Sequential pattern mining of multimodal data streams in dyadic interactions

Sequential pattern mining is used to find frequent data sequences over time. When sequential patterns are generated, the newly arriving patterns may not be identified as frequent sequential patterns due to the existence of old data and sequences. Progressive sequential pattern mining aims to find the most up-to-date sequential patterns given that obsolete items will be deleted from the sequences. When sequences come with multiple data streams, it is difficult to maintain and update the current sequential patterns. Even worse, when we consider the sequences across multiple streams, previous methods cannot efficiently compute the frequent sequential patterns. In this work, we propose an efficient algorithm PSP-AMS to address this problem. PSP-AMS uses a novel data structure PSP-MS-tree to insert new items, update current items, and delete obsolete items. By maintaining a PSP-MS-tree, PSP-AMS efficiently finds the frequent sequential patterns across multiple streams. The experimental results show that PSP-AMS significantly outperforms previous algorithms for mining of progressive sequential patterns across multiple streams on synthetic data as well as real data.

Sequential pattern mining is applicable in a wide range of applications since many types of data sets are in a time related format. Besides mining sequential patterns in a single dimension, mining multidimensional sequential patterns can give us more informative and useful patterns. Due to the huge increase in data volume and also quite large search space, efficient solutions for finding patterns in multidimensional sequence data are nowadays very important. For this reason, developing a parallel algorithm is necessary. In this paper, we present a multidimensional sequence model and a parallel algorithm follows the level-wise approach and all participating processors or workers generate candidate sequence and count their supports independently. Simulation experiments show good load balancing and scalable and acceptable speedup over different processors and problem sizes. dense data sets such as telecommunications, where there are many and long frequent patterns, performance of these algorithms degrades incredibly. On the one side, parallel and distributed computing is expected to relieve current mining methods from the sequential bottleneck, providing the ability to scale to massive datasets, and improving the response time. Achieving good performance on today's multiprocessor systems is a non-trivial task. The main challenges include synchronization and communication minimization, work-load balancing, find good data layout and data decomposition, and disk I/O minimization, which is especially important for data mining (4). On the other side, the basic single dimension can not satisfy the requirement of multi-attribute analysis, which is often the case in actual system practice. To address this problem, multidimensional sequence pattern mining is developed. The most time consuming operation in the discovery process of sequential patterns is the computation of the frequency of the occurrences of interesting subsequences in the sequence database. However, the number of sequential patterns grows exponentially and the task of finding all sequential patterns requires a lot of computational resources, which make it an ideal candidate for parallel processing. However, sequential pattern mining suffers from a scalability problem in both memory use and computational time when a dataset size is large. To perform mining on large datasets, we propose to parallelize this task. We demonstrate that our parallel algorithm can effectively alleviate the scalability problem. In this paper, we present a model for multidimensional sequence pattern and then propose a parallel algorithm for mining sequential patterns from multidimensional sequence data. The algorithm is based on the data parallelism strategy of pattern growth. The experimental results show that our parallel algorithms usually achieve good load balancing and scalability.

Supporting sequential pattern mining from data streams is nowadays a relevant problem in the area of data stream mining research. Actual proposals available in the literature are based on the well-known PrefixSpan approach and are, indeed, able to effectively bound the error of discovered patterns. This approach foresees the idea of dividing the target stream in a collection of manageable chunks, i.e., pieces of stream, in order to gain into effectiveness and efficiency. Unfortunately, mining patterns from stream chunks indeed introduce additional errors with respect to the basic application scenario where the target stream is mined continuously, in a non-batch manner. This is due to several reasons. First, since batches are processed individually, patterns that contain items from two consecutive batches are lost. Secondly, in most batch-based approaches, the decision about the frequency of a pattern is done locally inside a single batch. Thus, if a pattern is frequent in the stream but its items are scattered over different batches, it will be continuously pruned out and will never become frequent due to the algorithm’s lack of the “complete-picture” perspective. In order to address so-delineated pattern mining problems, this paper introduces and experimentally assesses BFSPMiner, a Batch-Free Sequential Pattern Miner algorithm for effectively and efficiently mining patterns in streams without being constrained to the traditional batch-based processing. This allows us, for instance, to discover frequent patterns that would be lost according to alternative batch-based stream mining processing models. We complement our analytical contributions by means of a comprehensive experimental campaign of BFSPMiner against real-world data stream sets and in comparison with current batch-based stream sequential pattern mining algorithms.

Incremental sequential patterns for multivariate temporal association rules mining

순차 패턴 탐색은 데이터 마이닝의 주요 기법 중의 하나로서 웹기반 시스템, 전자상거래, 생물정보학 및 USN 환경 등과 같은 여러 컴퓨터 응용 분야에서 생성되는 데이터를 효율적으로 분석하기 위하여 널리 활용되고 있다. 한편 이들 응용 분야에서 생성되는 정보들은 근래들어 한정적인 데이터 집합이 아닌 구성요소가 지속적으로 생성되는 데이터 스트림 형태로 생성되고 있다. 이러한 상황을 고려하여 데이터 스트림에서 순차패턴 탐색에 대한 연구들도 활발히 진행되고 있다. 하지만 이전의 연구들은 주로 분석 대상 데이터 스트림에서 단순 순차패턴을 구하는 과정에서 마이닝 수행 시간이나 메모리 사용량 등을 줄이는데 초점을 맞추고 있으며, 따라서 해당 데이터 스트림의 특성을 효율적으로 표현할 수 있는 보다 중요하고 의미있는 패턴들을 탐색하기 위한 연구는 거의 진행되지 못하고 있다. 본 논문에서는 데이터 스트림에서 보다 의미있는 순차패턴을 탐색하기 위한 방법으로 구성요소의 발생 간격 제한 조건을 활용한 빈발 순차패턴 탐색 방법을 제안한다. 먼저 발생 간격 정의 기준 및 발생 간격제한 빈발 순차패턴의 개념을 제시하고, 이어서 데이터 스트림에서 발생 간격 제한 조건을 적용하여 빈발 순차패턴을 효율적으로 탐색할 수 있는 마이닝 방법을 제안한다. Sequential pattern mining is one of the essential data mining tasks, and it is widely used to analyze data generated in various application fields such as web-based applications, E-commerce, bioinformatics, and USN environments. Recently data generated in the application fields has been taking the form of continuous data streams rather than finite stored data sets. Considering the changes in the form of data, many researches have been actively performed to efficiently find sequential patterns over data streams. However, conventional researches focus on reducing processing time and memory usage in mining sequential patterns over a target data stream, so that a research on mining more interesting and useful sequential patterns that efficiently reflect the characteristics of the data stream has been attracting no attention. This paper proposes a mining method of sequential patterns over data streams with a gap constraint, which can help to find more interesting sequential patterns over the data streams. First, meanings of the gap for a sequential pattern and gap-constrained sequential patterns are defined, and subsequently a mining method for finding gap-constrained sequential patterns over a data stream is proposed.

As an important branch of data mining, sequential patterns mining has been extensively studied. Based on sequential patterns mining, Structural Relation Patterns (SRPs) mining is proposed for mining relations among sequences, these relations are generally hidden behind sequential patterns. Upon the previous researches, the concepts of concurrent relation pattern and exclusive relation pattern are redefined; the definitions of ordered relation pattern and iterate relation pattern are given. The properties of SRPs are discussed, and they form a theoretical foundation for further study of structural relation patterns and relative mining algorithms. Beside, the thinking of mining associate relations among sequential patterns is proposed. SRPs mining is significant in practical applications same as sequential patterns mining.

Modeling human behavior on the Web is an essential goal of Web usage mining and is often performed by sequential pattern mining (SPM). The similarity between data elements and the variability of human behavior often result in the decrease of the support value of some patterns and thus limit the number of patterns mined. This is often associated with a loss of information. As the data can take the form of multiple data sources that represent multiple views of the area of interest, they can be used to handle this similarity and variability. Traditional approaches form a unique dataset from these data. However, the associated mining task is complex and generates a large and redundant set of sequential patterns. This work proposes G_SPM, a pattern mining algorithm designed for behavioral pattern mining that takes advantage of multi-source data to handle the similarity, while adopting a selective mining strategy to limit the complexity of the mining process and the increase in the number of patterns. It considers the behavioral data source as the main source, and exploits complementary sources only when similarity is suspected. It forms frequent general patterns that represent sets of similar behavioral patterns with a limited frequency, while controlling the level of generality. Experimental results confirm that G_SPM succeeds in mining general patterns and thus in handling the problem of item similarity. In addition G_SPM outperforms traditional approaches in terms of runtime and redundancy of the resulting set of patterns.

We present a data visualization approach that allows users to interactively analyse sequential movement patterns and automatically extract rules for marine animal behaviour. This enhances the ability of oceanic experts to understand marine ecosystems. Traditional time-series analysis, basic statistical methods, and single-view visualizations often fail to capture the intricate spatial and temporal dynamics of marine ecosystems. Our visual analytics approach, that leverages sequential pattern mining and rule mining techniques, integrates SPMF's Vertical Mining of Maximal Sequential Patterns (VMSP) algorithm and T-Rule Growth algorithm to efficiently extract and visualize movement patterns and behavioural rules of marine animals. The system also incorporates automated pre-processing to clean and prepare large datasets, reducing the time and effort required for manual handling. The visualization presents oceanic data from detection stations, allowing researchers to discover patterns in the movements of marine species. Our interface also provides geographical visualization of selected patterns or rules and helps users to navigate and find the relevant information. Key principles of our model include multi-view visualizations that provide comprehensive insights by displaying data from various perspectives simultaneously, and geo-spatial time-varying data visualization that captures and represents the dynamic movements of marine species. This approach allows domain experts to interactively explore and analyse sequential patterns and rules, enhancing their ability to make informed decisions. Additionally, feedback received, and insights gained from specialized professionals was used to determine if this approach would be a beneficial complement to existing practice.

In this paper, we present a new algorithm, Weighted Sequential support affinity pattern mining in which a new measure, sequential s-confidence is suggested. By using the measure, sequential patterns with support affinity are generated. A comprehensive performance study shows that WSAffinity is efficient and scalable in weighted sequential pattern mining. Moreover, it generates fewer but important sequential patterns.Keywordsdata miningsequential pattern miningsupport affinity pattern

In this paper the problem of Contiguous Item Sequential Pattern (CISP) Mining is presented as a sequential pattern mining problem under two constraints. First, each element in a sequence consists of only one item. Second, items appearing in the sequences that contain a pattern must be adjacent with respect to the underlying order as they appear in the pattern. Even though the problem of CISP mining can be solved by using previous approaches on sequential pattern mining under a general constraint description framework, this may lead to poor performance due to the large searching space. To efficiently solve this problem, a new data structure, UpDown Tree, is proposed for CISP mining. UpDown Tree based approach can greatly improve the efficiency of CISP mining in terms of both time and memory comparing to previous approaches. An extensive experimental study has shown promising results with our approach.

Structural Relation Patterns (SRPs) mining is proposed for mining relations among sequences, these relations are generally hidden behind sequential patterns. Concurrent Sequential Pattern (CSP) and Exclusive Sequential Pattern (ESP) are two important parts of SRP, called Structural Relation Sequential Pattern (SRSP). Upon the previous researches, the concepts of SRSP are redefined; the properties of SRSP are discussed; the algorithms for mining SRSPs are studied. All of these form a theoretical foundation for further study of structural relation patterns and relative mining algorithms. SRSPs mining is significant in practical applications same as sequential patterns mining.

Sequential pattern mining is an important mining task for discovering sequential patterns along with their insight relationships in many real-world applications. In practice, sequence databases are kept changing over the time along with their business. For some reasons, some sequences in the database are asked to be deleted from the database. In order to have a synchronization of discovered sequential patterns with the database from which they have been discovered, the sequential pattern mining task is re-considered with many challenges. As the number of deleted sequences is often smaller than the size of the entire database, re-mining from scratch the updated database might incur a high cost because sequential pattern mining is a computationally expensive task. In this paper, our work aims at an efficient incremental mining solution to the sequential pattern mining task with sequence deletions. Different from the existing works, we propose an expanded prefix tree by extending the existing prefix tree with additional structures for capturing more necessary information for the incremental mining process. Based on this tree, we propose an incremental sequential pattern mining algorithm, SPMD, for finding a complete set of sequential patterns with no re-scanning the original database, when a number of sequences in the database are deleted. Experimental results on the benchmark databases have confirmed that our SPMD algorithm outperforms the re-mining from scratch with the PrefixSpan algorithm with less running time.

Targeted mining of contiguous sequential patterns

Sequential pattern mining is one of the most studied and challenging tasks in data mining. However, the extension of well-known methods from many other classical patterns to sequences is not a trivial task. In this paper we study the notion of δ-freeness for sequences. While this notion has extensively been discussed for itemsets, this work is the first to extend it to sequences. We define an efficient algorithm devoted to the extraction of δ-free sequential patterns. Furthermore, we show the advantage of the δ-free sequences and highlight their importance when building sequence classifiers, and we show how they can be used to address the feature selection problem in statistical classifiers, as well as to build symbolic classifiers which optimizes both accuracy and earliness of predictions.

In data mining field, sequential pattern mining can be applied in divers applications such as basket analysis, web access patterns analysis, and quality control in manufactory engineering, etc. Many methods have been proposed for mining sequential patterns. However, conventional methods only consider the occurrences of itemsets in customer sequences. The sequential patterns discovered by these methods are called as positive sequential patterns, i.e., such sequential patterns only represent the occurrences of itemsets. In practice, the absence of a frequent itemset in a sequence may imply significant information. We call a sequential pattern as negative sequential pattern, which also represents the absence of itemsets in a sequence. The two major difficulties in mining sequential patterns, especially negative ones, are that there may be huge number of candidates generated, and most of them are meaningless. In this paper, we proposed a method for mining strong positive and negative sequential patterns, called PNSPM. In our method, the absences of itemsets are also considered. Besides, only sequences with high degree of interestingness will be selected as strong sequential patterns. An example was taken to illustrate the process of PNSPM. The result showed that PNSPM could prune a lot of redundant candidates, and could extract meaningful sequential patterns from a large number of frequent sequences.