In most of the business areas, many organizations are running similar trades and maintaining comparable databases. These organizations have noticed the importance of analyzing results obtained from the union of databases owned by different organizations. However, they do not want to disclose their contents to others since some of the contents are sensitive and private. Recently, preference-based queries have drawn massive attention in the database community. Particularly, the top-k dominating queries have been studied extensively, which selects the k objects that are better than other objects based on the `domination score'. In this paper, we have considered the top-k dominating queries on the combined databases of different organizations. We propose a secure framework for multi-party top-k dominating queries, where individual organizations do not need to expose their private databases to others. We analyze the privacy of our proposed framework and also evaluate its performance for various settings.

Recommendation systems address the "information overload" problem by filtering out items that customers may not be interested. Collaborative Filtering (CF) is the most popular technique which recommends items to a user based on similar users' preferences and/or similar items records by utilizing the user-item rating matrix. However, such a large amount of explicit feedbacks are not always available. Furthermore, user preferences are changing over time. The Conventional CF cannot capture the temporal dynamics of recommendations well. In this paper, we apply Deep Recurrent Neural Networks (DRNNs) to CF and generate dynamic, personalized recommendations by utilizing user purchase history big data. Experiments on the MovieLens dataset show relative improvements over previously reported results.

Selection of good locations in a map is an indispensable function in many applications. In order to select specific locations, we have to specify detailed selection criteria. However, it is not easy especially for users of mobile devices. Therefore, we used an idea of skyline queries, which are known to be easy and effective to retrieve interesting data from a database. In our previous work, we have proposed area skyline query that selects good locations in a map. However, the query is not fast enough for handling “big data”. We simplify and revise the algorithm of the query in this paper by using MapReduce framework so that we can use it for big data. Experiments' results demonstrate that the performance and scalability are superior to previous area skyline algorithm and are able to handle big data.

We consider the problem of finding a set of attribute values that give a high quality binary segmentation of a database. The quality of a segmentation is defined by an objective function suitable for the user's objective, such as "mean squared error," "mutual information," or "/spl chi//sup 2/" each of which is defined in terms of the distribution of a given target attribute. Our goal is to find value groups on a given conditional domain that split databases into two segments, optimizing the value of an objective function. Though the problem is intractable for general objective functions, there are feasible algorithms for finding high quality binary segmentations when the objective function is convex, and we prove that the typical criteria mentioned above are all convex. We propose two practical algorithms, based on computational geometry techniques, which find a much better value group than conventional heuristics.

We consider the problem of finding neighboring class sets. Objects of each instance of a neighboring class set are grouped using their Euclidean distances from each other. Recently, location-based services are growing along with mobile computing infrastructure such as cellular phones and PDAs. Therefore, we expect to see the development of spatial databases that contains very large number of access records including location information. The most typical type would be a database of point objects. Records of the objects may consist of "requested service name," "number of packet transmitted" in addition to x and y coordinate values indicating where the request came from. The algorithm presented here efficiently finds sets of "service names" that were frequently close to each other in the spatial database. For example, it may find a frequent neighboring class set, where "ticket" and "timetable" are frequently requested close to each other. By recognizing this, location-based service providers can promote a "ticket" service for customers who access the "timetable."

We discuss data mining based on association rules for two numeric attributes and one Boolean attribute. For example, in a database of bank customers, Age and Balance are two numeric attributes, and CardLoan is a Boolean attribute. Taking the pair (Age, Balance) as a point in two-dimensional space, we consider an association rule of the form ((Age,Balance) ∈P)⇒(CardLoan = Yes), which implies that bank customers whose ages and balances fall within a planar region P tend to take out credit card loans with a high probability.We consider two classes of regions, rectangles and admissible (i.e., connected and x-monotone) regions. For each class, we propose efficient algorithms for computing the regions that give optimal association rules for gain, support, and confidence, respectively. We have implemented the algorithms for admissible regions as well as several advanced functions based on them in our data mining system named SONAR (System for Optimized Numeric Association Rules), where the rules are visualized by using a graphic user interface to make it easy for users to gain an intuitive understanding of rules.

Given a huge database, we address the problem of finding association rules for numeric attributes, such as Permission to make digital/hard copies of all or pati of this material for personal or claasroom usc is granted without fee provided that the copies are not made or distributed for pro~t or cornmesvial advantage, the copyright notice, the title of the publication and Its date appear, and notice is given that cop yright is by permission of the ACM, Inc.To copy otherwise, to repubtish, to post on servers or to redistribute to lists, requires specific permission and/or fee.PODS '96, Montreal Quebec Canada

We discuss data mining based on association rules for two numeric attributes and one Boolean attribute. For example, in a database of bank customers, "Age" and "Balance" are two numeric attributes, and "CardLoan" is a Boolean attribute. Taking the pair (Age, Balance) as a point in two-dimensional space, we consider an association rule of the form((Age, Balance) ∈ P) ⇒ (CardLoan = Yes),which implies that bank customers whose ages and balances fall in a planar region P tend to use card loan with a high probability. We consider two classes of regions, rectangles and admissible (i.e. connected and x-monotone) regions. For each class, we propose efficient algorithms for computing the regions that give optimal association rules for gain, support, and confidence, respectively. We have implemented the algorithms for admissible regions, and constructed a system for visualizing the rules.

No abstract available.