Privacy-preserving data mining is a novel research direction in data mining and statistical databases, where data mining algorithms are analyzed for the side effects they incur in data privacy. There have been two types of privacy proposed concerning data mining. The first type of privacy, called output privacy, is that the data is altered so that the mining result will preserve certain privacy. The second type of privacy, called input privacy, is that the data is manipulated so that the mining result is not affected or minimally affected. In output privacy, given specific rules to be hidden, many data altering techniques for hiding association, classification and clustering rules have been proposed. However, to specify hidden rules, entire data mining process needs to be executed. For some applications, we are only interested in hiding certain sensitive predictive rules that contain given items. A predictive association rule set is the smallest rule set that makes the same prediction as the whole association rule set by confidence priority. In this work, we assume that only sensitive items are given and propose two algorithms, ISL (increase support of LHS) and DSR (decrease support of RHS), to replace data by unknowns in database so that sensitive predicative rules containing specified items on the left hand side of rule cannot be inferred through association rule mining. Examples illustrating the proposed algorithms are given. The characteristics of the algorithms are analyzed. The efficiency of the proposed approach is further compared with Saygin etc. approach. It is observed that our approach required less number of databases scanning and prune more number of hidden rules. However, our approach must hide all rules containing the hidden items on the left hand side, where Saygin etc approach can hide any specific rule.

Privacy-preserving data mining is a novel research direction in data mining and statistical databases, where data mining algorithms are analyzed for the side effects they incur in data privacy. For example, through data mining, one is able to infer sensitive information, including personal information or even patterns, from nonsensitive information or unclassified data. There have been two types of privacy concerning data mining. The first type of privacy is that the data is altered so that the mining result will preserve certain privacy. The second type of privacy is that the data is manipulated so that the mining result is not affected or minimally affected. Given specific rules to be hidden, many data altering techniques for hiding association, classification and clustering rules have been proposed. However, to specify hidden rules, entire data mining process needs to be executed. For some applications, we are only interested in hiding certain sensitive predicative rules that contain given items. In this work, we assume that only sensitive items are given and propose two algorithms, ISL (Increase Support of LHS) and DSR (Decrease Support of RHS), to modify data in database so that sensitive predicative rules containing specified items on the left hand side of rule cannot be inferred through association rule mining. Examples illustrating the proposed algorithms are given. The characteristics of the algorithms are analyzed. The efficiency of the proposed approach is further compared with Verykios etc. [2001, 2004] approach. It is shown that our approach required less number of databases scanning and prune more number of hidden rules. However, our approach must hide all rules containing the hidden items on the left hand side, where Verykios etc approach can hide any specific rule.

Privacy-preserving data mining [Agrawal, R., et al., May 2000] has recently emerged to address one of the negative sides of data mining technology: the threat to individual privacy. For example, through data mining, one is able to infer sensitive information, including personal information or even patterns, from non-sensitive information or unclassified data. There have been two broad approaches for privacy-preserving data mining. The first approach is to alter the data before delivery to the data miner so that real values are obscured. The second approach assumes the data is distributed between two or more sites, and these sites cooperate to learn the global data mining results without revealing the data at their individual sites. Given specific rules to be hidden, many data altering techniques for hiding association, classification and clustering rules have been proposed. However, to specify hidden rules, entire data mining process needs to be executed. For some applications, we are only interested in hiding certain sensitive items. In this work, we assume that only sensitive items are given and propose two algorithms to modify data in database so that sensitive items cannot be inferred through association rules mining algorithms. Examples illustrating the proposed algorithms are given. The efficiency of the proposed approach is further compared with Dasseni etc. (2001) approach. It is shown that our approach required less number of databases scanning and prune more number of hidden rules. However, our approach must hide all rules containing the hidden items on the right hand side, where Dasseni's approach can hide specific rules.

Most conventional data mining algorithms identify the relation among transactions with binary values. Transactions with quantitative values are, however, commonly seen in real world applications. In the past, we proposed a fuzzy mining algorithm based on the Apriori approach to explore interesting knowledge from the transactions with quantitative values. This paper proposes another new fuzzy mining algorithm based on the AprioriTid approach to find fuzzy association rules from given quantitative transactions. Each item uses only the linguistic term with the maximum cardinality in later mining processes, thus making the number of fuzzy regions to be processed the same as that of the original items. The algorithm therefore focuses on the most important linguistic terms for reduced time complexity.

Web mining of browsing patterns including simple sequential patterns and sequential patterns with browsing times has been studied recently However, most of these works focus on mining browsing patterns of web pages directly In this work, we introduce the problem of mining generalized browsing patterns on cross-levels of a taxonomy comprised of web pages. An algorithm called Apriori-Level is proposed to discover these cross-level generalized browsing patterns. Example demonstrating the proposed approach is given. Comparison with the Generalized Sequential Patterns (GSP) algorithm [16] shows that our approach generates fewer candidate sequences. The proposed algorithm thus promotes the efficiency of discovering coarser granularity of web browsing patterns.

Due to the increasing use of very large databases and data warehouses, mining useful information and helpful knowledge from transactions is evolving into an important research area. Most conventional data-mining algorithms identify the relationships among transactions using binary values and find rules at a single concept level. Transactions with quantitative values and items with hierarchy relation are, however, commonly seen in real-world applications. In this paper, we thus introduce the problem of mining fuzzy generalized association rules from quantitative data. A fuzzy mining algorithm based on Srikant and Agrawal's method is proposed for extracting implicit generalized knowledge from transactions stored as quantitative values. It integrates fuzzy-set concepts and generalized data mining technologies to achieve this purpose. Items in rules may be from any level of the given taxonomy. The effect of numbers of fuzzy regions on the performance of the proposed algorithm is also discussed.

Web mining of browsing patterns including simple sequential patterns and sequential patterns with browsing times has been studied recently. However, most of these works focus on mining browsing patterns of Web pages directly. In this work, we introduce the problem of mining browsing patterns on cross-levels of a taxonomy comprised of Web pages. In addition, browsing time is considered and processed using fuzzy set concepts to form linguistic terms. The proposed algorithm thus discovers cross-level relevant browsing behavior from linguistic data and promotes the discovery of coarsened granularity of Web browsing patterns.

World-wide-web applications have grown very rapidly and have made a significant impact on computer systems. Among them, web browsing for useful information may be most commonly seen. Due to its tremendous amounts of use, efficient and effective web retrieval has thus become a very important research topic in this field. Data mining is the process of extracting desirable knowledge or interesting patterns from existing databases for a certain purpose. In this paper, we use the data mining techniques to discover relevant browsing behavior from log data in web servers, thus being able to help make rules for retrieval of web pages. The browsing time of a customer on each web page is used to analyze the retrieval behavior. Since the data collected are numeric, fuzzy concepts are used to process them and to form linguistic terms. A sophisticated web-mining algorithm is thus proposed to find relevant browsing behavior from the linguistic data. Each page uses only the linguistic term with the maximum cardinality in later mining processes, thus making the number of fuzzy regions to be processed the same as the number of the pages. Computational time can thus be greatly reduced. The patterns mined out thus exhibit the browsing behavior and can be used to provide some appropriate suggestions to web-server managers.

Data mining of association rules from items in transaction databases has been studied extensively in recent years. In order to discover more practical rules, domain knowledge such as taxonomies of items [9] and similarity among items [11] have been considered to produce generalized association rules and similar association rules respectively. However, these algorithms deal with only transactions with binary values whereas transactions with quantitative values are more commonly seen in real-world applications. This paper thus proposes a new data-mining algorithm for extracting fuzzy knowledge from transactions stored as quantitative values. The proposed algorithm integrates fuzzy set concepts and the a priori mining algorithm to find fuzzy similar association rules in given transaction data sets where similarity relations are assumed among database items. The rules discovered here thus promote coarser granularity of association rules and exhibit quantitative regularity under similarity relations.

The discovery of functional dependencies (FDs) in relational databases is an important data mining problem. Most current work assumes that the database is static, and a database update requires rediscovering all the FDs by scanning the entire old and new database repeatedly. In this work, we present an efficient data mining algorithm to incrementally discover all FDs in the presence of a new set of tuples added to an old database. Based on the concept of tuple partitions and the monotonicity of FDs, we avoid re-scanning of the database and thereby reduce the computation time. The computational complexity of the proposed algorithm is analyzed. A comparison the with the pair-wise comparison-based incremental approach is also presented. The results show that an improved computation time is achieved, while extra space is required for partitions by our approach

Transaction data in real-world applications usually consist of quantitative values, so designing a sophisticated data-mining algorithm that is able to deal with various types of data presents a challenge to workers in this research field. Since sequential patterns are also very important for real-world applications, this paper focuses on finding fuzzy sequential patterns from quantitative data. A new mining algorithm is proposed, which integrates the fuzzy-set concepts and the AprioriAll algorithm. It first transforms quantitative values in transactions into linguistic terms, then filters them to find sequential patterns by modifying the AprioriAll mining algorithm. Each quantitative item uses only the linguistic term with the maximum cardinality in later mining processes, thus making the number of fuzzy regions to be processed the same as the number of the original items. The patterns mined out thus exhibit the sequential quantitative regularity in databases and can be used to provide some Suggestions to appropriate supervisors

This paper introduces the problem of mining similar association rules from transaction databases. Mining of association rules among items in transaction databases has been studied extensively in recent years. Various forms of data types such as quantitative and categorical data in databases as well as domain knowledge such as taxonomies have been considered in the mining algorithms. In this work, we present a data mining algorithm to discover similar association rules where similarity relations are assumed among database items. The proposed algorithm is a generalization of the apriori mining algorithm of association rules. The results developed here can be applied to cross-marketing, customer segmentation and prediction.

This paper presents a set of data mining methods to facilitate the discovery of inclusion dependencies (IND) from database relations. An inclusion dependency states that values in column of one relation must appear as values in column of some other relation. The mining of inclusion dependencies consists of two basic tasks: validation and searching. Many set and non-set oriented data mining techniques of inclusion dependencies for static database relations have been proposed. In this work, we propose three new set-oriented (SQL Language) validation techniques and one heuristic searching method of inclusion dependencies for static and dynamic databases. Computational complexities of the new validation techniques are analyzed, Numerical simulation of the heuristic searching method is performed. The results show good computational efficiencies at the cost of minor space requirements. The methods developed here can be applied to database maintenance and database reverse engineering.

In this paper, we use the data mining techniques to discover relevant browsing behavior from log data in web servers. The browsing time of a customer on each web page is used to analyze the retrieval behavior. Since the data collected are numeric, fuzzy concepts are used to process them and to form linguistic terms. A sophisticated web-mining algorithm is thus proposed to find relevant browsing behavior from the linguistic data. Each page uses only the linguistic term with the maximum cardinality in later mining processes, thus making the number of fuzzy regions to be processed the same as the number of the pages. Computational time can thus be greatly reduced.

Expert systems have been widely used in domains where mathematical models cannot be easily built, human experts are not available or the cost of querying an expert is high. Machine learning or data mining can extract desirable knowledge or interesting patterns from existing databases and ease the development bottleneck in building expert systems. In the past we proposed a method [Hong, T.P., Wang, T.T., Wang, S.L. (2000). Knowledge acquisition from quantitative data using the rough-set theory. Intelligent Data Analysis (in press).], which combined the rough set theory and the fuzzy set theory to produce all possible fuzzy rules from quantitative data. In this paper, we propose a new algorithm to deal with the problem of producing a set of maximally general fuzzy rules for coverage of training examples from quantitative data. A rule is maximally general if no other rule exists that is both more general and with larger confidence than it. The proposed method first transforms each quantitative value into a fuzzy set of linguistic terms using membership functions and then calculates the fuzzy lower approximations and the fuzzy upper approximations. The maximally general fuzzy rules are then generated based on these fuzzy approximations by an iterative induction process. The rules derived can then be used to build a prototype knowledge base in a fuzzy expert system.

In simple flow-shop problems, each machine operation center includes just one machine. If at least one machine center includes more than one machine, the scheduling problem becomes a flexible flow-shop problem. Flexible flow-shops can thus be thought of as generalizations of simple flow-shops. C. Sriskandarajah and S.P. Sethi (1989) proposed a heuristic algorithm for solving flexible flow-shop problems for two machine centers. In this paper, we extend their algorithm to solve flexible flow-shop problems for more than two machine centers. The heuristic Gupta algorithm is adopted as the kernel for achieving this purpose. Since this problem is NP-complete, optimal solutions seem unnecessary, especially when the number of jobs is large.

Present a data mining method for discovering fuzzy functional dependencies from quantitative data. A number of algorithms for mining functional dependencies from a crisp relational data model have been proposed recently. The concept of functional dependency is based on the assumption that data are either fully dependent or fully independent. However, this assumption is too restrictive for real-world applications. Our paper thus proposes a method to extract fuzzy functional dependencies from quantitative data. The dependencies discovered contain not only the conventional functional dependencies but also semantic dependencies between attributes. The discovered dependencies call be used in the reconstruction of the conceptual structure of relations and applied further to the integration of knowledge bases and databases as well as reverse engineering

We present a method for handling null queries that contains interval-valued ambiguous attributes. Null queries are queries that elicit a null answer from the database. Ambiguous attributes in null queries are attributes that are not defined in the original database schema but can be compounded from multiple rigid attributes in schema. Ambiguous attributes that are compounded from simple numbers were first studied by Nomura et al. (1992). In this work, we extend ambiguous attributes to fuzzy databases which contain data types of numbers and interval values. Fuzzy database management systems that can handle this type of ambiguous attributes not only can reduce the occurrences of null answers but also provide an improved query environment