Periodic Pattern Mining Research Articles

Periodic pattern mining in weighted dynamic networks

Periodic pattern mining in weighted dynamic networks

Supergraph based periodic pattern mining in dynamic social networks

In dynamic networks, periodically occurring interactions express especially significant meaning. However, these patterns also could occur infrequently, which is why it is difficult to detect while working with mass data. To identify such periodic patterns in dynamic networks, we propose single pass supergraph based periodic pattern mining SPPMiner technique that is polynomial unlike most graph mining problems. The proposed technique stores all entities in dynamic networks only once and calculate common sub-patterns once at each timestamps. In this way, it works faster. The performance study shows that SPPMiner method is time and memory efficient compared to others. In fact, the memory efficiency of our approach does not depend on dynamic network’s lifetime. By studying the growth of periodic patterns in social networks, the proposed research has potential implications for behavior prediction of intellectual communities.

On the Use of Rough Set Theory for Mining Periodic Frequent Patterns

This paper presents a new Apriori based approach for mining periodic frequent patterns from the temporal database. The proposed approach utilizes the concept of rough set theory for obtaining reduced representation of the initially considered temporal database. In order to consider only the relevant items for analyzing seasonal effects, a decision attribute festival has been considered. It has been observed that the proposed approach works fine for the analysis of the seasonal impact on buying behavior of customers. Considering the capability o f approach for the analysis of seasonal profitability concern, decision making, and future marketing may use it for the important decision - making process for the uplifting of sell.

Discovering Productive Periodic Frequent Patterns in Transactional Databases

Periodic frequent pattern mining is an important data mining task for various decision making. However, it often presents a large number of periodic frequent patterns, most of which are not useful as their periodicities are due to random occurrence of uncorrelated items. Such periodic frequent patterns would most often be detrimental in decision making where correlations between the items of periodic frequent patterns are vital. To enable mine the periodic frequent patterns with correlated items, we employ a correlation test on periodic frequent patterns and introduce the productive periodic frequent patterns as the set of periodic frequent patterns with correlated items. We finally develop the productive periodic frequent pattern (PPFP) framework for mining our introduced productive periodic frequent patterns. PPFP is efficient and the productiveness measure removes the periodic frequent patterns with uncorrelated items.

Mining and Analysis of Periodic Patterns in Weighted Directed Dynamic Network

Periodic patterns are mined individually on structural and weight aspects of an interaction in a dynamic network. However, these interactions possess a direction aspect too. Moreover, some applications require patterns on both aspects i) on direction and ii) on weight of directed interactions for a better understanding of their behaviour. To the authors' knowledge, no such work is available that mines both types of periodic patterns in a single run. To overcome this limitation, the authors propose a framework to mine periodic patterns on both the aspects. The framework first mines periodic patterns on direction, and then only the edges present in the patterns obtained are considered further for patterns on weight of directed interactions. Further, the patterns are being analysed to develop a better understanding of the dynamic network. To do so, a set of six parameters explained later in the text is proposed to study the behaviour of interactions at microscopic level. The framework is tested on real world and synthetic datasets. The results highlight its practical scalability and prove its efficiency.

A hash-based algorithm for mining non-redundant asynchronous periodic patterns

Periodic patterns occur in a wide variety of datasets like time series, temporal, spatiotemporal and biological datasets. Asynchronous periodic pattern mining discovers the patterns that appear significantly with strict periodicity in one or more subsequences called valid segments with tolerable disturbance between the valid segments. The state-of-the-art asynchronous periodic pattern mining algorithms proposed in the literature mine many redundant asynchronous periodic patterns. This paper proposes an algorithm to mine non-redundant asynchronous periodic patterns. It is a variation of a four-phase algorithm named SMCA. The proposed algorithm mines precise and concise set of non-redundant patterns whose number is in orders of magnitude smaller than that of previous methods. The proposed algorithm is explained using a hypothetical dataset and its performance is evaluated using three real data sets which are stock market data from Bombay Stock Exchange, Weather data from Cambridge University and Migratory Zebras' data from MoveBank.

A Hash-based Algorithm for Mining Non-redundant Asynchronous Periodic Patterns

Periodic patterns occur in a wide variety of datasets like time series, temporal, spatiotemporal and biological datasets. Asynchronous periodic pattern mining discovers the patterns that appear significantly with strict periodicity in one or more subsequences called valid segments with tolerable disturbance between the valid segments. The state-of-the-art asynchronous periodic pattern mining algorithms proposed in the literature mine many redundant asynchronous periodic patterns. This paper proposes an algorithm to mine non-redundant asynchronous periodic patterns. It is a variation of a four-phase algorithm named SMCA. The proposed algorithm mines precise and concise set of non-redundant patterns whose number is in orders of magnitude smaller than that of previous methods. The proposed algorithm is explained using a hypothetical dataset and its performance is evaluated using three real data sets which are stock market data from Bombay Stock Exchange, Weather data from Cambridge University and Migratory Zebras' data from MoveBank.

An efficient approach to mine flexible periodic patterns in time series databases

Periodic pattern mining in time series databases is one of the most interesting data mining problems that is frequently appeared in many real-life applications. Some of the existing approaches find fixed length periodic patterns by using suffix tree structure, i.e., unable to mine flexible patterns. One of the existing approaches generates periodic patterns by skipping intermediate events, i.e., flexible patterns, using apriori based sequential pattern mining approach. Since, apriori based approaches suffer from the issues of huge amount of candidate generation and large percentage of false pattern pruning, we propose an efficient algorithm FPPM (Flexible Periodic Pattern Mining) using suffix trie data structure. The proposed algorithm can capture more effective variable length flexible periodic patterns by neglecting unimportant or undesired events and considering only the important events in an efficient way. To the best of our knowledge, ours is the first approach that simultaneously handles various starting position throughout the sequences, flexibility among events in the mined patterns and interactive tuning of period values on the go. Complexity analysis of the proposed approach and comparison with existing approaches along with analytical comparison on various issues have been performed. As well as extensive experimental analyses are conducted to evaluate the performance of proposed FPPM algorithm using real-life datasets. The proposed approach outperforms existing algorithms in terms of processing time, scalability, and quality of mined patterns.

Predicting database workloads through mining periodic patterns in database audit trails

Information about the periodic changes of intensity and structure of database workloads plays an important role in performance tuning of functional components of database systems. Discovering the patterns in workload information, such as audit trails, traces of user applications, and sequences of dynamic performance views, is a complex and time-consuming task. This work investigates a new approach to analysis of information included in the database audit trails. In particular, it describes the transformations of information included in the audit trails into a format that can be used for discovering the periodic patterns in the fluctuations of database workloads. It presents an algorithm that finds elementary periodic patterns through nested iterations over a four-dimensional space of execution plans of SQL statements and positional parameters of the patterns. It proposes a collection of composition rules for the derivations of complex periodic patterns from the elementary and other complex patterns and it shows how to use such rules to predict the future workload levels.

Mining Association Rules in Dengue Gene Sequence with Latent Periodicity

The mining of periodic patterns in dengue database is an interesting research problem that can be used for predicting the future evolution of dengue viruses. In this paper, we propose an algorithm called Recurrence Finder (RECFIN) that uses the suffix tree for detecting the periodic patterns of dengue gene sequence. Also, the RECFIN finds the presence of palindrome which indicates the possibilities of formation of proteins. Further, this paper computes the periodicity of nucleic acid and amino acid sequences of any length. The periodicity based association rules are used to diagnose the type of dengue. The time complexity of the proposed algorithm is O(n2). We demonstrate the effectiveness of the proposed approach by comparing the experimental results performed on dengue virus serotypes dataset with NCBI-BLAST algorithm.

An Asynchronous Periodic Sequential Pattern Mining Algorithm with Multiple Minimum Item Supports for Ad Hoc Networking

The original sequential pattern mining model only considers occurrence frequencies of sequential patterns, disregarding their occurrence periodicity. We propose an asynchronous periodic sequential pattern mining model to discover the sequential patterns that not only occur frequently but also appear periodically. For this mining model, we propose a pattern-growth mining algorithm to mine asynchronous periodic sequential patterns with multiple minimum item supports. This algorithm employs a divide-and-conquer strategy to mine asynchronous periodic sequential patterns in a depth-first manner recursively. We describe the process of algorithm realization and demonstrate the efficiency and stability of the algorithm through experimental results.

Efficient Mining of Partial Periodic Patterns with Individual Event Support Thresholds Using Minimum Constraints

Partial periodic patterns are commonly seen in real-life applications and provide useful prediction with uncertainty. Most previous approaches have set a single minimum support threshold for all events to assume they have similar frequencies which is not practical for real-world applications. Instead of setting a single minimum support threshold for all events, Chen et al. proposed an FP-tree-like algorithm to allow multiple minimum supports for reflecting the natures of the events. However, such a tree-based algorithm encountered an efficiency problem while period length is long or event sequential orders in period segments are varied. Under the circumstance, many tree branches are created and much execution time is spent to find partial periodic patterns. In this paper, we thus propose a projection-based algorithm which examines only prefix subsequences and projects only corresponding postfix subsequences with multiple minimum supports to quickly find the partial periodic patterns in a recursive process. Experiments on both synthetic and real-life datasets show that the proposed algorithm is more efficient than the previous one.

Wireless Spectrum Occupancy Prediction Based on Partial Periodic Pattern Mining

Cognitive radio appears as a promising technology to allocate wireless spectrum between licensed and unlicensed users in an efficient way. When unlicensed users opportunistically utilize spectrum holes, prediction models that infer the availability of spectrum holes can help to improve the spectrum extraction rate and reduce the collision rate. In this paper, a spectrum occupancy prediction model based on Partial Periodic Pattern Mining (PPPM) is introduced. The mining aims at identifying frequent spectrum occupancy patterns that are hidden in the spectrum usage of a channel. The mined frequent patterns are then used to predict future channel states (i.e., busy or idle). Based on the prediction, unlicensed users are able to utilize spectrum holes aggressively without introducing significant interference to licensed users. PPPM outperforms traditional Frequent Pattern Mining (FPM) by considering real patterns that do not repeat perfectly due to noise, sensing errors, and irregular behaviors. Using real-world Wi-Fi and personal communication service (PCS) activities, we show a significant reduction on miss rate in channel state prediction. With the proposed prediction mechanism, the performance of Dynamic Spectrum Access (DSA) is substantially improved. Further, we extend the three-state PPPM to an N-state PPPM to predict the duration of high/low utilization in a channel. The frequent patterns of channel utilization duration are critical in optimizing channel switch strategies. The high prediction accuracy is validated with data collected in the paging bands.

Analysis of Various Periodicity Detection Algorithms in Time Series Data with Design of New Algorithm

Time series datasets consist of sequence of numeric values obtained over repeated measurements of time. They are Popular in many applications such as stock market analysis, power consumption, economic and sells forecasting, temperature etc. Periodic pattern mining or periodicity detection is process of finding periodic patterns in time series database. It has a number of applications, such as prediction, forecasting, detection of unusual activities, etc. Periodicity mining needs to give more attention as its increased need in real life applications. The types of periodicities are symbol periodicity, sequence periodicity and segment periodicity and they should be identified even in the presence of noise in the time series database. There are number of algorithms exists for periodic pattern mining. Those algorithms have some advantages and disadvantages. In this paper, we have compared different periodicity mining algorithms and given plan for developing efficient periodicity mining algorithm which detect symbol periodicity, sequence periodicity and segment periodicity and noise-resilient .

A two-phase approach for mining weighted partial periodic patterns

Partial periodic pattern mining has recently become an important issue in the field of data mining due to its wide applications in many businesses. A partial periodic pattern considers part of but not all the events within a specific period length, repeating with high frequency in an event sequence. Traditional partial periodic pattern mining, however, only considered the frequencies of patterns, but did not consider events that might have different importance. The study thus proposes a weighted partial periodic patterns mining algorithm to resolve this problem. To increase the efficiency, the two-phase upper-bound weighted model based on segmental maximum weights is adopted to prune unimportant candidates in early stage. Then the weighted partial periodic patterns are discovered from the candidate patterns. Finally, the experimental results on synthetic datasets and a real oil dataset show that the weighted partial periodic pattern mining is more practical to assist users for decision making.

Detecting P2P bots by mining the regional periodicity

Peer-to-peer (P2P) botnets outperform the traditional Internet relay chat (IRC) botnets in evading detection and they have become a prevailing type of threat to the Internet nowadays. Current methods for detecting P2P botnets, such as similarity analysis of network behavior and machine-learning based classification, cannot handle the challenges brought about by different network scenarios and botnet variants. We noticed that one important but neglected characteristic of P2P bots is that they periodically send requests to update their peer lists or receive commands from botmasters in the command-and-control (C&C) phase. In this paper, we propose a novel detection model named detection by mining regional periodicity (DMRP), including capturing the event time series, mining the hidden periodicity of host behaviors, and evaluating the mined periodic patterns to identify P2P bot traffic. As our detection model is built based on the basic properties of P2P protocols, it is difficult for P2P bots to avoid being detected as long as P2P protocols are employed in their C&C. For hidden periodicity mining, we introduce the so-called regional periodic pattern mining in a time series and present our algorithms to solve the mining problem. The experimental evaluation on public datasets demonstrates that the algorithms are promising for efficient P2P bot detection in the C&C phase.

Constraint Based Periodic Pattern Mining in Multiple Longest Common Subsequences

Constraint Based Periodic Pattern Mining in Multiple Longest Common Subsequences

A Framework for Periodic Outlier Pattern Detection in Time-Series Sequences

Periodic pattern detection in time-ordered sequences is an important data mining task, which discovers in the time series all patterns that exhibit temporal regularities. Periodic pattern mining has a large number of applications in real life; it helps understanding the regular trend of the data along time, and enables the forecast and prediction of future events. An interesting related and vital problem that has not received enough attention is to discover outlier periodic patterns in a time series. Outlier patterns are defined as those which are different from the rest of the patterns; outliers are not noise. While noise does not belong to the data and it is mostly eliminated by preprocessing, outliers are actual instances in the data but have exceptional characteristics compared with the majority of the other instances. Outliers are unusual patterns that rarely occur, and, thus, have lesser support (frequency of appearance) in the data. Outlier patterns may hint toward discrepancy in the data such as fraudulent transactions, network intrusion, change in customer behavior, recession in the economy, epidemic and disease biomarkers, severe weather conditions like tornados, etc. We argue that detecting the periodicity of outlier patterns might be more important in many sequences than the periodicity of regular, more frequent patterns. In this paper, we present a robust and time efficient suffix tree-based algorithm capable of detecting the periodicity of outlier patterns in a time series by giving more significance to less frequent yet periodic patterns. Several experiments have been conducted using both real and synthetic data; all aspects of the proposed approach are compared with the existing algorithm InfoMiner; the reported results demonstrate the effectiveness and applicability of the proposed approach.

Constraint Based Periodic Pattern Mining in Multiple Longest Common Subsequences

The periodicity search in longest common subsequences in multiple strings has a number of application, is an interesting data mining problem. Periodicity is very common practice in longest common subsequence mining algorithm. This work introduces a new parallel algorithm for finding periodicity in multiple strings. Few existing algorithms lacks in poor scalability, lacks in finding all longest pattern, and for finding symbol, partial and full periodicity. We designed the algorithm using FP-tree for finding periodicity for most common longest substring in multiple sources. We introduce a parallel algorithm for Constraint Based Periodic Pattern Mining (CBPPM) algorithm, which takes O( kN ) for finding periodicity and O((N×L×h)/p) time for MLCS pattern. We tested parallel algorithm on a coarse-grained multi-computer (BSP/CGM) model with p<√m processors that takes O(N×L√p) space per processor, with O(log p) communication rounds. We derive a practical implementation that works better for arbitrary length of input sequence. The algorithm is noise resilient, and shown its performance in presence of replacement, insertion, deletion, or mixture of these types of noise. We experimented with synthetic and real data reveals a near linear speedup with scalable performance. The comparative study shows algorithm's applicability and effectiveness, generally more noise resilient.

Projection-based partial periodic pattern mining for event sequences

Partial periodic pattern mining is one of the important issues in the field of data mining due to its practical applications. A partial periodic pattern consists of some periodic and non-periodic events in a specific period length, and is repeated with high frequency in an event sequence. In the past, a max-subpattern hit set algorithm was developed to discover partial periodic patterns, but its drawback is spending a large amount of time in calculating frequency counts from the redundant candidate nodes. In this study, we thus adopt an efficient encoding strategy to speed up the efficiency of processing period segments in an event sequence, and combined with the projection method to quickly find the partial periodic patterns in the recursive process. Finally, the experimental results show the superior performance of the proposed approach.