Sequential Pattern Mining

The election of organizational leaders, especially in organizations whose
members have diverse backgrounds and interests, can cause various
problems. Problems in the selection of school organization leaders include
the absence of an objective selection of organizational leadership candidates
because they are selected based on comparisons between candidates without
considering the criteria in the selection of organizational leadership
candidates. Research related to the multi-attributive border approximation
area comparison (MABAC) and simplified pivot pairwise relative criteria
importance assessment (PIPRECIA-S) methods has never been conducted so
far, so it is a reference in conducting this research using the MABAC and
PIPRECIA-S methods. This study aims to select the head of the school
organization using the MABAC method and PIPRECIA-S weighting can
increase the objectivity of the criteria assessment results by relying on
calculations from the PIPRECIA-S weighting method. Based on the
selection results using the MABAC method and PIPRECIA-S weighting,
candidate 1 was recommended as the leader of the school organization
because it achieved rank 1 with a total score of 0.293. The contribution of
this research is to help in the selection of the head of the organization using the PIPRECIA-S and MABAC methods as a decision-making solution.

A parallel algorithm based on a lattice theoretic approach is proposed to find the rules among patterns that sustain sequential nature in the multi-stream time series data. The human motion data captured by motion capturing system considered as multi-stream multidimensional data is used as real time data set. The data set is transformed into sequences of symbols of lower dimension due to its complex nature. The relations among symbols are expressed as "rules". The proposed algorithm is implemented on a Distributed Shared Memory (DSM) multiprocessors system. The experimental results justify the efficiency of finding rules from the sequences of the patterns for time series data by achieving significant speed up comparing with the previous reported algorithm.

On December 10, 2001 the U.S. Department of under the regulatory arrangements of the DOC, Commerce (DOC) imposed steep antidumping duties Argentina's beekeepers never had a chance of defending against honey imports from Argentina and China ranging themselves. For example, responding to the DOC's froIm 32.6 percent to 183.8 percent, and a countervailing lengthy and sophisticated questionnaires that sought to duty against Argentina of 5.9 percent. A previous determine cost of production went beyond the capacities antidumping investigation in 1995 ended with a of poor beekeepers. In the absence of information, the suspension "agreement" that curtailed U.S. imports from DOC resorted to evidence presented by the petitioners China by around 30 percent. Millions of beekeepers which was riddled with errors. The available evidence around the world, most of them poor, make a living suggests that had beekeepers been capable of responding from honey production, and a free and competitive to the questionnaires, the margin of dumping would had world market would help raise their standards of living. been lower, if at all existent. This and other evidence Nevertheless, the sequential pattern of increasing and discussed by Nogues suggest the urgent need to widening protectionism followed by the United States, introduce reforms into the World Trade Organization the world's top importer, to include successful exporters antidumping and subsidy agreements. At the minimum under the effects of its contingent protection measures what is required is a consensus that all respondents be sends a clear message that other countries should think given the same opportunity by the international trade twice before investing in expandiig honey exports to the rules. The author argues that at present this is not the United States. case and offers suggestions for reforms. In addition to looking into the trade effects of these contingent protection measures, Nogues concludes that This paper-a product of Trade, Development Research Group-is part of a larger effort in the group to improve trade policy work for development. Copies of the paper are available free from the World Bank,

Conceptual clustering is a discovery process that groups a set of data in the way that the intra-cluster similarity is maximized and the inter-cluster similarity is minimized. Traditional clustering algorithms employ some measure of distance between data points in n-dimensional space. However, not all data types can be represented in a metric space, therefore no natural distance function is available for them. We address the problem of clustering sequences of categorical values. We present a measure of similarity for the sequences and an agglomerative hierarchical algorithm that uses frequent sequential patterns found in the database to efficiently generate the resulting clusters. The algorithm iteratively merges smaller, similar clusters into bigger ones until the requested number of clusters is reached.

Due to the increasing use of mobile phones and their increasing capabilities, huge amount of usage and location data can be collected. Location prediction is an important task for mobile phone operators and smart city administrations to provide better services and recommendations. In this work, we propose a sequence mining based approach for location prediction of mobile phone users. More specifically, we present a modified Apriori-based sequence mining algorithm for the next location prediction, which involves use of multiple support thresholds for different levels of pattern generation. The proposed algorithm involves a new support definition, as well. We have analyzed the behaviour of the algorithm under the change of threshold through experimental evaluation and the experiments indicate improvement in comparison to conventional Apriori-based algorithm.

The abundance of information on online purchasing websites makes it challenging for customers to locate products that match their preferences. However, the cold-start problem arises when there isn't enough previous data, making it harder to make accurate recommendations for new customers or products. The enormous number of possible customers and products in a recommendation system leads to sparse data, which makes it harder to generate relevant recommendations and causes the sparsity problem. In addition, existing e-commerce recommender systems have difficulty making accurate product recommendations because they disregard individual consumer characteristics. In order to overcome these limitations, a hybrid recommender system combining ontology and sequential pattern mining (SPM) techniques is proposed. The strategy entails constructing an ontology that encompasses customer and product-related knowledge in the e-commerce domain. This ontology is then utilized to calculate customer preference similarities and generate predictions for the intended customer. The SPM algorithm is applied to the results of collaborative filtering to generate personalized recommendations for the customer. Experiments have demonstrated that the hybrid recommender system outperforms existing methods and resolves the cold-start problem and data sparsity in e-commerce recommender systems effectively. Even with limited initial data, the system generates accurate and individualized recommendations based on ontological domain knowledge and the customer's sequential purchase patterns. By integrating ontology and sequential pattern mining, this strategy enhances the precision and individualization of the e-commerce industry's recommendation process.

Sequence mining intends to discover rules from diverse datasets by implementing Rule Mining Algorithms with efficient data structures and data retrieval techniques. Traditional algorithms struggle in handling variable support measures which may involve repeated reconstruction of the underlying data structures with changing thresholds. To address these issues the premiere Sequence Mining Algorithm, AprioriAll is implemented against an Educational and a Financial Dataset, using the HASH and the TRIE data structures with scan reduction techniques. Primary idea is to study the impact of data structures and retrieval techniques on the rule mining process in handling diverse datasets. Performance Evaluation Matrices- Support, Confidence and Lifts are considered for testing the efficacies of the algorithm in terms of memory requirements and execution time complexities. Results unveil the excellence of Hashing in tree construction time and memory overhead for fixed sets of pre-defined support thresholds. Whereas, TRIE may avoid reconstruction and is capable of handling dynamic support thresholds, leading to shorter rule discovery time but higher memory consumption. This study highlights the effectiveness of Hash and TRIE data structures considering the dataset characteristics during rule mining. It underscores the importance of appropriate data structures based on dataset features, scanning techniques, and user-defined parameters.

This paper addresses the problem of discovering a sound Workflow net (WFN) from sample event traces representing the behavior of a process. A current challenging problem deals with discovering a suitable model from incomplete logs since it is not possible to ensure that a log contains all executions of a process. A method that allows building WFN from logs including a reduced number of traces is presented. It is based on the concept of invariant occurrence between activities, which yields sets of activities named conjoint occurrence classes, which are used to infer the causal and concurrent behavior not exhibited in the log. Procedures derived from the method have been implemented in the ProM framework. Experimental results show that the algorithm can discover models form logs including less traces compared to other standing discovery algorithms.

In laser hair removal, it is preferred for many clinical and technical reasons to use small to medium size laser spotsizes. However, manually aiming a small laser beam hundreds of times to cover a larger skin area can lead to uneven coverage and can result in missed areas and excess heating due to pulse stacking. For this reason, large beam diameters are sometimes used for hair removal. This can lead to ineffective treatments, increased occurrence of side effects and more pain to the patient. The latest Fotona SOE technology allows practitioners to apply small to medium beams without sacrificing the safety and efficacy of the treatment. This is achieved by utilizing computer controlled laser scanner mirrors to automatically place the laser beam over a larger skin area in a perfect non-sequential pattern.

We propose a higher order mining (HOM) approach for modelling, monitoring and analyzing district heating (DH) substations' operational behaviour and performance. HOM is concerned with mining over patterns rather than primary or raw data. The proposed approach uses a combination of different data analysis techniques such as sequential pattern mining, clustering analysis, consensus clustering and minimum spanning tree (MST). Initially, a substation's operational behaviour is modeled by extracting weekly patterns and performing clustering analysis. The substation's performance is monitored by assessing its modeled behaviour for every two consecutive weeks. In case some significant difference is observed, further analysis is performed by integrating the built models into a consensus clustering and applying an MST for identifying deviating behaviours. The results of the study show that our method is robust for detecting deviating and sub-optimal behaviours of DH substations. In addition, the proposed method can facilitate domain experts in the interpretation and understanding of the substations' behaviour and performance by providing different data analysis and visualization techniques.

A real world challenging task of the web master of an organization is to match the needs of user and keep their attention in their web site. So, only option is to capture the intuition of the user and provide them with the recommendation list. Web usage mining is a kind of data mining method that provide intelligent personalized online services such as web recommendations, it is usually necessary to model users" web access behavior. Web usage mining includes three process, namely, preprocessing, pattern discovery and pattern analysis. After the completion of these three phases the user can find the required usage patterns and use this information for the specific needs. The data abstraction is achieved through data preprocessing. The aim of discovering frequent sequential access patterns in Web log data is to obtain information about the navigational behavior of the users. In the proposed system, an efficient sequential pattern mining algorithm is used to identify frequent sequential web access patterns. The access patterns are retrieved from a Graph, which is then used for matching and generating web links for recommendations.

Association rule mining may be a crucial and area of the data mining methodology. It aims at finding some hidden information or relationships among the attributes of the data. Association rule mining may be a stimulating area of data mining analysis that discovers correlations between wholly different item sets throughout event information. To enhance the efficiency and accuracy of the proposed work this is incorporating additional measures in simple Genetic Algorithmic rule for such data processing issues. We have used Genetic algorithmic rule to optimize the generated rules and take fitness operate for the improvement and realize the optimum solutions. To compare these two implementations on the generation and fitness they needed in proposed work.

The World Wide Web provides colossal volume of data to the internet users and grows at a rapid pace every day. The server creates log files regarding the page, IP address of the user, agent, operating system and time stamp and this data is mined to extract useful information using web usage mining. The primary objective of this paper is to find the browsing behavior pattern of the users from the log files using a novel algorithm named "Frequent Pages Constellation Algorithm". The FPC Algorithm groups the frequent pages browsed by the users and the results produced is utilized for personalization, forecast market vogues, enhances marketing strategies, improve website page positioning, site restricting etc. The experimental results showcased that the proposed FPC algorithm generates very less candidates, takes very less

A great number of individuals are anxious to exploit the internet's wealth of information. It can be employed to further enhance the existing data. However, the primary challenge lies in uncovering the valuable information that is concealed within HTML elements. This study proposes a framework for web usage mining that examines web server log files using sequential pattern mining approaches. Web log patterns reveal information about user behavior, preferences, and website interactions. Preprocessing of the web data was carried out. The primary objective of preprocessing is to enhance data integrity while decreasing the volume of information that requires evaluation. Prior to inputting the data into the pattern discovery phase, it is necessary to eliminate noise by resolving the challenge of distinguishing between different users and sessions. To identify frequent sequential access in large, low-support data sets, a method for mining sequential patterns is developed. A sequential pattern mining technique identifies recurring sequential patterns in multidimensional web log files with minimum support provided. Multidimensional sequential pattern mining is primarily concerned with enhancing the standard of the patterns the user received back. PrefixSpan algorithm has been used to extract tabular as well as unstructured data from HTML tag. Prefix prunes some web info by calculating the support value at different nodes in the represented projected sub-database and snipe away huge portions of the representation that are guaranteed not to create any outcomes. The system is implemented in Matlab programming language. In the domain of web mining, Matlab has been employed to extract valuable information from the web, including user records and content. When mining extensive sequences containing numerous records, in particular, the method substantially reduces execution time and eliminates enormous memory access costs. The PrefixSpan algorithm enhanced with the starting position and innertagcount parameters has better performance than Markov model and GSP algorithm with execution time of 2.35seconds.

The signalling mechanism of monetary policy is of vital importance as it conveys the central bank’s assessment of the economy and its future outlook. In the Indian context, changes in the policy environment
since the later half of the 1990s have brought about shifts in the operating procedure of monetary policy. As a result, beside the traditional instruments, new indirect instruments have emerged as tools of monetary
policy signalling. Against this backdrop, we examine the efficacy and robustness of alternative monetary policy instruments in transmitting policy signals and its impact on financial market behaviour. Employing
a SVAR model, we ascertain whether the gradual emphasis on indirect instruments have facilitated the task of conveying the monetary policy stance and also provide evidence of asymmetric response of financial
markets to monetary policy shocks.

Sequential pattern mining is advantageous for several applications for example, it finds out the sequential purchasing behavior of majority customers from a large number of customer transactions. However, the existing researches in the field of discovering sequential patterns are based on the concept of frequency and presume that the customer purchasing behavior sequences do not fluctuate with change in time, purchasing cost and other parameters. To acclimate the sequential patterns to these changes, constraint are integrated with the traditional sequential pattern mining approach. It is possible to discover more user-centered patterns by integrating certain constraints with the sequential mining process. Thus in this paper, monetary and compactness constraints in addition to frequency and length are included in the sequential mining process for discovering pertinent sequential patterns from sequential databases. Also, a CFML-PrefixSpan algorithm is proposed by integrating these constraints with the original PrefixSpan algorithm, which allows discovering all CFML sequential patterns from the sequential database. The proposed CFML-PrefixSpan algorithm has been validated on synthetic sequential databases. The experimental results ensure that the efficacy of the sequential pattern mining process is further enhanced in view of the fact that the purchasing cost, time duration and length are integrated with the sequential pattern mining process.

There has been an increase in our capabilities of both generating and collecting data with the progress of humanity and development of technology. The ultimate intent of this massive data collection is to utilize it for different reasons that include process efficiency, system automation and competitive benefits among others. This is achieved by determining both implicit and explicit unidentified patterns in data that can direct the process of decision making. Data mining is a process of finding rules and patterns that describe the characteristic properties of the data. Sequential pattern mining is one of the x Chapter 7 discusses the need of fuzziness in incremental mining to give more simple and practical sequential patterns. The concept of fuzzy sequential pattern summarization using member functions and knowledge of linguistics is explained. The role of lattice structure for fuzzy pattern summarization is further discussed. A data structure called ProgresLattice that is based on standard lattice structure is introduced in this chapter. An algorithm that makes use of fuzziness, proposed Progreslattice and incremental mining is proposed in Chapter 7 to mine fuzzy sequential patterns. The algorithm is explained with the help of example along with its stepwise execution. This efficiency of this algorithm is compared with another fuzzy sequential pattern mining algorithm called as Nancy, L.P. et al.s' algorithm. Chapter 8 concludes this thesis by giving the application of the proposed algorithms. A model called ‚Constraint guided progressive sequential mining waterfall model for CRM (CGP-SMW)" is proposed in this chapter that shows the stages in CRM application where the use of these proposed algorithms can be done. The chapter further gives various application of any algorithm for incremental mining of sequential patterns from progressive databases. The limitation of this study and future research direction in this area is also suggested.

The problem of sequential pattern mining is one of the several issues that have deserved particular attention on the general problem of data mining. Despite the important developments in the last years, the best algorithm in the area (PrefixSpan) does not deal with gap constraints and consequently doesn't allow for the introduction of background knowledge into the process. In this paper we present the generalization of the PrefixSpan algorithm to deal with gap constraints, using a new method to generate projected databases. Studies on performance and scalability were conducted in synthetic and real-life datasets, and the respective results are presented.

Indonesia is a country with the world's largest tropical peatlands of about 14.9 million hectares that have important roles that support life. Unfortunately, there were many fires in peatlands. In Kalimantan, peatland fires which are characterized by hotspot occurrences reached an average of 25.1% in this decade. According to experts and field forest fire fighters, fire hotspots that appear in a sequence of two to three days at the same location has a high potential of becoming a forest fire. This study aims to determine the sequential patterns of hotspots occurrence, classify satellite image data and identify the fire spot. Fire spot identification was done using hotspots sequences patterns that were overlaid with burned area classification results. Sequential pattern mining using the Prefix Span algorithm was applied to identify sequences of hotspot occurrence. Meanwhile, classification using Maximum Likelihood method was applied to satellite image Landsat 7 to identify burned areas in Pulang Pisau and Palangkaraya Central Kalimantan and Pontianak in West Kalimantan. Furthermore, sequence patterns were overlaid with image classification results. The results show that in Pulang Pisau, 26.19% of sequences patterns are located in burned areas and 72.62% sequence patterns are in the buffer of burned area within a radius of one kilometer. As for Palangkaraya, there are 62.50% sequences patterns are located in burned areas and 87.50% sequence patterns are in the buffer of burned area with the radius of one kilometer. It is concluded that in Pulang Pisau and Palangkaraya there are respectively 72.62% and 87.50% fire hotspots which are strong indicators of peatland fires.

Temporal facts, the facts for characterizing events that hold in specific time periods, are attracting rising attention in the knowledge graph (KG) research communities. In terms of quality management, the introduction of time restrictions brings new challenges to maintaining the temporal consistency of KGs and detecting potential temporal conflicts. Previous studies rely on manually enumerated temporal constraints to detect conflicts, which are labor-intensive and may have granularity issues. We start from the common pattern of temporal facts and constraints and propose a pattern-based temporal constraint mining method, PaTeCon. PaTeCon uses automatically determined graph patterns and their relevant statistical information over the given KG instead of human experts to generate time constraints. Specifically, PaTeCon dynamically attaches class restriction to candidate constraints according to their measuring scores.We evaluate PaTeCon on two large-scale datasets based on Wikidata and Freebase respectively. The experimental results show that pattern-based automatic constraint mining is powerful in generating valuable temporal constraints.

In recent years, recommendation systems have become vital for businesses to enhance customer satisfaction and generate revenue across various domains, such as e-commerce and entertainment. Deep learning techniques have significantly improved the accuracy and efficiency of these systems. However, there is a lack of comprehensive literature reviews that classify and summarize the latest deep learning techniques used in recommendation systems. Additionally, some existing review papers either overlook state-of-the-art techniques or focus narrowly on specific domains. To address these research gaps, this systematic review comprehensively analyzes the literature on deep learning techniques in recommendation systems, specifically employing term classification. The study highlights the emerging research area of domain classification, which has shown promising results in applying deep learning techniques to domains such as social networks, e-commerce, and e-learning. This review offers insights into the deep learning techniques used across different recommendation systems and provides suggestions for future research. This work fills a critical research gap and serves as a valuable resource for researchers and practitioners interested in deep learning techniques for recommendation systems.

Tourists share opinions about Points of Interest (POIs) through online posts and social media platforms. Opinion mining is a popular technique for extracting feedback from tourists who visited various places hidden in reviews, which are used in several tourist applications that generally reflect their preference towards POI. On the other hand, a trip schema is difficult for tourists because they must pick up sequential POIs in unknown areas that meet their limitations and preferences. However, most prior trip suggestion methods are suboptimal for several reasons, including that they do not consider valuable user reviews and rely exclusively on left-to-right unidirectional discovery sequence models. This study proposes a Neural Network-Long Short-Term Memory (LSTM) POI recommendation system for calculating user similarity based on opinions and preferences. In addition, it presents a method for discovering sequential trip recommendations with Bidirectional Encoder Representations from...

Upon integration of internet and its various applications and increase of internet pages, access to information in search engines becomes difficult. To solve this problem, web page recommendation systems are used. In this paper, recommender engine are improved and web usage mining methods are used for this purpose. In recommendation system, clustering was used for classification of users’ behavior. In fact, we implemented usage mining operation on the data related to each user for making its movement pattern. Then, web pages were recommended using neural network and markov model. So, performance of recommendation engine was improved using user’s movement patterns and clustering and neural network and Markov model, and obtained better results than other methods. To predict the data recovery quality on web, two factors including accuracy and coverage were used

A search was implemented for a microbial lipase capable of bioconverting a diester (dimethyl 5-(3-(2-(7-chloroquinolin-2-yl)ethyl)phenyl)4,6dithianonanedioate) to its S-ester acid, an intermediate in the production of Verhikast (a leukotriene receptor antagonist). Required properties of the sought-after enzyme included a high enantiomeric selectivity (e.e. > 98 %), the formation of only trace amounts of diacid and a high bioconversion rate. This search yielded 57 lipase-producing microorganisms, 18 of which presented detectable bioconversion activity. Thirteen of these microbes were selected for further study based upon their lipase production level and enzyme stability at harvest. Despite their common enzymatic property, namely the hydrolysis of triglycerides, these lipase preparations presented diverse ester acid specific synthesis rates (from < 0.01/~g/unit/h to 0.98 #g/unit/h) and diacid formation levels (from 0 K to 35~). One of these microbes, identified as Pseudomonas aeruginosa (strain MB 5001), was found to produce a lipase having all of the above-listed required properties. The initial fermentation process developed in shake flasks was rapidly and successfully scaled up in 23-liter laboratory bioreactors, achieving a maximum production of 35 tmits/ml of lipase after 48 h of cultivation.

Text mining is a technique that helps the user to find useful information from a large amount of digital text document. It is a challenging issue to find accurate knowledge (or features) in text documents to help users to find what they want. Most existing text mining methods adopted term-based approaches, but they all suffer from the problem of polysemy and synonymy. Text mining is a technique that helps the user to find useful information from a large amount of digital text document. It is a challenging issue to find accurate knowledge (or features) in text documents to help users to find what they want. Most existing text mining methods adopted term-based approaches, but they all suffer from the problem of polysemy and synonymy. The effective usage and updating of discovered patterns is still an open research issue. Pattern deploying and pattern evolving method has also been proposed in order to refine the patterns that help in improving the effectiveness of pattern discovery. Th...

Web page recommendation system is an application of Web Usage Mining (WUM) approach, which specializes in predicting the user next browsing activity in real-time Web for personalized recommendations. To date, many works have been addressed in investigating the use of data mining techniques (e.g., Clustering) in Web page application. Most of the research efforts are utilized partitional clustering algorithms to discover user profiles in order to obtain a better quality of recommendations. However, the quality of current solutions has only managed to achieve accuracy within the range of 60-70%. This happens due to the weaknesses of partitive algorithm criterion in Web page recommendation system to overcome the overlapping profiles, which require more attention. In order to tackle above problem, a modified algorithm for Web page recommendation is proposed. The ultimate goal is to improve the recommendation quality which leads to increase the prediction accuracy. Hence, this study carri...

As the information available on the World Wide Web is increasing day-by-day, access to the websites is also increasing which results in huge amount of web log data (also called web usage data). Discovery and analysis of useful information from these web logs become a practical necessity. Frequent access pattern, which is the sequence of accesses pursued by users frequently, is one of the interesting and useful knowledge in practice. Web access pattern tree (WAP-tree) mining is a frequent pattern mining technique for web log access sequences, which first stores the original web access sequence database on a prefix tree for storing non-sequential data. WAP-tree algorithm then, mines the frequent sequences from the WAP-tree by recursively reconstructing intermediate trees, starting with suffix sequences and ending with prefix sequences. In this paper, we propose a more efficient algorithm named eWAP-mine (enhanced web access pattern mining algorithm), which is based directly on the initial conditional web access sequence base (1-CWASD) of each frequent event and eliminates the need for reconstructing intermediate conditional WAP-trees.

We study the problem of finding the k most frequent items in a stream of items for the recently proposed max-frequency measure. Based on the properties of an item, the maxfrequency of an item is counted over a sliding window of which the length changes dynamically. Besides being parameterless, this way of measuring the support of items was shown to have the advantage of a faster detection of bursts in a stream, especially if the set of items is heterogeneous. The algorithm that was proposed for maintaining all frequent items, however, scales poorly when the number of items becomes large. Therefore, in this paper we propose, instead of reporting all frequent items, to only mine the top-k most frequent ones. First we prove that in order to solve this problem exactly, we still need a prohibitive amount of memory (at least linear in the number of items). Yet, under some reasonable conditions, we show both theoretically and empirically that a memory-efficient algorithm exists. A prototype of this algorithm is implemented and we present its performance w.r.t. memory-efficiency on real-life data and in controlled experiments with synthetic data.

We propose a streaming algorithm, based on the minimal description length (MDL) principle, for extracting non-redundant sequential patterns. For static databases, the MDL-based approach that selects patterns based on their capacity to compress data rather than their frequency, was shown to be remarkably effective for extracting meaningful patterns and solving the redundancy issue in frequent itemset and sequence mining. The existing MDL-based algorithms, however, either start from a seed set of frequent patterns, or require multiple passes through the data. As such, the existing approaches scale poorly and are unsuitable for large datasets. Therefore, our main contribution is the proposal of a new, streaming algorithm, called Zips, that does not require a seed set of patterns and requires only one scan over the data. For Zips, we extended the Lempel-Ziv (LZ) compression algorithm in three ways: first, whereas LZ assigns codes uniformly as it builds up its dictionary while scanning the input, Zips assigns codewords according to the usage of the dictionary words; more heaviliy used words get shorter code-lengths. Secondly, Zips exploits also non-consecutive occurences of dictionary words for compression. And, third, the well-known space-saving algorithm is used to evict unpromising words from the dictionary. Experiments on one synthetic and two real-world large-scale datasets show that our approach extracts meaningful compressing patterns with similar quality to the state-of-the-art multi-pass algorithms proposed for static databases of sequences. Moreover, our approach scales linearly with the size of data streams while all the existing algorithms do not.

Association rules are commonly used to provide decision‐makers with knowledge that helps them to make good decisions. Most of the published proposals mine association rules without paying particular attention to temporal information. However, in real‐life applications data usually change over time or presenting different temporal situations. Therefore, the extracted knowledge may not be useful, since we may not know whether the rules are currently applicable or whether they will be applicable in the future. For this reason, in recent years, many methods have been proposed in the literature for mining temporal association rules, which introduce a greater predictive and descriptive power providing an additional degree of interestingness. One of the main problems in this research field is the lack of visibility most works suffer since there is no standard terminology to refer to it, making it difficult to find and compare proposals and studies in the field. This contribution attempts t...

Background: Biclustering is a critical task for biomedical applications. Order-preserving biclusters, submatrices where the values of rows induce the same linear ordering across columns, capture local regularities with constant, shifting, scaling and sequential assumptions. Additionally, biclustering approaches relying on pattern mining output deliver exhaustive solutions with an arbitrary number and positioning of biclusters. However, existing order-preserving approaches suffer from robustness, scalability and/or flexibility issues. Additionally, they are not able to discover biclusters with symmetries and parameterizable levels of noise. Results: We propose new biclustering algorithms to perform flexible, exhaustive and noise-tolerant biclustering based on sequential patterns (BicSPAM). Strategies are proposed to allow for symmetries and to seize efficiency gains from item-indexable properties and/or from partitioning methods with conservative distance guarantees. Results show BicSPAM ability to capture symmetries, handle planted noise, and scale in terms of memory and time. BicSPAM also achieves the best match-scores for the recovery of hidden biclusters in synthetic datasets with varying noise distributions and levels of missing values. Finally, results on gene expression data lead to complete solutions, delivering new biclusters corresponding to putative modules with heightened biological relevance. Conclusions: BicSPAM provides an exhaustive way to discover flexible structures of order-preserving biclusters. To the best of our knowledge, BicSPAM is the first attempt to deal with order-preserving biclusters that allow for symmetries and that are robust to varying levels of noise. Background Biclustering tasks over real-value matrices aim to discover sub-matrices (biclusters) where a subset of rows exhibit a correlated pattern over a subset of columns. However, existing approaches impose the selection of specific patterns of correlation, which often leads to incomplete solutions. A simple yet powerful direction to accommodate more flexible patterns-order-preserving patterns-was introduced by Ben-Dor et al. [1]. A bicluster is order-preserving if there is a permutation of its columns under which the sequence of values in every row is strictly increasing. These biclusters capture biclusters with shifting and scaling patterns of gene expression, and are, additionally, critical to detect other meaningful profiles as the progression of a disease or cellular response in distinct stages. Order-preserving biclustering can be applied

Sequential pattern mining (SPM) under gap constraint is a challenging task. Many efficient specialized methods have been developed but they are all suffering from a lack of genericity. The Constraint Programming (CP) approaches are not so effective because of the size of their encodings. In [7], we have proposed the global constraint PREFIX-PROJECTION for SPM which remedies to this drawback. However, this global constraint cannot be directly extended to support gap constraint. In this paper, we propose the global constraint GAP-SEQ enabling to handle SPM with or without gap constraint. GAP-SEQ relies on the principle of right pattern extensions. Experiments show that our approach clearly outperforms both CP approaches and the state-of-the-art cSpade method on large datasets.

Sequential pattern mining under constraints is a challenging data mining task. Many efficient ad hoc methods have been developed for mining sequential patterns, but they are all suffering from a lack of genericity. Recent works have investigated Constraint Programming (CP) methods, but they are not still effective because of their encoding. In this paper, we propose a global constraint based on the projected databases principle which remedies to this drawback. Experiments show that our approach clearly outperforms CP approaches and competes well with ad hoc methods on large datasets.

Sequential pattern mining under various constraints is a challenging data mining task. The paper provides a generic framework based on constraint programming to discover sequence patterns defined by constraints on local patterns (e.g., gap, regular expressions) or constraints on patterns involving combination of local patterns such as relevant subgroups and top-k patterns. This framework enables the user to mine in a declarative way both kinds of patterns. The solving step is done by exploiting the machinery of Constraint Programming. For complex patterns involving combination of local patterns, we improve the mining step by using dynamic CSP. Finally, we present two case studies in biomedical information extraction and stylistic analysis in linguistics.

Iconicity can help with a complex aspect of language learning, i.e., mapping a specific word-form onto its referent, allowing for faster language learning. Previous experiments have focused on iconicity in simplistic, individual word learning contexts. However, language learners often face high ambiguity in real-world learning contexts. We explored how context impacts the way iconic signs are learned with the understanding that iconicity may not always be advantageous. In a sign language crosssituational word-learning experiment, we manipulated the degree of similarity between iconic properties of signs and semantic categories of referents. We found iconic over arbitrary signs were more helpful for learning regardless of learning environment. However, when looking at only iconic signs across different learning environments, learners were less accurate at learning signs in the same iconic neighborhood (i.e., phonologically related in an iconic way such as signs that occur on the nose for nosebased signs), but not when signs were unrelated, or semantically similar (i.e., both signs referring to two similar animals). Learners were slowest and least accurate when the signs were both semantically and iconically similar to each other. These results highlight implications for real-world word learning, demonstrating that iconicity is less privileged for learning in some contexts.

Data mining methods extract knowledge from huge amounts of data. Recently with the explosion of mobile technologies, a new type of data appeared. The resulting databases can be described as spatiotemporal data in which spatial information (e.g., the location of an event) and temporal information (e.g., the date of the event) are included. In this article, we focus on spatiotemporal patterns extraction from this kind of databases. These patterns can be considered as sequences representing changes of events localized in areas and its near surrounding over time. Two algorithms are proposed to tackle this problem: the first one uses a priori strategy and the second one is based on pattern-growth approach. We have applied our generic method on two different real datasets related to: 1) pollution of rivers in France; and 2) monitoring of dengue epidemics in New Caledonia. Additionally, experiments on synthetic data have been conducted to measure the performance of the proposed algorithms.