High temporal fuzzy utility pattern analysis extracts patterns by considering temporal factors at which items appear, as well as expressing patterns in a linguistic way that humans can easily understand based on the fuzzy theory. However, in real world stream data environments, data often involve inherent uncertainty, such as inaccuracies in sensor data. Nevertheless, previous studies utilizing the temporal aspect for pattern analysis and fuzzy set theory do not explicitly account for such uncertainty or inaccuracies. Motivated by the limitation, we introduce a scalable method to extract high temporal fuzzy utility patterns within dynamic uncertain data streams. By incorporating uncertainty associated with items appearing in each transaction, our method extracts uncertainty-based high temporal fuzzy utility patterns that more accurately reflect real world conditions. The pattern expansion process is conducted efficiently with a list structure and various pruning methods. Experimental results indicate the proposed approach exhibits superior performance on runtime and memory usage compared to previous methods under diverse membership functions. Moreover, our method exhibits the highest scalability for increasing data volumes and maintains stable sensitivity to threshold variation while exhibiting the completeness and significance of extracted results. Evaluations under various uncertainty distributions show the superior performance of our approach, and effectiveness evaluations indicate contributions of proposed components. The case study involving the concept drifting evaluation indicates the applicability of our approach in various real-world environments with better performance. The implementation is available at https://github.com/sejongdmlab/ TFUN.

On the Internet of Things (IoT) environment, interconnected devices continuously share generated data in real time, typically collected with a specific purpose. To ensure efficiency, IoT technology must deliver only the most relevant insights into these devices. High utility pattern mining is a technique that extracts important knowledge, and there has been research on performance improvements to efficiently mine these patterns in dynamic environments. Although approaches with a prelarge concept have been introduced to mining high utility patterns in data deletion environments, the state-of-the-art method relies on inefficient data structures, making them unsuitable for real-time analysis with IoT data. To overcome these limitations, this article proposes a novel utility pattern mining approach with prelarge concept for dynamic IoT environments, where data is deleted because of sensor errors or storage constraints. The actual utilities of the prelarge patterns are maintained to skip the verification and improve communication delays. It can optimize processing time and memory consumption due to storing fewer large or prelarge patterns based on the actual utility. The proposed method operates in an efficient list-based manner, enabling effective search space pruning when the rescan condition is met and generating compact data structures through transaction merging. The experiments indicate that our algorithm is outstanding regarding processing time and scalability with minimal compromise in memory consumption compared with the existing methods, while extracting the exact patterns. Additionally, an analysis that replicates real IoT environments demonstrates that the proposed method is sufficiently applicable in real-world settings.

Utility-driven pattern analysis is a fundamental method for analyzing noteworthy patterns with high utility for diverse quantitative transactional databases. Recently, various approaches have emerged to handle large, dynamic database environments more efficiently by reducing the number of data scans and pattern expansion operations with the pre-large concept. However, existing pre-large-based high utility pattern mining methods either fail to handle real-time transaction modifications or require additional data scans to validate candidate patterns. In this paper, we propose a novel efficient utility-driven pattern mining algorithm using the pre-large concept for transaction modifications. Our method incorporates a single-scan-based framework through the management of actual utility values and discovers high utility patterns without candidate generation for efficient utility-driven dynamic data analysis in the modification environment. We compared the performance of the proposed method with state-of-the-art methods through extensive performance evaluation utilizing real and synthetic datasets. According to the evaluation results and a case study, the suggested method performs a minimum of 1.5 times faster than state-of-the-art methods alongside minimal compromise in memory, and it scaled well with increases in database size. Further statistical analyses indicate that the proposed method reduces the pattern search space compared to the previous method while delivering a complete set of accurate results without loss.

Pattern mining, one of the data analysis approaches, provides meaningful assistance for various business services, such as product recommendation and marketing. However, certain real-world data contain uncertain characteristics, and some business services want to consider the uncertainty of data. Uncertain pattern mining is an advanced technique for discovering more useful patterns from uncertainty-driven data with uncertain information about items. However, although many business services create and process incremental data in real-time, most of the previous uncertain pattern mining techniques have limitations in analyzing incremental data since they mainly focus on processing static data. To address the limitations, we present a list-based uncertain pattern mining method that effectively analyzes incremental uncertainty-driven data in real time by scanning stream data only once. In addition, uncertainty-driven data analytics can be executed efficiently due to the list structure that is effective in construction and mining. The tests of performance for runtime, memory consumption, and scalability are performed using real datasets and synthetic datasets, which illustrate that the suggested technique reveals outstanding performance compared to state-of-the-art algorithms. The additional case study evaluations with concept-drifting tests as well as accuracy and significance tests demonstrate the practical applications of the algorithm and the quality of the extracted results.

In a manufacturing factory, product lines are organized by several constituents and exhibit a profit value, i.e., income from products. Erasable patterns are less profitable patterns whose gain, i.e., the sum of product profits, does not exceed a user-defined threshold. Mining erasable patterns provides the necessary information to users who want to increase profits by erasing less profitable patterns. There are requirements for a method which efficiently manages uncertain databases in incremental environments to identify erasable patterns that consider uncertainty. Because our novel technique uses a list structure, it is more efficient at finding erasable patterns from incremental databases. Moreover, accumulated stream data should be handled efficiently to identify new useful patterns in both additional data and the existing data. In this article, an algorithm using a list-based structure is proposed to extract erasable patterns containing valuable knowledge from uncertain databases in real time with effective and productive performance. In order to derive erasable patterns from continuously accumulated stream databases, the structure efficiently manages the information gathered from the previous database. Extensive performance and pattern quality evaluations were conducted using real and synthetic datasets. The results show that the algorithm performs up to seven times faster than state-of-the-art erasable pattern mining algorithms on real datasets and scales adeptly on synthetic datasets while delivering reliable and significant result patterns.

Many researchers have been investigating and applying a new trend of data mining , namely high occupancy itemset mining. Frequent itemset mining often returns a large set of itemsets, but businesses need a smaller set of inputs to investigate or send into a recommendation system to quickly make decisions. Applying an occupancy measure to a support-based mining framework will thus bring many benefits for decision support systems , while managers will benefit by having a new method to visualize reports and analyze data more efficiently. Similar to frequent itemset mining, mining high occupancy itemsets can be applied on any transaction database. In this research, we apply additional conditions to eliminate unqualified itemsets and integrate the property of equivalence class to reduce the runtime of the k-itemsets generation process. Moreover, a new theorem is stated and applied to a specific class of databases so that it is not necessary to calculate the upper-bound occupancy, and this speeds up the process as well as reduces memory requirements with regard to generating high occupancy itemsets. We develop two new algorithms, fast high occupancy itemset mining (FHOI) and depth first search (DFS) for high occupancy itemset mining (DFHOI) to solve the problem. Our new algorithms are examined experimentally using different databases to evaluate its performance in term of runtime and memory usage.

Recent studies on heterogeneous information network (HIN) embedding-based recommendations have encountered challenges. These challenges are related to the data heterogeneity of the associated unstructured attribute or content (e.g., text-based summary/description) of users and items in the context of HIN. In order to address these challenges, in this article, we propose a novel approach of semantic-aware HIN embedding-based recommendation, called SemHE4Rec. In our proposed SemHE4Rec model, we define two embedding techniques for efficiently learning the representations of both users and items in the context of HIN. These rich-structural user and item representations are then used to facilitate the matrix factorization (MF) process. The first embedding technique is a traditional co-occurrence representation learning (CoRL) approach which aims to learn the co-occurrence of structural features of users and items. These structural features are represented for their interconnections in terms of meta-paths. In order to do that, we adopt the well-known meta-path-based random walk strategy and heterogeneous Skip-gram architecture. The second embedding approach is a semantic-aware representation learning (SRL) method. The SRL embedding technique is designed to focus on capturing the unstructured semantic relations between users and item content for the recommendation task. Finally, all the learned representations of users and items are then jointly combined and optimized while integrating with the extended MF for the recommendation task. Extensive experiments on real-world datasets demonstrate the effectiveness of the proposed SemHE4Rec in comparison with the recent state-of-the-art HIN embedding-based recommendation techniques, and reveal that the joint text-based and co-occurrence-based representation learning can help to improve the recommendation performance.

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.

Renewable energy is critical to the energy industry, even more so given the recent surge in oil and gas prices. As a result, the oil and gas sector is constantly preparing for the coming energy transition. This raises the question of when the oil and gas industry should invest in renewable energy sources. This article examines the history of oil company involvement in renewable energy to address this question. It also establishes and quantifies a relationship between renewable energy investment and oil and gas prices, followed by a projection of renewable energy investment based on oil and gas prices. A decision support system based on crude oil prices and renewable energy stocks is developed to predict renewable energy investment decisions. According to the experiments, the prediction model used provides favorable mean square values compared to other models. An algorithm based on the price of oil determines whether to invest in shares of renewable energy companies. The performance of a company is evaluated based on its expected profitability.

Software-defined vehicular networks (SDVN) can help analyze and reconfigure networks. Massive data generation in autonomous vehicles can lead to issues in network configuration, routing, network characteristics, and system load factors. Load balancing in vehicle sensors helps reduce delays and improve resource utilization. In this paper, we propose a load balancing algorithm to map sensor data, vehicles and data centers performing tasks. A dynamic convergence method is proposed to help identify vehicle system load factors and compare their termination criteria. We also propose a packet-level intrusion detection model. After all load balancing, the model can track the attack on the network. The proposed model further combines the entropy-based active learning and the attention-based model to efficiently identify the attacks. Experiments are then conducted on the standard KDD data to validate the developed models with and without an attention-based active learning mechanism. Our experimental results show that the load balancing mechanism is able to achieve more performance gains than previous techniques. Moreover, the results show that the developed model can improve the decision boundary by using a pooling strategy and an entropy uncertainty measure.

Abstract Recently, revealing more valuable information except for quantity value for a database is an essential research field. High utility itemset mining (HAUIM) was suggested to reveal useful patterns by average-utility measure for pattern analytics and evaluations. HAUIM provides a more fair assessment than generic high utility itemset mining and ignores the influence of the length of itemsets. There are several high-performance HAUIM algorithms proposed to gain knowledge from a disorganized database. However, most existing works do not concern the uncertainty factor, which is one of the characteristics of data gathered from IoT equipment. In this work, an efficient algorithm for HAUIM to handle the uncertainty databases in IoTs is presented. Two upper-bound values are estimated to early diminish the search space for discovering meaningful patterns that greatly solve the limitations of pattern mining in IoTs. Experimental results showed several evaluations of the proposed approach compared to the existing algorithms, and the results are acceptable to state that the designed approach efficiently reveals high average utility itemsets from an uncertain situation.

Internet-delivered Psychological Treatment (IDPT) has been shown to be an effective method for improving psychological disorders. Natural language processing (NLP) requires an appropriate set of linguistic features for word representation and emotion segmentation. For psychological applications, models must be trained on extensive and diverse datasets to achieve expert-level performance. Labeling psychological texts authorized by patients is challenging because emotional biases can lead to incorrect segmentation of emotions and labeling emotional data is time consuming. In this paper, we propose an assistance tool for psychologists to explore the emotional aspects of mentally ill individuals. We first use an NLP-based method to create emotional lexicon embeddings, and then apply attention-based deep clustering. The learned representation is then used to visualize the emotional aspect of the text authorized by patients. We expand the patient authored text using synonymous semantic expansion. A latent semantic representation based on context is clustered using EANDC, which is a Explainable Attention Network-based Deep adaptive Clustering model. We use similarity metrics to select a subset of the text and then improve the explainability of learning using a curriculum-based optimization method. The experimental results show that synonym expansion based on the emotion lexicon increases accuracy without affecting the results. The attention method with bidirectional LSTM architecture achieved 0.81 ROC in a blind test. The self-learning based embedding then visualizes the weighted attention words and helps the psychiatrist to improve his explanatory power of the qualitative match for clinical notes and the remedy. The method helps in labeling text and improves the recognition rate of symptoms of mental disorders.

Utility pattern mining is a branch of data mining that extracts valid patterns by considering the quantity and weight of the items. In addition, utility occupancy pattern mining, which considers the quantity, importance, and proportion of the pattern in the transaction, has been proposed. Despite this advantage, there is no utility seizing approach to handle the dynamically generated data flows. As electronics are interconnected and intelligent systems are constructed, data is generated in real-time and accumulated rapidly. Therefore, a method to read data immediately in a dynamic environment and efficiently analyze massive data is required. To overcome the limitations of the existing utility occupancy methods, we propose a novel mining approach, HUOMI, which performs quickly on an increasing database. The suggested algorithm has an optimized data structure and an improved pruning technique, which can respond to the dynamic environment promptly. To indicate the effectiveness of the proposed method, performance evaluations were conducted on real and synthetic data sets. In the experimental results, the suggested algorithm showed a better performance than the other state-of-the-art algorithms.

Recent advances in deep learning have tremendously leveraged the performance of network representation learning (NRL). Multiple deep learning-based NRL models have been proposed recently to effectively handling primitive tasks of information network analysis and mining (INAM) domain, including link prediction (LP). LP is considered as an important one due to its multiple applications in many disciplines. In the recent few years, LP in dynamic networks has attracted a lot of attention from researchers to propose novel algorithms for better capturing both rich structural and evolutional information of complex information networks (INs). However, recent models are mainly concentrated on preserving the sequential representations of a given network over time. They have largely ignored other important structural features, such as: intracommunity which contributes to the creation of links between network nodes. In this article, we propose a novel community-driven dynamic NRL technique upon the RNN+GCN framework, called: ComGCN. Specifically, the ComGCN model is a combination of microscopic (node embedding-based) and mesoscopic (intracommunity-based) dynamic network embedding approach which enable effectively handling the LP problem in context of dynamism. Extensive experiments on real-world dynamic networks demonstrated the effectiveness of the proposed model compared with recent state-of-the-art baselines.

Closed sequential pattern (CSP) mining is an optimization technique in sequential pattern mining because they produce more compact representations. Additionally, the runtime and memory usage required for mining CSPs is much lower than the sequential pattern mining. This task has fascinated numerous researchers. In this study, we propose a novel approach for closed clickstream pattern mining using C-List (CCPC) data structure. Closed clickstream pattern mining is a more specific task of CSP mining that has been lacking in research investment; nevertheless, it has promising applications in various fields. CCPC consists of two key steps: It initially builds the SPPC-tree and the C-List for each frequent 1-pattern and then determines all frequently closed clickstream 1-patterns; next, it constructs the C-List for each frequent k-pattern and mines the remaining frequently closed k-patterns. The proposed method is optimized by modifying the SPPC-tree structure and a new property is added into each node element in both the SPPC-tree and C-Lists to quickly prune nonclosed clickstream. Experimental results conducted on several datasets show that the proposed method is better than the previous techniques and improves the runtime and memory usage in most cases, especially when using low minimum support thresholds on the huge databases.

In industrial areas, most of databases are dynamic databases, and the volume of the databases has grown with the passage of time. Especially, pattern mining for incremental database needs different approaches from static database because the profit or the accuracy of the previously inserted data can be reduced. Since data is time- sensitive, the recent data has a relatively higher value than the old data. In this paper, we suggest the damped window based average utility driven data analytics for intelligent systems, which the damped window reflects the importance according to the arrival time of the transactions. The proposed mining approach adopts novel data structure, which modify the importance of item as the passage of time, and it improves mining efficiency with several pruning strategies and without generating candidate patterns. To evaluate the performance of the proposed mining approach, we conducted various experiments using several real and synthetic data sets. The result of the experiments presented that the suggested method performs better in terms of runtime and memory usage than the other state-of-the-art mining techniques. Moreover, through the scalability experiments, which changed the number of different items or transactions, we verified that the proposed algorithm maintained a stable performance under various environmental changes.

The Internet of Things (IoT) environment includes things that exchange information using data generated through sensors. The IoT technology can be executed efficiently by providing the necessary information to things instead of transmitting all the data. There is a need for an algorithm that can extract meaningful information for the things from the data. High utility pattern mining, which can handle the characteristics of real-world databases better than traditional pattern mining methods, has been actively researched. The traditional high utility pattern mining techniques find meaningful patterns from static databases. Therefore, these techniques are not suitable for the dynamically changing databases in the real world. In order to solve this problem, a variety of methods that consider the modifications, deletions, and insertions of transactions are proposed. The prelarge concept, which can be efficiently operated by reducing the rescanning of the original database using two thresholds, was also proposed in order to handle the limitations of the methods that are used for static databases. In the prelarge concept, the large patterns and the prelarge patterns that are discovered are maintained and used for the next transaction change. In this article, we proposed a new method pre-large based incremental high utility pattern mining for transaction modification (PIHUP-MOD) implemented as a tree structure in order to handle the transaction modifications. The method has an efficient structure and techniques to mine high utility patterns in the transaction modifications with the prelarge concept. We conducted the performance evaluation, and the experimental results on the real datasets and the synthetic datasets showed that the proposed approach has a better performance than the state-of-the-art approaches in terms of the runtime, memory usage, and scalability.

Over the past decade, high-utility itemset mining (HUIM) has received widespread attention that can emphasize more critical information than was previously possible using frequent itemset mining (FIM). Unfortunately, HUIM is very similar to FIM since the methodology determines itemsets using a binary model based on a pre-defined minimum utility threshold. Additionally, most previous works only focused on single, small datasets in HUIM, which is not realistic to any real-world scenarios today containing big data environments. In this work, the fuzzy-set theory and a MapReduce framework are both utilized to design a novel high fuzzy utility pattern mining algorithm to resolve the above issues. Fuzzy-set theory is first involved and a new algorithm called efficient high fuzzy utility itemset mining (EFUPM) is designed to discover high fuzzy utility patterns from a single machine. Two upper-bounds are then estimated to allow early pruning of unpromising candidates in the search space. To handle the large-scale of big datasets, a Hadoop-based high fuzzy utility pattern mining (HFUPM) algorithm is then developed to discover high fuzzy utility patterns based on the Hadoop framework. Experimental results clearly show that the proposed algorithms perform strongly to mine the required high fuzzy utility patterns whether in a single machine or a large-scale environment compared to the current state-of-the-art approaches.

Databases that deal with the real world have various characteristics. New data is continuously inserted over time without limiting the length of the database, and a variety of information about the items constituting the database is contained. Recently generated data has a greater influence than the previously generated data. These are called the time-sensitive non-binary stream databases, and they include databases such as web-server click data, market sales data, data from sensor networks, and network traffic measurement. Many high utility pattern mining and stream pattern mining methods have been proposed so far. However, they have a limitation that they are not suitable to analyze these databases, because they find valid patterns by analyzing a database with only some of the features described above. Therefore, knowledge-based software about how to find meaningful information efficiently by analyzing databases with these characteristics is required. In this article, we propose an intelligent information system that calculates the influence of the insertion time of each batch in a large-scale stream database by applying the sliding window model and mines recent high utility patterns without generating candidate patterns. In addition, a novel list-based data structure is suggested for a fast and efficient management of the time-sensitive stream databases. Moreover, our technique is compared with state-of-the-art algorithms through various experiments using real datasets and synthetic datasets. The experimental results show that our approach outperforms the previously proposed methods in terms of runtime, memory usage, and scalability.

Data mining is a method for extracting useful information that is necessary for a system from a database. As the types of data processed by the system are diversified, the transformed pattern mining techniques for processing these type of data have been proposed. Unlike the traditional pattern mining methods, erasable pattern mining is a technique for finding the patterns that can be removed by coming with a small profit. Erasable pattern mining should be able to process data by considering both the environment that the data are generated from and the characteristics of the data. An uncertain database is a database that is composed of uncertain data. Since erasable patterns discovered from uncertain data contain significant information, these patterns need to be extracted. In addition, databases gradually increase, because the data from various fields is generated and accumulated over data streams. Data streams should be processed as intelligently as possible to provide the useful data to the system in real time. In this paper, we propose an efficient erasable pattern mining algorithm that processes uncertain data that is generated over data streams. The uncertain erasable patterns discovered through the suggested technique are more meaningful information by considering the probability of the item and the profit. Moreover, the proposed method can perform efficient mining operations by using both tree and list structures. The performance of the suggested algorithm is verified through the performance tests compared with state-of-the-art algorithms using real data sets and synthetic data sets.

Natural language processing is a useful processing technique of language data, such as text and speech. Sequence labeling represents the upstream task of many natural language processing tasks, such as machine translation, text classification, and sentiment classification. In this paper, the focus is on the sequence labeling task, in which semantic labels are assigned to each unit of a given input sequence. Two frameworks of latent variable conditional random fields (CRF) models (called LVCRF-I and LVCRF-II) are proposed, which use the encoding schema as a latent variable to capture the latent structure of the hidden variables and the observed data. Among the two designed models, the LVCRF-I model focuses on the sentence level, while the LVCRF-II works in the word level, to choose the best encoding schema for a given input sequence automatically without handcraft features. In the experiments, the two proposed models are verified by four sequence prediction tasks, including named entity recognition (NER), chunking, reference parsing and POS tagging. The proposed frameworks achieve better performance without using other handcraft features than the conventional CRF model. Moreover, these designed frameworks can be viewed as a substitution of the conventional CRF models. In the commonly used LSTM-CRF models, the CRF layer can be replaced with our proposed framework as they use the same training and inference procedure. The experimental results show that the proposed models exhibit latent variable and provide competitive and robust performance on all three sequence prediction tasks.

A low complexity pilot allocation scheme is proposed for a large OFDM block with null subcarriers. The proposed scheme allocates pilots to the edge part of the active subcarrier region according to the 2nd order polynomial and to the middle part of the active subcarrier region according to the 1st order polynomial or the comb‐type pilot pattern. To find a parameter of the 2nd order polynomial, the proposed scheme applies exhaustive search of a single parameter by using an integer unit resolution. It is shown by simulation that the proposed scheme is a close‐to‐optimal pilot allocation scheme yielding better symbol error rate (SER) performance than the traditional 3rd and 5th order polynomial‐based schemes although the proposed scheme has lower computation and implementation complexity.

Blind selected mapping (BSLM) is a technique to reduce the peak to average power ratio (PAPR) of OFDM by using a set of scrambling sequences. Since the side information (SI) on the scrambling sequence selected by the transmitter is not directly sent to the receiver in BSLM, the SI needs to be estimated by the receiver. In this paper, a BSLM scheme that applies a new SI estimation approach is proposed. The proposed BSLM scheme estimates the SI by selecting the minimum of the candidate metrics that are computed with the received pilot signal, the pilot symbol sequence, and the scrambling sequences. It is shown by simulation that the proposed BSLM scheme yields better SI error rate (SIER) and bit error rate (BER) performance and lower implementation complexity than four traditional BSLM schemes.

Weighted itemset mining, which is one of the important areas in frequent itemset mining, is an approach for mining meaningful itemsets considering different importance or weights for each item in databases. Because of the merit of the weighted itemset mining, various related works have been studied actively. As one of the methods in the weighted itemset mining, FWI (Frequent Weighted Itemset) mining calculates weights of transactions from weights of items and then finds FWIs based on the transaction weights. However, previous FWI mining methods still have limitations in terms of runtime and memory usage performance. For this reason, in this paper, we propose two algorithms for mining FWIs more efficiently from databases with weights of items. In contrast to the previous approaches storing transaction IDs for mining FWIs, the proposed methods employ new types of prefix tree structures and mine these patterns more efficiently without storing any transaction ID. Through extensive experimental results in this paper, we show that the proposed algorithms outperform state-of-the-art FWI mining algorithms in terms of runtime, memory usage, and scalability.

Abstract As a core area in data mining, frequent pattern (or itemset) mining has been studied for a long time. Weighted frequent pattern mining prunes unimportant patterns and maximal frequent pattern mining discovers compact frequent patterns. These approaches contribute to improving mining performance by reducing the search space. However, we need to consider both the downward closure property and patterns' subset checking process when integrating these different methods in order to prevent unintended pattern losses. Moreover, it is also essential to extract valid patterns with faster runtime and less memory consumption. For this reason, in this paper, we propose more efficient maximal weighted frequent pattern (MWFP) mining approaches based on tree and array structures. We describe how to handle these problems more efficiently, maintaining the correctness of our method. We develop two types of maximal weighted frequent mining algorithms based on weight ascending order and support descending order and compare these two algorithms to conclude which is more suitable for MWFP mining. In addition, comprehensive tests in this paper show that our algorithms are more efficient and scalable than state‐of‐the‐art algorithms, and they also have the correctness of the MWFP mining in terms of their pattern generation results.

Various association rule mining techniques, such as frequent itemset mining, sequence itemset mining, and high utility itemset mining, have been studied to reveal valuable knowledge hidden from large databases. Among these techniques, high utility itemset mining has been researched actively by many researchers because of its characteristics that can find more meaningful itemsets compared to those of other approaches by considering the utility of each item in a given database. In recent years, mining high utility itemsets over data streams has emerged as an interesting topic because many users want to obtain valuable information from stream data, which are continually generated at rapid rates. However, in these environments, most of the previous high utility itemset mining methods cannot efficiently work in terms of both runtime and memory usage. In addition, since they conduct their mining processes without any consideration of transactions' arrival-time, it is hard for these methods to sufficiently fulfill the needs of users when they want to obtain only up to date, relevant information over data streams. In this paper, we propose a new tree-based algorithm that mines recent high utility itemsets over data streams. On the basis of the time decaying model, our algorithm diminishes the utilities of transactions according to their arrival-time in order to assign larger weights to recent data compared to those of older ones. Moreover, the algorithm regularly updates the utility information in its tree data structure and prunes the nodes with the utility values less than a user-specified minimum value. Thereby, the algorithm can maintain a reasonable memory usage bound by avoiding memory use that is unessential. Experimental results demonstrate that our algorithm can mine recent high utility itemsets from varying stream data while consuming smaller computational resources than those of the existing algorithms.

In frequent pattern mining, items are considered as having the same importance in a database and their occurrence are represented as binary values in transactions. In real-world databases, however, items not only have relative importance but also are represented as non-binary values in transactions . High utility pattern mining is one of the most essential issues in the pattern mining field, which recently emerged to address the limitation of frequent pattern mining. Meanwhile, tree construction with a single database scan is significant since a database scan is a time-consuming task. In utility mining, an additional database scan is necessary to identify actual high utility patterns from candidates. In this paper, we propose a novel tree structure, namely SIQ-Tree (Sum of Item Quantities), which captures database information through a single-pass. Moreover, a restructuring method is suggested with strategies for reducing overestimated utilities. The proposed algorithm can construct the SIQ-Tree with only a single scan and decrease the number of candidate patterns effectively with the reduced overestimation utilities, through which mining performance is improved. Experimental results show that our algorithm outperforms a state-of-the-art one in terms of runtime and the number of generated candidates with a similar memory usage.

Since the concept of frequent pattern mining was proposed, there have been many efforts to obtain useful pattern information from large databases. As one of them, applying weight conditions allows us to mine weighted frequent patterns considering unique importance of each item composing databases, and the result of analysis for the patterns provides more useful information than that of considering only frequency or support information. However, although this approach gives us more meaningful pattern information, the number of patterns found from large databases is extremely large in general; therefore, analyzing all of them may become inefficient and hard work. Thus, it is essential to apply a method that can selectively extract representative patterns from the enormous ones. Moreover, in the real-world applications, unexpected errors such as noise may occur, which can have a negative effect on the values of databases. Although the changes by the error are quite small, the characteristics of generated patterns can be turned definitely. For this reason, we propose a novel algorithm that can solve the above problems, called AWMax (an algorithm for mining Approximate weighted maximal frequent patterns (AWMFPs) considering error tolerance). Through the algorithm, we can obtain useful AWMFPs regardless of noise because of the consideration of error tolerance. Comprehensive performance experiments present that the proposed algorithm has more outstanding performance than previous state-of-the-art ones.

A time-reversal (TR)-based transmission using pre-distortion of the transmitted waveforms for intersymbol interference (ISI) alignment is proposed. This scheme differs from the previous TR-based pre-filtering schemes in that it uses distinctively designed pre-distorted waveforms that are based on the transmitted symbol information. In the proposed pre-distortion scheme, the successively received waveforms carrying adjacent transmitted symbols are aligned so that the received symbol power can be intensified. Since the overlapped portions of the successively transmitted waveforms from the proposed pre-distortion scheme are correlated, the power of the overall transmitted signal that is formed by time-shifting and adding the pre-distorted waveforms becomes larger than the sum of the individual pre-distorted waveform powers. In order to adjust the pre-distorted waveform power without measuring the actual power of the overall transmitted signal, an upper bound on the average power of the overall transmitted signal was derived analytically under a low SNR assumption. Simulation showed that the proposed pre-distortion scheme outperformed both conventional TR and the minimum mean square error (MMSE)-based pre-filtering when the ISI was serious with a high transmission rate.

Generally, association rule mining uses only a single minimum support threshold for the whole database. This model implicitly assumes that all items in the database have the same nature. In real applications, however, each item can have different nat

Based on the frequent pattern mining, closed frequent pattern mining and weighted frequent pattern mining have been studied to reduce the search space and discover important patterns. In the previous definition of weighted closed patterns, supports of patterns are only considered to compute the closures of the patterns. It means that the closures of weighted frequent patterns cannot be perfectly checked. Moreover, the usefulness of weighted closed frequent patterns depends on the presence of frequent patterns that have supersets with the exactly same weighted support. However, from the errors such as noise, slight changes in items' supports or weights by them have significantly negative effects on the mining results, which may prevent us from obtaining exact and valid analysis results since the errors can break the original characteristics of items and patterns. In this paper, to solve the above problems, we propose a concept of robust weighted closed frequent pattern mining, and an approximate bound is defined on the basis of the concept, which can relax requirements for precise equality among patterns' weighted supports. Thereafter, we propose a weighted approximate closed frequent pattern mining algorithm which not only considers the two approaches but also suggests fault tolerant pattern mining in the noise constraints. To efficiently mine weighted approximate closed frequent patterns, we suggest pruning and subset checking methods which reduce search space. We also report extensive performance study to demonstrate the effectiveness, efficiency, memory usage, scalability, and quality of patterns in our algorithm.

Maximal frequent pattern mining has been suggested for data mining to avoid generating a huge set of frequent patterns. Conversely, weighted frequent pattern mining has been proposed to discover important frequent patterns by considering the weighted

In data mining area, weight based sequential pattern mining has been suggested to find important sequential patterns by considering the weights of sequential patterns. More extensions with weight constraints have been proposed such as mining weighted

There have been many studies on mining frequent itemset (or pattern) in the data mining field because of its broad applications in mining association rules, correlations, graph patterns, constraint based frequent patterns, sequential patterns, and many other data mining tasks. One of major challeng es in frequent pattern mining is a huge number of result patterns. As the minimum threshold becomes lower, an exponentially large number of itemsets are generated. Therefore, pruning unimportant patterns effectively in mining process is one of main topics in frequent pattern mining. In weighted frequent pattern mining, not only support but also weight are used and important patterns can be detected. In this paper, we propose two efficient algorithms for mining weighted frequent itemsets in which the main approaches are to push weight constraints into the Apriori algorithm and the pattern growth algorithm respectively. Additionally, we show how to maintain the downward closure property in mining weighted frequent itemsets. In our approach, the normalized weights within the weight range are used according to the importance of items. A weight range is used to restrict weights of items and a minimum weight is utilized to balance between weight and support of items for pruning the search space. Our approach generates fewer but important weighted frequent itemsets in large databases, particularly dense databases with low minimum supports. An extensive performance study shows that our algorithm outperforms previous mining algorithms. In addition, it is efficient and scalable.

In this paper, we present a new algorithm, Weighted Interesting Pattern mining (WIP) in which a new measure, weight-confidence, is developed to generate weighted hyperclique patterns with similar levels of weights. A weight range is used to decide weight boundaries and an h-confidence serves to identify strong support affinity patterns. WIP not only gives a balance between the two measures of weight and support, but also considers weight affinity and/or support affinity between items within patterns so more valuable patterns can be generated. A comprehensive performance study shows that WIP is efficient in weighted frequent pattern mining. Moreover, it generates fewer but more valuable patterns for users.

Researchers have proposed weighted frequent itemset mining algorithms that reflect the importance of items. The main focus of weighted frequent itemset mining concerns satisfying the downward closure property. All weighted association rule mining algorithms suggested so far have been based on the Apriori algorithm. However, pattern growth algorithms are more efficient than Apriori based algorithms. Our main approach is to push the weight constraints into the pattern growth algorithm while maintaining the downward closure property. In this paper, a weight range and a minimum weight constraint are defined and items are given different weights within the weight range. The weight and support of each item are considered separately for pruning the search space. The number of weighted frequent itemsets can be reduced by setting a weight range and a minimum weight, allowing the user to balance support and weight of itemsets. WFIM generates more concise and important weighted frequent itemsets in large databases, particularly dense databases with low minimum support, by adjusting a minimum weight and a weight range.