Extract Frequent Patterns Research Articles

An Improved Algorithm for Extracting Frequent Gradual Patterns

Frequent gradual pattern extraction is an important problem in computer science widely studied by the data mining community. Such a pattern reflects a co-variation between attributes of a database. The applications of the extraction of the gradual patterns concern several fields, in particular, biology, finances, health and metrology. The algorithms for extracting these patterns are greedy in terms of memory and computational resources. This clearly poses the problem of improving their performance. This paper proposes a new approach for the extraction of gradual and frequent patterns based on the reduction of candidate generation and processing costs by exploiting frequent itemsets whose size is a power of two to generate all candidates. The analysis of the complexity, in terms of CPU time and memory usage, and the experiments show that the obtained algorithm outperforms the previous ones and confirms the interest of the proposed approach. It is sometimes at least 5 times faster than previous algorithms and requires at most half the memory.

Association Rule Mining for Dynamic Error Classification in the Automotive Manufacturing Industry

Association rule mining (ARM) is a data mining technique that extracts relationships between variables in dataset. This paper presents an application of ARM for dynamic error classification in the automotive manufacturing industry. The method involves the extraction of frequent patterns from historical vehicle configuration data to build a rule-based system for error classification. The presented approach includes frequent item sets, rule mining and the estimation of conditional error probabilities. The results show that ARM is a promising approach for dynamic error classification in the automotive manufacturing industry and has potential to improve product quality.

Improving the Efficiency of Distributed Utility Item Sets Mining in Relation to Big Data

High utility pattern mining is an analytical approach used to identify sets of items that exceed a specific threshold of utility values. Unlike traditional frequency-based analysis, this method considers user-specific constraints like the number of units and benefits. In recent years, the importance of making informed decisions based on utility patterns has grown significantly. While several utility-based frequent pattern extraction techniques have been proposed, they often face limitations in handling large datasets. To address this challenge, we propose an optimized method called improving the efficiency of Distributed Utility itemsets mining in relation to big data (IDUIM). This technique improves upon the Distributed Utility item sets Mining (DUIM) algorithm by incorporating various refinements. IDUIM effectively mines item sets of big datasets and provides useful insights as the basis for information management and nearly real-time decision-making systems. According to experimental investigation, the method is being compared to IDUIM and other state algorithms like DUIM, PHUI-Miner, and EFIM-Par. The results demonstrate the IDUIM algorithm is more efficient and performs better than different cutting-edge algorithms.

Profiling Public Transit Passenger Mobility Using Adversarial Learning

It is important to capture passengers’ public transit behavior and their mobility to create profiles, which are critical for analyzing human activities, understanding the social and economic structure of cities, improving public transportation, assisting urban planning, and promoting smart cities. In this paper, we develop a generative adversarial machine learning network to characterize the temporal and spatial mobility behavior of public transit passengers, based on massive smart card data and road network data. The Apriori algorithm is extended with spatio-temporal constraints to extract frequent transit mobility patterns of individual passengers based on a reconstructed personal trip dataset. This individual-level pattern information is used to construct personalized feature vectors. For regular and frequent public transit passengers, we identify similar transit mobility groups using spatio-temporal constraints to construct a group feature vector. We develop a generative adversarial network to embed public transit mobility of passengers. The proposed model’s generator consists of an auto-encoder, which extracts a low-dimensional and compact representation of passenger behavior, and a pre-trained sub-generator containing generalization features of public transit passengers. Shenzhen City is taken as the study area in this paper, and experiments were carried out based on smart card data, road network data, and bus GPS data. Clustering analysis of embedding vector representation and estimation of the top K transit destinations were conducted, verifying that the proposed method can profile passenger transit mobility in a comprehensive and compact manner.

CIP- Efficient Method for Mining Frequent Itemsets from Data streams using Landmark Window Model

Continuous stream transactions like network monitoring, retail market data analysis and stock market prediction need the “frequent patterns” to be detected recurrently. Literature suggests that several pattern mining solutions are being developed over years. Still lot of challenges need to be addressed due to rapidness in generation of continuous, unbounded and ordered data real time. Hence extraction of frequent patterns from recent data will improve the analysis of stream data. In this article, a new landmark window model CIP (candidate indexing and pruning) is considered for mining the datasets. CIP allows us to mine over entire history of data streams, which improves the accuracy. This article also proposes the candidate indexed sub (CIS)-tree scheme to extract the essential information from each incoming transactions of data streams. Our proposal is compared with the existing “improved data stream mining” (ISDM) for maximal frequent itemsets algorithm. Extensive experimental analyses prove the superiority of the proposed CIP over the popular ISDM in terms of accuracy and time complexity for high-speed data stream. This article also covers up a case study where the proposed approach is applied for an application called “web prefetching”.

Frequent Event Pattern Extraction of Drilling Time Series Using Change Point Detection and Event Sequence Generation

In drilling processes, non-stationary phases corresponding to shifts between operating conditions and changes in downhole formations typically lead to false alarms. Extracting these frequent event patterns is critical to build drilling process monitoring and fault diagnosis models. This study aims to extract the frequent event patterns associated with non-stationary phases in drilling time series. In this way, diversified information related to signal changes under normal conditions can be obtained, which is beneficial for suppressing false alarms and improving fault detection performance. The main contributions of this study are twofold: 1) a non-stationary phase detection method is proposed to extract drilling frequent event patterns based on t-distributed stochastic neighbor embedding and relative unconstrained least-squares importance fitting; 2) an event sequence generation method is proposed to express drilling frequent event patterns with a group of symbols. The effectiveness of the proposed method is demonstrated by data from a real drilling project.

Hybrid Recommender System for Mental Illness Detection in Social Media Using Deep Learning Techniques.

Recommender systems are chiefly renowned for their applicability in e-commerce sites and social media. For system optimization, this work introduces a method of behaviour pattern mining to analyze the person's mental stability. With the utilization of the sequential pattern mining algorithm, efficient extraction of frequent patterns from the database is achieved. A candidate sub-sequence generation-and-test method is adopted in conventional sequential mining algorithms like the Generalized Sequential Pattern Algorithm (GSP). However, since this approach will yield a huge candidate set, it is not ideal when a large amount of data is involved from the social media analysis. Since the data is composed of numerous features, all of which may not have any relation with one another, the utilization of feature selection helps remove unrelated features from the data with minimal information loss. In this work, Frequent Pattern (FP) mining operations will employ the Systolic tree. The systolic tree-based reconfigurable architecture will offer various benefits such as high throughput as well as cost-effective performance. The database's frequently occurring item sets can be found by using the FP mining algorithms. Numerous research areas related to machine learning and data mining are fascinated by feature selection since it will enable the classifiers to be swift, more accurate, and cost-effective. Over the last ten years or so, there have been significant technological advancements in heuristic techniques. These techniques are beneficial because they improve the search procedure's efficiency, albeit at the potential sacrifice of completeness claims. A new recommender system for mental illness detection was based on features selected using River Formation Dynamics (RFD), Particle Swarm Optimization (PSO), and hybrid RFD-PSO algorithm is proposed in this paper. The experiments use the depressive patient datasets for evaluation, and the results demonstrate the improved performance of the proposed technique.

Efficient mining of intra-periodic frequent sequences

Frequent Sequence Mining (FSM) is a fundamental task in data mining. Although FSM algorithms extract frequent patterns, they cannot discover patterns that periodically appear in the data. However, periodic trends are found in many areas such as market basket analysis, where discovering itemsets periodically purchased by customers can help understand periodic customer behavior. This is the task of Periodic Frequent Pattern Mining (PFPM). A major limitation common to traditional PFPM algorithms is that they reduce the periodicity between non-disjoint itemsets. They do not take into account the periods between disjoint itemsets. Thus, they find itemsets that appear periodically, but would fail to find a periodic appearance of distinct itemsets. To address this limitation, this paper extends the traditional problem of FSM with intra-periodicity and provides a theoretical background to extract intra-periodic frequent sequences. This leads to a new mining algorithm called Intra-Periodic Frequent Sequence Miner. Experimental results confirm its efficiency.

Discovering Patterns of Cardiovascular Disease and Diabetes in Myocardial Infarction Patients Using Association Rule Mining

Highlights: Association Rule Mining tools predict the association of early-onset Myocardial Infarction with Hypertension and Diabetes Mellitus. Association Rule Mining tools using clinical and biochemical attributes can predict the development of Hypertension and Diabetes Mellitus in Myocardial Infarction patients. Abstract: Cardiovascular diseases (CVDs) are a major cause of mortality in diabetic patients. Hypertensive patients are more likely to develop diabetes and hypertension contributes to the high prevalence of CVDs, in addition to dyslipidemia and smoking. This study was to find the different patterns and overall rules among CVD patients, including rules broken down by age, sex, cholesterol and triglyceride levels, smoking habits, myocardial infarction (MI) type on ECG, diabetes, and hypertension. The cross-sectional study was performed on 240 subjects (135 patients of ST-elevation MI below 45 years and 105 age matched controls). Association rule mining was used to detect new patterns for early-onset myocardial infarction. A hotspot algorithm was used to extract frequent patterns and various promising rules within real medical data. The experiment was carried out using "Weka'', a tool for extracting rules to find out the association between different stored real parameters. In this study, we found out various rules of hypertension like “Rule 6” says that if levels of BP Systolic > 131 mmHg, LpA2 > 43.2 ng/ml, hsCRP > 3.71 mg/L, initial creatinine > 0.5 mg/dl, and initial Hb ≤15 g/dl (antecedent), then the patient will have 88% chance of developing hypertension (consequent). Similarly for diabetes mellitus with finding their lift and confidence for different support like “Rule 6”, if MI type on ECG = ’Inferior Wall MI’ with STATIN=No, and levels of Triglycerides ≤325 (antecedent), then the patient had a 67% chance of developing diabetes mellitus. We concluded that early-onset myocardial infarction is significantly associated with hypertension and diabetes mellitus.Using association rule mining, we can predict the development of hypertension and diabetes mellitus in MI patients.

Deep learning based hashtag recommendation system for multimedia data

This work aims to provide a novel hybrid architecture to suggest appropriate hashtags to a collection of orpheline tweets. The methodology starts with defining the collection of batches used in the convolutional neural network. This methodology is based on frequent pattern extraction methods. The hashtags of the tweets are then learned using the convolution neural network that was applied to the collection of batches of tweets. In addition, a pruning approach should ensure that the learning process proceeds properly by reducing the number of common patterns. Besides, the evolutionary algorithm is involved to extract the optimal parameters of the deep learning model used in the learning process. This is achieved by using a genetic algorithm that learns the hyper-parameters of the deep architecture. The effectiveness of our methodology has been demonstrated in a series of detailed experiments on a set of Twitter archives. From the results of the experiments, it is clear that the proposed method is superior to the baseline methods in terms of efficiency.

Unsupervised Two-Stage Root-Cause Analysis for Integrated Systems

The increasing complexity and high cost of integrated systems have placed immense pressure on root-cause analysis and diagnosis. In light of artificial intelligence and machine learning, a large amount of intelligent root-cause analysis methods have been proposed. However, most of them need historical test data with root-cause labels from repair history, which are often difficult and expensive to obtain. We propose a two-stage unsupervised root-cause-analysis method in which no repair history is needed. In the first stage, a decision-tree model is trained with system test information to cluster the data in a coarse-grained manner. In the second stage, frequent-pattern mining is applied to extract frequent patterns in each decision-tree node to precisely cluster the data so that each cluster represents only a small number of root causes. The proposed method can accommodate both numerical and categorical test items. A combination of the L-method, cross validation, and Silhouette score enables us to automatically determine all hyperparameters. Two industry case studies with system test data demonstrate that the proposed approach significantly outperforms the state-of-the-art unsupervised root-cause-analysis method.

Prefix-Pruning-Based Distributed Frequent Trajectory Pattern Mining Algorithm

An important problem to be solved in smart city construction is how to improve the efficiency of mining frequent patterns that can be used for location prediction and location-based services of massive trajectory datasets. Owing to uncertain personal trajectory and non-explicit trajectory items, the existing sequence mining algorithms cannot be used directly. To solve this problem, this study proposes a distributed trajectory frequent pattern mining algorithm (SparkTraj) based on prefix pruning. First, a grouping and partitioning technique is used to abstract the original trajectory data and convert them into a common time series.Then, the generation of a redundant trajectory pattern is avoided by using the path adjacency pruning method. Second, to improve mining efficiency, SparkTraj is designed and implemented in Spark, which employs cluster memory computing. Finally, experiments on common datasets show that the proposed algorithm can effectively extract frequent trajectory patterns, and, in particular, deal with the massive amounts of trajectory data. Compared with common trajectory pattern mining algorithms, the SparkTraj algorithm not only improves the overall performance but also has good scalability.

RDF knowledge graph keyword type search using frequent patterns

With the rapid development of Semantic Web, the retrieval of RDF data has become a research hotspot. As the main method of data retrieval, keyword search has attracted much attention because of its simple operation. The existing RDF keyword search methods mainly search directly on RDF graph, which is no longer applicable to RDF knowledge graph. Firstly, we propose to transform RDF knowledge graph data into type graph to prune the search space. Then based on type graph, we extract frequent search patterns and establish a list from frequent search patterns to pattern instances. Finally, we propose a method of the Bloom coding, which can be used to quickly judge whether the information our need is in frequent search patterns. The experiments show that our approach outperforms the state-of-the-art methods on both accuracy and response time.

A Parallel Direct‐Vertical Map Reduce Programming model for an effective frequent pattern mining in a dispersed environment

SummaryData mining methods face various problems when extracting the data analysis process. These problems are solved based on the frequent pattern mining (FPM). In the data mining area, the frequent pattern (FP) extraction is a dominant embassy while processing the databases. Despite the fact that it has proven its importance in various mining provinces, the consistent research work of researchers has provided numerous efforts in this environment through substantial innovations. Based on the expected challenges of pattern mining, the time and memory consumption while mining the hidden FP increases over time by the conventional FPM approaches. This article proposes a Parallel Direct‐Vertical Map Reduce Programming model to effectively extract the FPM in a dispersed environment and analyzed using the Sales transaction dataset as UCI Machine Learning Repository. The proposed model has two phases of working, namely, (i) mapping and (ii) reduction. Here, the proposed Parallel Direct‐Vertical method is utilized while mapping. The parallelism of the proposed model is implemented and executed based on the number of clusters of databases in architecture‐based software and its final outcome shows that the proposed model is more efficient than the existing methods.

A novel algorithm for extracting frequent gradual patterns

The extraction of frequent gradual pattern is an important problem in computer science and largely studied by the scientist’s community of research in data mining. A frequent gradual pattern translates a recurrent co-variation between the attributes of a database. Many applications issues from many domains, such as economy, health, education, market, bio-informatics, astronomy or web mining, are based on the extraction of frequent gradual patterns. Algorithms to extract frequent gradual patterns in the large databases are greedy in CPU time and memory space. This raises the problem of improving the performances of these algorithms. This paper presents a technique for improving the performance of frequent gradual pattern extraction algorithms. The exploitation of this technique leads to a new, more efficient algorithm called SGrite. The experiments carried out confirm the interest of the proposed technique.

Memory-optimized distributed utility mining for big data

In recent days, social media, online services, smartphones, and the Internet of Things (IoT) produces large quantities of data every second. The generated data is structured, unstructured, or semi-structured and available in various formats. Therefore, traditional approaches are not sufficient to handle such kind of data effectively. High utility pattern mining is a famous study area of data analytics that incorporates utility measures to consider user-based constraints such as number of units and benefit, in addition to frequency statistics of datasets. It is also essential to make effective decisions, and their demand is increasing in the last decade. Several techniques have been suggested for utility-based frequent pattern extraction. However, these approaches are limited to data size and operate on standalone systems. We have proposed a parallel approach named distributed memory-optimized utility mining (DMOUM) for high utility-based frequent pattern mining to address this issue. A new pruning technique is implemented that greatly reduces processing time and memory use. The proposed approach can handle huge amounts of data, i.e., big data. DMOUM approach is implemented in cluster-node architecture using Spark. The results are validated on various real-time datasets and found proposed approach has better performance than cutting edge approaches in terms of run-time, memory consumption, and scalability measures. The proposed work will provide solutions to real-time issues such as health care, education, e-commerce, and so on.

Applying Computational Ethnography to Examine Nurses' Workflow Within Electronic Health Records.

The aim of this study was to describe computational ethnography as a contemporary and supplemental methodology in EHR workflow analysis and the relevance of this method to nursing research. We explore the use of audit logs as a computational ethnographic data source and the utility of data mining techniques, including sequential pattern mining (SPM) and Markov chain analysis (MCA), to analyze nurses' workflow within the EHRs. SPM extracts frequent patterns in a given transactional database (e.g., audit logs from the record). MCA is a stochastic process that models a sequence of states and allows for calculating the probability of moving from one state to the next. These methods can help uncover nurses' global navigational patterns (i.e., how nurses navigate within the record) and enable robust workflow analyses. We demonstrate hypothetical examples from SPM and MCA, such as (a) the most frequent sequential pattern of nurses' workflow when navigating the EHR using SPM and (b) transition probability from one record screen to the next using MCA. These examples demonstrate new methods to address the inflexibility of current approaches used to examine nursing EHR workflow. Within a clinical context, the use of computational ethnographic data and data mining techniques can inform the optimization of the EHR. Results from these analyses can be used to supplement the data needed in redesigning the EHR, such as organizing and combining features within a screen or predicting future navigation to improve the record that nurses use.

Efficient Methods for Clickstream Pattern Mining on Incremental Databases

Sequence pattern mining is a core task of data mining in various fields. Clickstream pattern mining, a variant of sequence pattern mining, is very useful in e-commerce, and is used to analyze, evaluate and predict online customer behaviors, attracting the interest of many researchers. Clickstream pattern mining has become important due to its wide range of applications. However, most previous approaches are not suitable to apply when clickstreams are inserted into a database, because this task is too time consuming. The challenge of this problem is how to find a solution to minimize the runtimes and reduce the number of times the original databases are scanned to reduce the computational cost in the mining process on incremental databases. In this paper, we propose two effective methods for mining clickstream patterns from incremental databases, named inCMUB and Eff-inCMUB, based on the pre-large concept. inCMUB inserts new clickstreams from an inserted database into the existing tree and mines all frequent clickstream patterns, while Eff-inCMUB is a new approach, and builds a new tree from the inserted database to find pre-large 1-patterns and then it updates the pre-large clickstream patterns mined from the original database to extract frequent clickstream patterns. The experiments show that our proposed methods outperform the SMUB algorithm in terms of runtimes, memory usage and scalability on real-word clickstream databases.

Corridor Detection from Large GPS Trajectories Datasets

Given the widespread use of mobile devices that track their geographical location, it has become increasingly easy to acquire information related to users’ trips in real time. This availability has triggered several studies based on user’s position, such as the analysis of flows of people in cities, and also new applications, such as route recommendation systems. Given a dataset of geographical trajectories in an urban metropolitan area, we propose a new algorithm to detect corridors. Corridors can be defined as geographical paths, with a minimum length, that are commonly traversed by a minimum number of different users. We propose an efficient strategy based on the Apriori algorithm to extract frequent trajectory patterns from the geo-spatial dataset. By discretizing the data and adapting the roles of itemsets and baskets of this algorithm to our context, we find the longest corridors formed by cells shared by a minimum number of trajectories. After that, we refine the results obtained with a subsequent filtering step, by using a Radius Neighbors Graph. To illustrate the algorithm, the GeoLife dataset is analyzed by following the proposed method. Our approach is relevant for transportation analytics because it is the base to detect lacking lines in public transportation systems and also to recommend to private users which route to take when moving from one part of the city to another on the basis of behavior of the users who provided their logs.

Mining Sequential Patterns with VC-Dimension and Rademacher Complexity

Sequential pattern mining is a fundamental data mining task with application in several domains. We study two variants of this task—the first is the extraction of frequent sequential patterns, whose frequency in a dataset of sequential transactions is higher than a user-provided threshold; the second is the mining of true frequent sequential patterns, which appear with probability above a user-defined threshold in transactions drawn from the generative process underlying the data. We present the first sampling-based algorithm to mine, with high confidence, a rigorous approximation of the frequent sequential patterns from massive datasets. We also present the first algorithms to mine approximations of the true frequent sequential patterns with rigorous guarantees on the quality of the output. Our algorithms are based on novel applications of Vapnik-Chervonenkis dimension and Rademacher complexity, advanced tools from statistical learning theory, to sequential pattern mining. Our extensive experimental evaluation shows that our algorithms provide high-quality approximations for both problems we consider.