Mining Multiple Databases Research Articles

A New Framework for Discovering Protein Complex and Disease Association via Mining Multiple Databases.

One important challenge in the post-genomic era is to explore disease mechanisms by efficiently integrating different types of biological data. In fact, a single disease is usually caused through multiple genes products such as protein complexes rather than single gene. Therefore, it is meaningful for us to discover protein communities from the protein-protein interaction network and use them for inferring disease-disease associations. In this article, we propose a new framework including protein-protein networks, disease-gene associations and disease-complex pairs to cluster protein complexes and infer disease associations. Complexes discovered by our approach is superior in quality (Sn, PPV and ACC) and clustering quantity than other four popular methods on three PPI networks. A systematic analysis shows that disease pairs sharing more protein complexes (such as Glucose and Lipid Metabolic Disorders) are more similar and overlapping proteins may have different roles in different diseases. These findings can provide clinical scholars and medical practitioners with new ideas on disease identification and treatment.

Global Data Fusion versus Local Pattern Fusion in Mining Multiple Databases: A Comparative Review

With the emergence of big data, mining distributed databases has become a critical task in the domain of discovery of knowledge from databases. Many of the traditional multiple-database mining methods developed until now have emphasized mining the mono-database, which is a pool of all the local databases merged at a central site; local patterns discovered at local sites are not analyzed in mono-database mining. However, in real-world applications, data collected from multiple databases may be duplicitous and unreliable. Therefore, developing methods to discover reliable, high-quality knowledge from multiple databases is a challenging task when mining multi-sourced data. This paper scrupulously reviews all the existing methods for mining multiple and distributed databases based on global data fusion and local pattern fusion techniques. The research issues and recently developed methods, which involves local pattern analysis in multi-database mining, are also discussed.

A State-of-the-Art Review of Knowledge Discovery in Multiple Databases

AbstractKnowledge discovery in multiple databases offers many opportunities and challenges. We have given a number of motivating points on knowledge discovery in multiple databases. In view of further studies on this aspect, we highlight some domains that generated numerous problems on multiple related databases. Activities related to data preprocessing in a multi-database mining environment are also discussed. Important techniques of mining multiple databases are outlined. Many interesting patterns that originated out of multi-database environments are highlighted. We shall witness more research outcomes and investigations as the number of multi-database domains is on the rise.

Clustering local frequency items in multiple databases

Frequent items could be considered as a basic type of patterns in a database. In the context of multiple data sources, most of the global patterns are based on local frequency items. A multi-branch company transacting from different branches often needs to extract global patterns from data distributed over the branches. Global decisions could be taken effectively using such patterns. Thus, it is important to cluster local frequency items in multiple databases. An overview of the existing measures of association is presented here. For the purpose of selecting the suitable technique of mining multiple databases, we have surveyed the existing multi-database mining techniques. A study on the related clustering techniques is also covered here. The notion of high frequency itemsets is introduced here, and an algorithm for synthesizing supports of such itemsets is designed. The existing clustering technique might cluster local frequency items at a low level, since it estimates association among items in an itemset with a low accuracy, and thus a new algorithm for clustering local frequency items is proposed. Due to the suitability of measure of association A2, association among items in a high frequency itemset is synthesized based on it. The soundness of the clustering technique has been shown. Numerous experiments are conducted using five datasets, and the results on different aspects of the proposed problem are presented in the experimental section. The effectiveness of the proposed clustering technique is more visible in dense databases.