Showing papers by "Pang-Ning Tan published in 2001"

Mining Indirect Associations in Web Data

Abstract: Web associations are valuable patterns because they provide useful insights into the browsing behavior of Web users. However, there are two major drawbacks of using current techniques for mining Web association patterns, namely, their inability to detect interesting negative associations in data and their failure to account for the impact of site structure on the support of a pattern. To address these issues, a new data mining technique called indirect association is applied to the Web click-stream data. The idea here is to find pairs of pages that are negatively associated with each other, but are positively associated with another set of pages called the mediator. These pairs of pages are said to be indirectly associated via their common mediator. Indirect associations are interesting patterns because they represent the diverse interests of Web users who share a similar traversal path. These patterns are not easily found using existing data mining techniques unless the groups of users are known a priori. The effectiveness of indirect association is demonstrated using Web data from an academic institution and an online Web store.

Clustering Earth Science Data: Goals, Issues and Results

Abstract: * This work was partially supported by NASA grant # NCC 2 1231 and by Army High Performance Computing Research Center contract number DAAH04-95-C-0008. The content of this work does not necessarily reflect the position or policy of the government and no official endorsement should be inferred. Access to computing facilities was provided by AHPCRC and the Minnesota Supercomputing Institute. ABSTRACT This paper reports on recent work applying data mining to the task of finding interesting patterns in earth science data derived from global observing satellites, terrestrial observations, and ecosystem models. Patterns are “interesting” if ecosystem scientists can use them to better understand and predict changes in the global carbon cycle and climate system. The initial goal of the work reported here (which is only part of the overall project) is to use clustering to divide the land and ocean areas of the earth into disjoint regions in an automatic, but meaningful, way that enables the direct or indirect discovery of interesting patterns. Finding “meaningful” clusters requires an approach that is aware of various issues related to the spatial and temporal nature of earth science data: the “proper” measure of similarity between time series, removing seasonality from the data to allow detection of non-seasonal patterns, and the presence of spatial and temporal autocorrelation (i.e., measured values that are close in time and space tend to be highly correlated, or similar). While we have techniques to handle some of these spatiotemporal issues (e.g., removing seasonality) and some issues are not a problem (e.g., spatial autocorrelation actually helps our clustering), other issues require more study (e.g., temporal autocorrelation and its effect on time series similarity). Nonetheless, by using the Kmeans as our clustering algorithm and taking linear correlation as our measure of similarity between time series, we have been able to find some interesting ecosystem patterns, including some that are well known to earth scientists and some that require further investigation.