In this paper, we introduce a new class of Association Rules (ARs) named "Multi-Relation Association Rules" which in contrast to primitive ARs (that are usually extracted from multi-relational databases), each rule item consists of one entity and several relations. These relations indicate indirect relationship between entities. Consider the following Multi-Relation Association Rule where the first item consists of three relations live in, nearby and humid: "Those who live in a place which is near by a city with humid climate type and also are younger than20®their health condition is good". A new algorithm called MRARis proposed to extract such Rules from directed graphs with labeled edges which are constructed from RDBMSs or semantic web data. Also, the question "how to convert RDBMS data or semantic web data to a directed graph with labeled edges?"is answered. In order to evaluate the proposed algorithm, some experiments are performed on a sample dataset and also a real-world drug semantic web dataset. Obtained results confirm the ability of the proposed algorithm in mining Multi-Relation Association Rules.

Association Rules are among important techniques in data mining which are used for extracting hidden patterns and knowledge in large volumes of data. Association Rules help individuals and organizations take strategic decisions and improve their business processes. Extracted Association Rules from a database contain important and confidential information that if published, the privacy of individuals may be threatened. Therefore, the process of hiding sensitive Association Rules should be performed prior to sharing the database. This is done through changing the database transactions. These changes must be made in such a way that all sensitive Association Rules are hidden and a maximum number of non-sensitive Association Rules are extractable from the sanitized database. In fact, a balance is to be established between hiding the sensitive Rules and extracting the non-sensitive Rules. A new algorithm is presented in this paper to create a balance between preserving privacy and extracting knowledge. The items of sensitive Rules are clustered in the proposed algorithm, in order to reduce changes. In fact, reduction of changes and clustering of Rules are applied in order to reduce the side effects of the hiding process on non-sensitive Rules.

Association Rules are one of the data and web mining techniques which aim to discover the frequent patterns among itemsets in a transactional database. Frequent patterns and correlation between itemsets in datasets and databases are extracted by these interesting Rules. The Association Rules are positive or negative, and each has its own specific characteristics and definitions. The mentioned algorithms of the discovery of Association Rules are always facing challenges, including the extraction of only positive Rules, while negative Rules in databases are also important for a manager’ s decision making. Also, the threshold level for support and confidence criteria is always manual with trial and error by the user and the proper place or the characteristics of datasets is not clear for these Rules. This research analyses the behavior of the negative Association Rules based on trial and error. After analyzing the available algorithms, the most efficient algorithm is implemented and then the negative Rules are extracted. This test repeats on several standard datasets to evaluate the behavior of the negative Rules. The analyses of the achieved outputs reveal that some of the interesting patterns are detected by the negative Rules, while the positive Rules could not detect such helpful Rules. This study emphasizes that extracting only positive Rules for covering Association Rules is not enough.

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.