Many parallelization techniques have been proposed to enhance the performance of Apriori-like frequent itemset mining algorithms. Characterized by both map and reduce functions, MapReduce has emerged excels in datasets terabyte scale or larger either homogeneous heterogeneous clusters. Minimizing scheduling overhead each map-reduce phase maximizing utilization nodes are keys successful implementations. In this paper, we propose three algorithms, named SPC, FPC, DPC, investigate effective...

Objective: The number of patients requiring prolonged mechanical ventilation (PMV) is increasing worldwide, but the weaning outcome prediction model in these still lacking. We hence aimed to develop an explainable machine learning (ML) predict successful PMV using a real-world dataset. Methods: This retrospective study used electronic medical records admitted 12-bed respiratory care center central Taiwan between 2013 and 2018. three ML models, namely, extreme gradient boosting (XGBoost),...

Mining of sequential patterns in a transactional database is time consuming due to its complexity. Maintaining present non-trivial task after update, since appended data sequences may invalidate old and create new ones. In contrast re-mining, the key improve mining performance proposed incremental update algorithm effectively utilize discovered knowledge. By counting over instead entire updated most cases, fast filtering found last successive reductions candidate together make efficient on possible.

The importance of the data stored in smart phones is increased as more applications are deployed and executed. Once phone damaged or lost, valuable information treasured device lost altogether. If cloud storage can be integrated with services for periodical backup a mobile client, risk minimized. However, important might uncovered by malicious third party during retrieval transmission using wireless without proper authentication protection. Therefore, this paper, we design an archive...

The advancement of explainable recommendations aims to improve the quality textual explanations for recommendations. Traditional methods primarily used Recurrent Neural Networks (RNNs) or their variants generate personalized explanations. However, recent research has focused on leveraging Transformer architectures enhance by extracting user reviews and incorporating features from interacted items. Nevertheless, previous studies have failed fully exploit relationship between ratings more In...

Sequential patterns in customer transactional databases are commonly mined for E-Commerce recommendations. In many practical applications, the absence of certain item-sets and sequences could have important implications. Mining frequent comprising not only occurrence but also will increase accuracy product A sequential pattern containing at least one absent itemset is called a negative pattern. this paper, we formulate problem mining by introducing constraints propose an algorithm PNSP...

Sequential pattern mining is a challenging issue because of the high complexity temporal discovering from numerous sequences. Current approaches either require frequent database scanning or generation several intermediate databases. As databases may fit into ever-increasing main memory, efficient memory-based discovery sequential patterns becoming possible. In this paper, we propose memory indexing approach for fast mining, named MEMISP. During whole process, MEMISP scans sequence only once...

The mining of closed sequential patterns has attracted researchers for its capability using compact results to preserving the same expressive power as traditional mining. Many studies have shown that constraints are essential applications patterns. However, time not been incorporated into sequence yet. Therefore, we propose an algorithm called CTSP pattern with constraints. loads database memory and constructs time-indexes facilitate both closure checking, within growth framework. index sets...

Common sequential pattern mining algorithms handle static databases. Once the data change, previous result will be incorrect, and we need to restart entire process for new updated sequence database. Previous approaches, within either Apriori-based or projection-based framework, mine patterns in a forward manner. Considering incremental characteristics of sequence-merging, develop novel technique, called backward mining, efficient discovery. We propose an algorithm, BSPinc, using strategy....

Lungs and kidneys are two vital frequently injured organs among critically ill patients. In this study, we attempt to develop a weaning prediction model for patients with both respiratory renal failure using an explainable machine learning (XML) approach. We used the eICU collaborative research database, which contained data from 335 ICUs across United States. Four ML models, including XGBoost, GBM, AdaBoost, RF, were used, feature windows, at 48 h. The model's explanations presented domain,...

Many modern applications such as sensor networks produce probabilistic data. These data are collected into an uncertain database. To interpret uncertainty and to mine frequent patterns in database, all possible certain databases considered, which generates exponential number of combinations makes the mining problem highly complicated. In practice, is interactive, discovery itemsets database even more challenging. The objective interactive shorten time that required obtain desired iterated...

In response to globalization, International Financial Reporting Standards (IFRS) has become the norm of global capital markets. Companies preparing financial statements using IFRS may make situation fully disclosed. Nevertheless, an overestimated accrual expense a balance sheet not only underestimate earnings data, but also increase cash outflows statement flows. When is underestimated, corporate will inflate statistics. addition, problem funds shortage occur upon actual payment because...

The discovery of sequential patterns, which extends beyond frequent item-set finding association rule mining, has become a challenging task due to its complexity. Essentially, user would specify minimum support threshold with respect the database find out desired patterns. mining process is usually iterative since must try various thresholds obtain satisfactory result. Therefore, time-consuming be repeated several times. However, current approaches are inadequate for such long execution time...

Mining frequent itemsets in an uncertain database is a highly complicated problem. Most algorithms focus on improving the mining efficiency with assumption that static. Uncertain databases, however, are constantly updated newly appended transactions like certain databases. Some patterns may become obsolete and new ones emerge due to updates. Remining whole from scratch very time-consuming owing frequentness probabilities computations. To tackle this maintenance problem, we propose algorithm...

Image classification is an important technique for effective content-based multimedia retrieval. Many classifiers have been proposed while frequent patterns based approaches received many attentions in recent years. In this paper, we image approach utilizing sequential discovered from distinct classes. The segmented and low-level features are extracted as a sequence of feature-sets. Sequence-rules collected each class conflict rules resolved by rule pruning. Useful then selected to form the...