High Utility Patterns Research Articles

SOHUPDS+: An Efficient One-phase Algorithm for Mining High Utility Patterns over a Data Stream

Existing algorithms for mining high utility patterns over a data stream are two-phase algorithms that are not scalable due to the large number of candidates generation in the first phase, particularly when the minimum utility threshold is low. Moreover, in the second phase, the algorithm needs to scan the database again to find out actual utility for candidates. In this paper, we propose one-phase algorithm SOHUPDS+ to mine high utility itemsets in the current sliding window of the data stream with respect to absolute or relative minimum utility threshold. To facilitate SOHUPDS+, we propose a data structure IUDataListSW+ , which stores and maintains utility and upper-bound values of the items in the current sliding window when sliding window advances. In addition, we propose a transaction merging strategy, called BitmapTransactionMerging , which saves execution time for utility and upper-bound values computations in denser datasets. Moreover, we propose update strategies to utilize mined high utility patterns from the previous sliding window to update high utility patterns in the current sliding window. The results of experiments illustrate that SOHUPDS+ is more efficient than the state-of-the-art algorithms in terms of execution time as well as memory usage in most of the experiments on various datasets.

Modified Genetic Algorithm for Efficient High-Utility Itemset Mining

In pattern mining, high-utility itemset mining (HUIM) is useful for discovering high-utility patterns. The study of HUIM using heuristic techniques reflects issues in producing better offspring. It is ineffective in terms of search space organization, population diversity, and utility calculation, which impact runtime and accuracy. It is observed that very few researchers have experimented with genetic algorithm (GA) and are still facing the same issues as mentioned before. To overcome these problems, a novel approach is proposed for HUIM using modified GA and optimized local search (HUIM-MGALS) with six potential contributions. First is linking the utility with the Bitmap dataset to reduce utility access time, leading to effective search space organization. Second, HUIM-MGALS employs a fitness scaling strategy to avoid redundancy. Third, a high-utility itemset (HUI) revision strategy is employed to explore significant HUIs. Modified population diversity maintenance strategy and iterative crossover help to preserve significant HUIs and improve search capability as fourth and fifth contributions. Sixth, the use of multiple mutations refines the wasted individuals to boost accuracy. Extensive experimentation showed that HUIM-MGALS significantly outperforms the presented algorithms, up to 8.6 times faster. It also demonstrates superior HUI discovery capabilities for both sparse and dense datasets. This is supported by the modified population diversity maintenance strategy, which is proved to be the most impactful modification for HUI discovery in HUIM-MGALS.

Method for Mining High-utility Patterns in Transaction Stream Data based on Linked List Structure

Mining valuable patterns in data streams presentsa significant challenge in the field of data mining. Thistask is crucial as it allows for the identification of highlyprofitable item sets within transaction databases. However, asnew transactions are continually added, new valuable patternsemerge, thus changing the usefulness of previously analyzeddata. It is essential to promptly update information regardingthese changes to enable effective business decision-making.Consequently, existing mining methods applied to transactionflow datasets require considerable time to identify newpatterns and update information related to new transactions.This article focuses on the research and proposal of a newtransaction stream data mining method called High-UtilityStream Linked-List Mining. The method utilizes a linkedlist structure known as the High-Utility Stream Linked List(HUSLL) to store information about patterns in the database.Mining and updating transaction information are directlyperformed on the HUSLL structure. Experimental resultsdemonstrate that this novel mining method exhibits moreefficient execution times compared to previous solutions.

Effective approaches for mining correlated and low-average-cost patterns

High-utility pattern mining is a data-mining research field that focuses on identifying patterns that exceed a specified threshold for utility values. These utility values primarily indicate positive benefits such as profits. However, in actual scenarios, certain applications and demands require the discovery of patterns at a lower cost. Hence, the concept of low-average-cost itemset mining is introduced to evaluate patterns based on their average costs. This is different from high-utility pattern mining, as the cost is minimized. In this study, we focus on extracting patterns with strong correlations and low average costs. Because existing techniques depend on the support metric to devise cost-based pruning strategies, these strategies are ineffective in extracting correlated and low-average-cost patterns. Therefore, we introduce a novel lower bound to efficiently limit the search space for patterns. Experimental analyses are conducted on eight databases to assess the performance of the mining algorithm in terms of execution time, memory usage, pruning effectiveness, and scalability. Additionally, the effects of different pruning techniques and item-sorting orders on the performance of the algorithm are investigated.

Stable top-k periodic high-utility patterns mining over multi-sequence

Stable top-k periodic high-utility patterns mining over multi-sequence

MLC-miner: Efficiently discovering multi-level closed high utility patterns from quantitative hierarchical transaction databases

High utility pattern mining (HUPM) extends frequent pattern mining (FPM) by including item significance and quantity, which determine their utility, in transaction databases. A pattern is a high utility if its utility satisfies the minimum threshold. Several efficient approaches have been developed to address this task. However, more extensive databases combined with lower thresholds can lead to an explosion in the number of discovered patterns. Use the closed version of the pattern to decrease patterns while keeping information. To improve real-world applicability, HUPM research now considers item connections. Categorizing items creates an abstraction model that can help decision-makers. This new mining task is generalized high utility pattern mining (GHUPM). However, GHUPM significantly increases the complexity of the mining task, resulting in longer mining times, higher memory usage, and a more significant number of discovered patterns. To date, none of the proposed approaches in GHUPM have utilized the closed representation to shrink the size of the result set. This study introduces MLC-Miner, a multi-level closed high utility pattern extraction technique, to fill this research gap. MLC-Miner combines several efficient methods to enhance mining performance and reduce memory footprints. Experimental results show that MLC-Miner can handle up to 5 million transactions efficiently.

Constraint-based high utility mobile trajectory pattern mining for security systems

In security system, high utility pattern mining of a large number of users mobile trajectories is helpful to analyze user behavior patterns, and enhance the internal prevention of the security system.Currently, the frequent pattern mining for mobile trajectory in security systems do not take into account the differences in the safety levels between staying points and the frequency of their occurrence. Existing high utility mobile trajectory pattern mining methods are unable to discover staying points that meet specific user requirements. To solve these problems, a constraint-based high utility mobile trajectory mining algorithm is proposed (names as HUIM-ILC-ACO). It takes into account the user’s stay time, stay frequency, and the safety level of each stopping point and calculates the utility value of each staying point by incorporating these factors through weighted computation. Based on this, the algorithm combines ant colony optimization with length constraints and item constraints to construct a method for mining high utility mobile trajectory patterns that better align with user interests. Experimental results on real datasets and a target mobile trajectory RFID dataset show that proposed algorithm is efficient in terms of runtime and pattern quantity, and it can effectively mine a pattern set that is closer to user interests.

Improved adaptive-phase fuzzy high utility pattern mining algorithm based on tree-list structure for intelligent decision systems

With the rapid development of AI and big data mining technologies, computerized medical decision-making has become increasingly prominent. The aim of high-utility pattern mining (HUPM) is to discover meaningful patterns in medical databases that contribute to maximizing the utility from the perspective of diagnosis. However, HUPM pays less attention to the interpretability and explainability of these patterns in medical decision-making scenarios. This paper proposes a novel algorithm called the Improved fuzzy high-utility pattern mining (IF-HUPM) to address this problem. First, the paper applies a fuzzy preprocessing method to divide the fuzzy intervals of a medical quantitative data set, which enhances the fuzziness and interpretability of the data. Next, in the process of IF-HUPM, both fuzzy tree and list structures are employed to calculate fuzzy high-utility values. By combining the characteristics of the one-stage and two-stage algorithms of HUPM, an adaptive-phase Fuzzy HUPM hybrid frame is proposed. The experimental results demonstrate that the proposed IF-HUPM algorithm enhances both accuracy and efficiency and the mining process requires less time and space on average.

Incremental Top-k High Utility Pattern Mining and Analyzing Over the Entire Accumulated Dynamic Database

Incremental Top-k High Utility Pattern Mining and Analyzing Over the Entire Accumulated Dynamic Database

A Survey on Association Rule Mining for Enterprise Architecture Model Discovery

AbstractAssociation Rule Mining (ARM) is a field of data mining (DM) that attempts to identify correlations among database items. It has been applied in various domains to discover patterns, provide insight into different topics, and build understandable, descriptive, and predictive models. On the one hand, Enterprise Architecture (EA) is a coherent set of principles, methods, and models suitable for designing organizational structures. It uses viewpoints derived from EA models to express different concerns about a company and its IT landscape, such as organizational hierarchies, processes, services, applications, and data. EA mining is the use of DM techniques to obtain EA models. This paper presents a literature review to identify the newest and most cited ARM algorithms and techniques suitable for EA mining that focus on automating the creation of EA models from existent data in application systems and services. It systematically identifies and maps fourteen candidate algorithms into four categories useful for EA mining: (i) General Frequent Pattern Mining, (ii) High Utility Pattern Mining, (iii) Parallel Pattern Mining, and (iv) Distribute Pattern Mining. Based on that, it discusses some possibilities and presents an exemplification with a prototype hypothesizing an ARM application for EA mining.

Scalable and Efficient Approach for High Temporal Fuzzy Utility Pattern Mining.

Fuzzy utility (FU) pattern mining with an advantage in human reasoning has become one of the interesting topics in studies of knowledge discovery. The discovered information in FU pattern mining from real-life quantitative databases with item profits is suitable for interpreting data from a human perspective because it is not expressed using numerical values but linguistic terms which consist of natural languages. State-of-the-art approaches in this literature provide extended results by considering temporal factors, such as seasons, which can be influential in real-life situations. However, they still suffer from scalability issues because they are based on level-wise approaches which generate a number of candidates. In this article, we propose a scalable and efficient approach with a novel data structure for mining high temporal FU patterns without generating candidates. Efficient pruning techniques and algorithms are presented to improve the performance of the proposed approach. Performance experiments on both real and synthetic datasets show that the suggested algorithm has better performance than the state-of-the-art algorithms in terms of runtime, memory usage, and scalability.

An Efficient Approach for Mining Closed High Utility Patterns in the Incremental Database

An Efficient Approach for Mining Closed High Utility Patterns in the Incremental Database

Efficient approach of high average utility pattern mining with indexed list-based structure in dynamic environments

Various studies on high utility pattern mining have been conducted to satisfy the emerging need to consider the characteristics of real-world databases, such as the importance and quantity of items. In the traditional utility-based framework, the mining result is influenced by the number of items in a pattern, or in some cases, single utilities of items. In order to overcome this drawback, high average utility pattern mining has been proposed. It provides more interesting results since it takes into account the average utility of patterns by considering their lengths. Methods based on this concept have emerged in recent years, including ones that target incremental environments. However, existing algorithms create an enormous number of candidate patterns or require complex operations during the mining process. To address this degradation, we propose a new and more efficient approach for mining high average utility patterns in dynamic environments. The proposed algorithm utilizes a data structure more efficient than previous ones, which takes the form of an indexed list. It also incorporates efficient realigning and mining techniques for handling incremental data and accurately mining results. Experimental results show the superiority of the proposed approach in terms of runtime, memory usage, scalability, and accuracy.

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.

Exercise gas exchange and echo-derived analysis in HFpEF based on patterns of oxygen peripheral utilization

Abstract Background Exercise limitation is a key signature of heart failure with preserved ejection fraction (HFpEF) syndrome evolution with studies highlighting a significant cardiac reserve limitation and others identifying a predominant limit in the peripheral VO2 kinetics and utilization. Aims To clarify the cardiopulmonary and hemodynamic response in HFpEF according to the high or low peripheral O2 utilization patterns during maximal exercise. Materials and Method 38 patients with a HFpEF diagnosis based on the ESC HFpEFF criteria underwent cardiopulmonary exercise test (CPET) combined with stress echocardiography. Patients were divided in two groups according to the median peak c(A-V) O2 difference (Group 1: < 11 mL/dL and Group 2: > 11 mL/dL). Cardiac output (CO) was calculated by echocardiography according to the formula LVOT velocity integral x LVOT area. Data were obtained after a familiarization CPET session. Results Group 1 and 2 presented similar rest demographic, left and right hemodynamic data (table). The relationship between CO and c (A-V) O2 in terms of VO2 isopleths at peak exercise were not related to any difference in CO (figure) Conclusions In HFpEF, an impairment in O2 utilization during exercise appears independent of the cardiac reserve response (similar peak CO increase) but rather peripheral mechanisms implicated in O2 diffusion/release seem to play a dominant role, emerging as selective therapeutic targets for improving functional capacity and symptoms.

Preface

It is our great pleasure to have you at the 2023 5th International Conference on Artificial Intelligence Technologies and Applications (ICAITA 2023), which took place in Changchun, China from June 30th to July 2nd, 2023 (hybrid conference). ICAITA 2023 was successfully held, and received unanimous praise from various experts and scholars all over the world.ICAITA 2023 existed as an international forum for academic communications between experts and scholars in the fields of artificial intelligence technologies and applications. It promoted the research and developmental activities in related fields and scientific information interchange among all the participants, establishing research connections for them to find global partners for potential collaboration.After months of well preparation and hard work, the Proceedings of ICAITA 2023 covering a bunch of papers, having been checked through rigorous review and processes to meet the requirements of publication, are smoothly published. These papers feature but are not limited to the following topics: Artificial Neural Networks, Computational Theories of Learning, Pervasive Computing and Ambient Intelligence, Fuzzy Logic and Methods, etc.We had about 300 participations both as speakers and participants, including senior experts, research talents, and young scholars from enterprises, research institutes, and universities worldwide. During the keynote speech part, each keynote speaker was given 30–40 minutes for keynote speech. Among them, Professor Philippe Fournier-Viger from Shenzhen University, China performed a keynote speech on Advances and Challenges for the Automatic Discovery of Interesting Patterns in Data. He first briefly reviewed early studies on designing algorithms for identifying frequent patterns. Then, he presented an overview of recent challenges and advances to identify other types of interesting patterns in more complex data. Topics that were discussed included high utility patterns, locally interesting patterns, and periodic patterns. Lastly, the SPMF open-source software was mentioned and opportunities related to the combination of pattern mining algorithms with traditional artificial intelligence techniques for intelligent systems were discussed. The insightful speeches of each keynote speakers triggered interesting discussions and a good number of informal talks between all the participants.We would like to express our gratitude to the organizers, keynote speakers and all the participants of ICAITA 2023, for their fruitful work and contribution to the success of the Conference. We are thankful to all the reviewers for providing constructive criticisms, stimulated comments and suggestions to the authors. Particularly, our special thanks go to Journal of Physics: Conference Series, for the endeavor of all its colleagues in publishing this paper volume.May this event be a spark that ignites many more new inventions in the fields related to artificial intelligence technologies and applications!The Committee of ICAITA 2023List of Committee Member is available in this pdf.

HEPM: High-efficiency pattern mining

Pattern mining (PM) is an important field of data mining and has gained considerable momentum recently, mainly owing to the massive growth of big data. PM often sets attentive objectives such as mining frequent or high utility patterns to obtain attractive patterns. High utility patterns address the defect of frequent patterns that cannot reveal the maximum profit. However, it neglects another vital factor, cost or investment. This paper proposes a new high-efficiency PM problem that considers both utility and investment. The problem aims to find patterns with the maximum profit-to-investment ratio. Our paper is devoted to studying high-efficiency itemsets in transaction databases. We first formulate the criteria for a high-efficiency PM problem. Subsequently, we propose a two-phase algorithm called HEPM and an improved one-phase algorithm called HEPMiner to discover high-efficiency patterns in a transaction database. We design a corresponding pruning strategy within HEPM to reduce the search space. In HEPMiner, we utilize a novel efficiency-list and an estimated efficiency co-occurrence structure in the pruning strategies to further improve the mining performance. Moreover, we derive the upper bounds of efficiency for both algorithms. The experimental results demonstrate the effectiveness and efficiency of our two algorithms.

Efficient approach for mining high-utility patterns on incremental databases with dynamic profits

High-utility itemset mining (HUIM) is one of the heavily studied fields of data mining, which is due to its high compatibility with real-world applications. HUIM is a process of extracting a complete set of interesting patterns by considering the importance and the quantity of each item. Incremental high-utility itemset mining (IHUIM) further increases the compatibility of HUIM by mining from a data stream instead of a static database. It does this by mining interesting patterns in a single database scan and storing prior knowledge in a compact data structure. However, there is a critical limitation, because the importance of each item has to be guaranteed to remain unchanged throughout the data stream. This limitation hinders the compatibility of IHUIM to the real-world applications, because the importance of an item changes from time to time in the real world. Conventional IHUIM approaches consequently have to use outdated information when mining interesting patterns. This paper proposes a novel problem of mining high-utility itemsets in an incremental and dynamic profit environment in order to account for the fluctuation of the item's importance. Furthermore, a novel approach to mining patterns in this type of environment is introduced using an efficient list structure and tight upper bounds. Experiments on real and synthetic datasets show that the proposed approach performs well compared to the state-of-the-art approaches in terms of the runtime and memory usage, and it scales better than the other approaches.

Mining skyline frequent-utility patterns from big data environment based on MapReduce framework

When the concentration focuses on data mining, frequent itemset mining (FIM) and high-utility itemset mining (HUIM) are commonly addressed and researched. Many related algorithms are proposed to reveal the general relationship between utility, frequency, and items in transaction databases. Although these algorithms can mine FIMs or HUIMs quickly, these algorithms merely take into account frequency or utility as a unilateral criterion for itemsets but the other factors (e.g., distance, price) could be also valuable for decision-making. A new skyline framework has been presented to mine frequent high utility patterns (SFUPs) to better support user decision-making. Several new algorithms have been proposed one after another. However, the Internet of Things (IoT), mobile Internet, and traditional Internet are generating massive amounts of data every day, and these cutting-edge standalone algorithms can not satisfy the new challenge of finding interesting patterns from this data. Big Data uses a distributed architecture in the form of cloud computing to filter and process this data to extract useful information. This paper proposes a novel parallel algorithm on Hadoop as a three-stage iterative algorithm based on MapReduce. MapReduce is used to divide the mining tasks of the whole large data set into multiple independent sub-tasks to find frequent and high utility patterns in parallel. Numerous experiments were done in this paper, and from the results, the algorithm can handle large datasets and show good performance on Hadoop clusters.

High utility pattern mining algorithm over data streams using ext-list.

High utility pattern mining algorithm over data streams using ext-list.