies trigger an alarm as they present unexpected behaviors that need to be specifically analyzed. The algorithm has been tested with Spanish electricity demand data. Results show that the prediction errors obtained when the model is updated considering novelties and anomalies are lower than the errors obtained when the model is not updated. Thus, the model adjusts in real-time to the new patterns of data providing accurate errors and real-time predictions.

A goal-oriented framework for ontology reuse

Ontologies have been successfully used to assign semantics in the Semantic Web context, to support integration of data from different systems or different sources, and to enable reasoning. However, building ontologies is not a trivial task. Ontology reuse can help in this matter. The search and selection of ontologies to be reused should consider the alignment between their scope and the scope of the ontology being developed. In this paper, we discuss how goal modeling can be helpful in this context and we present GO-FOR, a framework in which goals are the central elements to promote ontology reuse. GO-FOR comprises a conceptual architecture, a goal-oriented ontology development process and a supporting tool. In GO-FOR we introduce Goal-Oriented Ontology Patterns (GOOPs) as a new type of pattern to be applied to develop ontologies in a goal-oriented approach. Results of the use of GO-FOR to build an ontology used to integrate patient examination data are also shown in this paper.

Rules and exceptions: A Tolerance Principle account of the possessive suffix in Northern East Cree.

Debate around inflectional morphology in language acquisition has contrasted various rule- versus analogy-based approaches. This paper tests the rule-based Tolerance Principle (TP) against a new type of pattern in the acquisition of the possessive suffix -im in Northern East Cree. When possessed, each noun type either requires or disallows the suffix, which has a complex distribution throughout the lexicon. Using naturalistic video data from one adult and two children - Ani (2;01-4;03) and Daisy (3;08-5;10) - this paper presents two studies. Study 1 applies the TP to the input to extrapolate two possible sets of nested rules for -im and make predictions for child speech. Study 2 tests these predictions and finds that each child's production of possessives over time is largely consistent with the predictions of the TP. This paper finds the TP can account for the acquisition of the possessive suffix and discusses implications for language science and Cree language communities.

Evolution pattern mining on dynamic social network

Recently, due to the popularity of social websites and apps, considerable attention has been paid to the analysis of the structure of social networks. Clearly, social networks usually evolve over time; some new users and relationships are established; and some obsolete ones are removed. This dynamic feature definitely increases the complexity of pattern discovery. In this paper, we introduce a new representation to express the dynamic social network and a new type of pattern, the evolution pattern, to capture the interaction evolutions in a dynamic social network. Furthermore, a novel algorithm, evolution pattern miner (EPMiner), is developed to efficiently discover the evolution characteristics. EPMiner also employs some pruning strategies to effectively reduce the search space to improve the performance. The experimental results on several datasets show the efficiency and the scalability of EPMiner for extracting interaction evolution in dynamic networks. Finally, we apply EPMiner on real datasets to show the practicability of evolution pattern mining.

A New Artificial Immune System Based on Continuous Learning for Pattern Recognition

This paper presents a novel approach for pattern recognition based on continuous training inspired by the biological immune system operation. The main objective of this paper is to present a method capable of continually learn, i.e., being able to address new types of patterns without the need to restart the training process (artificial immune system with incremental learning). It is a useful method for solving problems involving a permanent knowledge extraction, e.g., 3D facial expression recognition, whose quality of the solutions is strongly dependent on a continuous training process. In this context, two artificial immune algorithms are employed: (1) the negative selection algorithm, which is responsible for the pattern recognition process and (2) the clonal selection algorithm, which is responsible for the learning process. The main application of this method is in assisting in decision-making on problems related to pattern recognition process. To evaluate and validate the efficiency of this method, the system has been tested on handwritten character recognition, which is a classic problem in the literature. The results show efficiency, accuracy and robustness of the proposed methodology.

Tristable Phenomenon in a Predator–Prey System Arising from Multiple Limit Cycles Bifurcation

In this paper, we consider a predator–prey system with Holling type III ratio-dependent functional response. Such a system can exhibit complex dynamical behavior such as bistable and tristable phenomena which contain equilibria and oscillating motions for certain parameter values. In particular, we show that the ratio-dependent predator–prey system can exhibit multiple limit cycles due to Hopf bifurcation, giving rise to coexistence of stable equilibria and stable periodic solutions. These solutions may reveal some new type of patterns of complex dynamical behaviors in predator–prey systems.

Prediction of spatial distribution of invasive alien pests in two-dimensional systems based on a discrete time model

Preventing and managing the invasion of alien species poses significant challenges. In this article, we studied the population density of invasive alien pests based on a discrete time model, and found a new type of pattern and a new type of pattern formation mechanism. This new pattern is caused by the uncertainty of two states in the system (caused by chaos), which is different from the Turing instability. At last, the formation conditions of the irregular speckle pattern were quantitatively analyzed. We believe that the findings of this paper can provide theoretical guidance for the prevention and management of the invasion of alien pests.

Frequent State Transition Patterns of Multivariate Time Series

Sequence pattern discovery is a key issue in multivariate time series analysis. Popular approaches first obtain the pattern of each single-variate time series and then obtain cross-variate associations. In this paper, we consider different variables at the same time during pattern construction, and propose a new type of pattern called State Transition pAttern with Periodic wildcard gaps (STAP). Compared to previous types, STAP reveals stronger cross associations among different variables and provides better interpretability for decision makers. We design an approach with two stages, namely frequent state discovery and pattern synthesis, to obtain frequent STAPs. We propose two pre-pruning and an Apriori-pruning techniques to speed up pattern discovery. We also propose two post-pruning techniques to simplify the output and a visualization way to support expert decision. Experimental results on four real-world datasets demonstrate 1) STAP captures the cross and temporal associations; 2) the five pruning and pattern synthesis techniques are quite effective; and 3) visualization technique greatly increases the readability of STAP.

Frequent pattern discovery with tri-partition alphabets

The concept of patterns is the basis of sequence analysis. There are various pattern definitions for biological data, texts, and time series. Inspired by the methodology of three-way decisions and protein tri-partition, this paper proposes a frequent pattern discovery algorithm for a new type of pattern by dividing the alphabet into strong, medium, and weak parts. The new type, called a tri-pattern, is more general and flexible than existing ones and is therefore more interesting in applications. Experiments were undertaken on data in various fields to reveal the universality of this new pattern. These include protein sequence mining, petroleum production time series analysis, and forged Chinese text keyword mining. The results show that tri-patterns are more meaningful and desirable than the existing four types of patterns. This study enriches the semantics of sequential pattern discovery and the application fields of three-way decisions.

Pattern Matching Performance Analysis Based on Linear Models with Information Backscattered from RFID Tags

Pattern matching localization algorithms are commonly proposed in complex indoor environments. Received signal strength indicator (RSSI) of radio signals backscattered from passive ultra high frequency radio frequency identification (RFID) tags is commonly used as the pattern in matching process. Considering RSSI is severely affected by multipath propagation and noise, etc., more types of patterns will be a better choice for further study. In recent years, phase information is becoming more and more important for RFID-based systems. As for the period ambiguity problem of wrapped phase information, phase difference of arrival (PDOA) of double-frequency signals can be utilized as a type of pattern instead of phase of arrival corresponding to the propagation distance. In this paper, we combine RSSI and PDOA as a brand new type of pattern named RP. In previous study, only localization accuracy is considered to prove effective. It is hard to show the matching performance of the type of pattern with the following process of localization algorithms. Here we take advantage of the linear relationship between pattern Euclidean distance and geographic locations Euclidean distance of tags fixed in reference points to evaluate the distinctiveness performance of RP in matching. The linear correlation of the model is tested at each reference point with experimental data captured from indoor scenario. By evaluating these models statistically, we think RP is better than RSSI and PDOA as a pattern.

Mining Frequent Itemsets in Correlated Uncertain Databases

Recently, with the growing popularity of Internet of Things (IoT) and pervasive computing, a large amount of uncertain data, e.g., RFID data, sensor data, real-time video data, has been collected. As one of the most fundamental issues of uncertain data mining, uncertain frequent pattern mining has attracted much attention in database and data mining communities. Although there have been some solutions for uncertain frequent pattern mining, most of them assume that the data is independent, which is not true in most real-world scenarios. Therefore, current methods that are based on the independent assumption may generate inaccurate results for correlated uncertain data. In this paper, we focus on the problem of mining frequent itemsets over correlated uncertain data, where correlation can exist in any pair of uncertain data objects (transactions). We propose a novel probabilistic model, called Correlated Frequent Probability model (CFP model) to represent the probability distribution of support in a given correlated uncertain dataset. Based on the distribution of support derived from the CFP model, we observe that some probabilistic frequent itemsets are only frequent in several transactions with high positive correlation. In particular, the itemsets, which are global probabilistic frequent, have more significance in eliminating the influence of the existing noise and correlation in data. In order to reduce redundant frequent itemsets, we further propose a new type of patterns, called global probabilistic frequent itemsets, to identify itemsets that are always frequent in each group of transactions if the whole correlated uncertain database is divided into disjoint groups based on their correlation. To speed up the mining process, we also design a dynamic programming solution, as well as two pruning and bounding techniques. Extensive experiments on both real and synthetic datasets verify the effectiveness and efficiency of the proposed model and algorithms.

Externally controlled anisotropy in pattern-forming reaction-diffusion systems.

The effect of centrifugal forces is analyzed in a pattern-forming reaction-diffusion system. Numerical simulations conducted on the appropriate extension of the Oregonator model for the Belousov-Zhabotinsky reaction show a great variety of dynamical behaviors in such a system. In general, the system exhibits an anisotropy that results in new types of patterns or in a global displacement of the previous one. We consider the effect of both constant and periodically modulated centrifugal forces on the different types of patterns that the system may exhibit. A detailed analysis of the patterns and behaviors observed for the different parameter values considered is presented here.

Discovery of Temporal Associations in Multivariate Time Series

Multivariate time series are common in many application domains, particularly in industrial processes with a large number of sensors installed for process monitoring and control. Often, such data encapsulate complex relations among individual series. This paper presents a new type of patterns in multivariate time series, referred to as temporal associations, to capture a wide range of local relations along and across individual series. A scalable algorithm is developed to discover frequent associations by incorporating (1) redundancy pruning of patterns in single time series and (2) two conditions to avoid over-counting the occurrences of associations, thus greatly reducing the space and runtime complexity of the discovery process. A statistical significance measure is also introduced for ranking and post-pruning discovered associations. To evaluate the proposed method, synthetic data sets and a real world data set taken from the time series mining repository as well as a large data set obtained from a delayed coking plant are used. The experiments demonstrated that the discovered associations capture the local relations in multiple time series and that the proposed method is scalable to large data sets.

Enterprise security pattern: a new type of security pattern

ABSTRACTIn recent years, most organizations have suffered attacks against their information systems. For this reason, organizations should seek support from enterprise security architectures (ESAs) in order to secure their information assets. Security patterns can help when building complex ESAs, but they have some limitations that reduce their usability. In this paper, we define the metapattern of a new type of security pattern called Enterprise Security Pattern. This new metapattern provides a model‐driven environment and combines all elements that must be considered when designing and building ESAs. We present here a precise meta‐model and four diagrams to describe the metapattern of the enterprise security patterns. When avoiding a security problem, organizations could use enterprise security patterns to provide their designers with an optimal and proven security guideline and so standardize the design and building of the ESA for that problem. Enterprise security patterns could also facilitate the selection and tailoring of security policies, patterns, mechanisms, and technologies when a designer is building ESAs. To illustrate our ideas, we present an instance of this new type of pattern, showing how it can be used. Copyright © 2014 John Wiley & Sons, Ltd.

Pattern Based Feature Construction in Semantic Data Mining

The authors propose a new method for mining sets of patterns for classification, where patterns are represented as SPARQL queries over RDFS. The method contributes to so-called semantic data mining, a data mining approach where domain ontologies are used as background knowledge, and where the new challenge is to mine knowledge encoded in domain ontologies, rather than only purely empirical data. The authors have developed a tool that implements this approach. Using this the authors have conducted an experimental evaluation including comparison of our method to state-of-the-art approaches to classification of semantic data and an experimental study within emerging subfield of meta-learning called semantic meta-mining. The most important research contributions of the paper to the state-of-art are as follows. For pattern mining research or relational learning in general, the paper contributes a new algorithm for discovery of new type of patterns. For Semantic Web research, it theoretically and empirically illustrates how semantic, structured data can be used in traditional machine learning methods through a pattern-based approach for constructing semantic features.

Rotation and flipping robust region binary patterns for video copy detection

Many video fingerprints have been proposed to handle the video transformations problems when the original contents are copied and redistributed. However, most of them did not take into account flipping and rotation transformations. In this paper, we propose a novel video fingerprint based on region binary patterns, aiming to realize robust and fast video copy detection against video transformations including rotation and flipping. We extract two complementary region binary patterns from several rings in keyframes. These two kinds of binary patterns are converted into a new type of patterns for the proposed video fingerprint which is robust against rotation and flipping. The experimental results demonstrated that the proposed video fingerprint is effective for video copy detection particularly in the case of rotation and flipping. Furthermore, our experimental results proved that the proposed method allows for high storage efficiency and low computation complexity, which is suitable for practical video copy system.

Discovery of fuzzy quantitative sequential patterns with multiple minimum supports and adjustable membership functions

The objective of mining quantitative sequential patterns is to discover complete sets of sequential patterns with quantities in databases. Although this novel type of pattern considers information, compared to traditional sequential patterns, it contains a sharp boundary problem; that is, when the quantity of an item is near the boundary of two predetermined quantity intervals, it can be either ignored or overemphasized. Hong et al. proposed a new type of pattern, called a fuzzy quantitative sequential pattern (FQSP), to solve this problem. Although this type of pattern provides summary information to expedite decision making, FQSPs confront unrealistic circumstances, that is, they offer only a unique minimum support (min_sup) to all items for the whole database. A higher min_sup fails to find rare items with higher profit, whereas a lower min_sup leads to a combinatorial explosion problem. Instead of this type of mechanism, we used the idea of multiple minimum supports to mine an FQSP. Furthermore, the utility of FQSP is also disturbed because this approach uses only a single membership function without considering the price–quantity relation. The idea of adjustable membership functions was also provided to address this problem. This study proposes a new model to discover an FQSP with both multiple minimum supports and adjustable membership functions. Experiments using synthetic and real datasets demonstrated the computational efficiency, scalability, and effectiveness of the model.

Mining flipping correlations from large datasets with taxonomies

In this paper we introduce a new type of pattern -- a flipping correlation pattern. The flipping patterns are obtained from contrasting the correlations between items at different levels of abstraction. They represent surprising correlations, both positive and negative, which are specific for a given abstraction level, and which "flip" from positive to negative and vice versa when items are generalized to a higher level of abstraction. We design an efficient algorithm for finding flipping correlations, the Flipper algorithm, which outperforms naïve pattern mining methods by several orders of magnitude. We apply Flipper to real-life datasets and show that the discovered patterns are non-redundant, surprising and actionable. Flipper finds strong contrasting correlations in itemsets with low-to-medium support, while existing techniques cannot handle the pattern discovery in this frequency range.