Large-scale DATA may consist of big, DISTRIBUTED, scattered, heterogeneous, irrelevant, misleading, real, and unrealistic DATA or any combination of them. Therefore, analyzing, creating value and DATA productivity is always an important and open challenge. Therefore, the purpose of this study is to present a new coalition architecture for generating valuable information for decision making among the masses of DATA. The proposed architecture, abbreviated ASMLDE, aims to develop and improve DATA MINING and semantic exploration, and to produce useful and high-quality rules consisting of four layers, seven components and six key elements. In the proposed architecture, conceptualization with 4v's process, insight into the volume and scale of DATA in the form of 3v's model and finally qualitative insight based on DATA thickness, are used for conceptualization and standardization of qualitative processes and more complex interpretations. This architecture, supported by ontology and agent MINING, reduces large search spaces and increases the speed and quality of DATA MINING operations due to the use of multi-agent systems. Automating exploration operations, reducing DATA complexity and business processes are also important achievements of the proposed architecture. To evaluate the proposed architecture, a large-scale DATAset of natural disasters and earthquake ontology classes from the DBpedia knowledge base have been used. The evaluation results obtained by exploring the semantic rules of the mentioned DATAset highlight the effectiveness and capabilities of the ASMLDE architecture in enhancing the quality of the semantic rules explored to fit the user need and reducing the large DATA MINING space over other similar frameworks and architectures.

Background: Provide a health care service to the patients with diabetes provides useful information that could be used to identify, treatment, following up and prevention of diabetes. Explore and investigation of large volumes of DATA requires effective and efficient methods for finding hiding patterns in the DATA. The use of various techniques of DATA MINING in particular Classification and Frequent patterns can be helpful.Methods: This article is a narrative review. We searched keywords related to application of DATA MINING in the field of diabetes, through related DATAbases, in English language articles published from 2005 to 2015. Also related articles in the selected articles list have been analyzed.Results: From the 2144 articles obtained in the initial search, 38 articles related to the subject of study, were selected. Several studies shown that classification and clustering algorithms, association rules and artificial intelligence are the most widely used DATA MINING techniques for predict the risk of diabetes has been successfully used.Conclusion: The important step in control of diabetes, use of the methods that could determine the possibility or lack of diabetes. According to studies conducted in this area seem to use DATA MINING techniques to prevent, treat and discover the connection between diabetes and its risk factors, can lead to significant improvements in the field of diabetes research and provide better health care for this group of patients.

MINING association rules in DISTRIBUTED DATAbases is an interesting problem in the context of parallel and DISTRIBUTED DATA MINING. A number of approaches have, so far, been proposed for DISTRIBUTED MINING of association rules. However, most of them consider all types of frequent itemsets the same, even though there are different types of itemsets in DISTRIBUTED DATAbases, e.g., derivable and non-derivable. In this study, a new application of deduction rules is introduced for DISTRIBUTED MINING of association rules which exploits the derivability of itemsets to reduce communication overhead and to enhance response time. A new algorithm is proposed which mines derivable and non-derivable frequent itemsets in a DISTRIBUTED DATAbase. Since the collection of derivable and non-derivable frequent itemsets form all frequent itemsets, our algorithm mines all frequent itemsets rather than a subset of them. In the algorithm, there is no need to scan local DATAbases and exchange messages in order to obtain support counts of derivable frequent itemsets, since each site can produce them autonomously. Experimental evaluations on horizontally partitioned real-life DATAsets show that such exploitation drastically reduces communication and also improves response time.  Therefore the new algorithm is useful when communication bandwidth is the main bottleneck.