Search

Association rules provide important knowledge that can be extracted from transactional databases. Owing to the massive exchange of information nowadays, databases become dynamic and change rapidly and periodically: new transactions are added to the database and/or old transactions are updated or removed from the database. Incremental mining was introduced to overcome the problem of maintaining previously generated association rules in dynamic databases. In this paper, we propose an efficient algorithm (IMIDB) for incremental itemset mining in large databases. The algorithm utilizes the trie data structure for indexing dynamic database transactions. Performance comparison of the proposed algorithm to recently cited algorithms shows that a significant improvement of about two orders of magnitude is achieved by our algorithm. Also, the proposed algorithm exhibits linear scalability with respect to database size.

Mining temporal association rules is a very interesting topic, which is applied in many applications nowadays. In temporal databases, each item has its own lifetime period, called exhibition period, which is different from other items. With the rapid increase of databases and new transactions added, the incremental mining is introduced to solve the problem of maintaining association rules in updated databases. The existing algorithms did not utilize the previously discovered rules when the database is updated. This paper presents an efficient algorithm for mining incremental temporal association rules. The proposed algorithm, called IndxTAR, utilizes two major components: a relatively new data structure and previously discovered frequent temporal itemsets to improve the performance of mining the incremental temporal association rules. Experiments on both real and synthetic datasets are conducted to compare IndxTAR algorithm performance with recently cited incremental temporal mining algorithms. The results show that IndxTAR algorithm overcomes other algorithms by many orders of magnitude and can efficiently process large databases with linear scalability. [ABSTRACT FROM AUTHOR]

Background: The risk of hemorrhagic transformation (HT) after cerebral ischemia is a major problem. There are no firm prognostic criteria to predict the risk of HT after the ischemic event. Objective: The aim of this study was to investigate the predictive factors of HT in acute ischemic stroke. Methods: Consecutive patients admitted with acute cerebral infarction were recruited and submitted to clinical evaluation including National Institute of Health Stroke Scale (NIHSS), laboratory investigations, baseline computed tomography (CT) scan, magnetic resonance imaging (MRI) and magnetic resonance angiography (MRA), follow up CT scan within one week from onset. Results: Forty nine patients out of 324 patients (15.1%) with ischemic stroke developed HT. Patients with HT had higher NIHSS score (P=0.001), larger volume of infarction (P=0.001), more prevalent early ischemic changes in baseline CT (P=0.001), microbleeds and significant intracranial arterial stenosis in MRA (P=0.001). Also, HT patients tended to have higher levels of random blood sugar (RBS) (P=0.0001), higher total leukocyte count (TLC) (P=0.002), higher values of international normalized ratio (INR) (P=0.02), and lower cholesterol levels (P=0.002). AF and cardioembolic stroke were more common among patients who developed HT (P=0.001). Conclusion: Cardioembolic stroke, stroke attributable to large artery atherosclerosis, AF, large infarction size, microbleeds, high RBS and INR on admission are independent predictors of HT. [ABSTRACT FROM AUTHOR]