A Bayesian Committee Machine
... There the question is posed, given the kernel function bj (x), is there a fixed basis function expansion which corresponds to these kernel functions. One answer is given by Mercer’s theorem which states that for symmetric and positive definite kernels a corresponding set of fixed basis function exis ...
... There the question is posed, given the kernel function bj (x), is there a fixed basis function expansion which corresponds to these kernel functions. One answer is given by Mercer’s theorem which states that for symmetric and positive definite kernels a corresponding set of fixed basis function exis ...
Evaluating a clustering solution: An application in the tourism market
... VFDT’s average memory allocation over the course of the run was 23MB while CVFDT’s was 16.5MB The average number of nodes in VFDT’s tree was 2696 and in CVFDT’s tree was 677(132: alternate tree, 545: main tree) ...
... VFDT’s average memory allocation over the course of the run was 23MB while CVFDT’s was 16.5MB The average number of nodes in VFDT’s tree was 2696 and in CVFDT’s tree was 677(132: alternate tree, 545: main tree) ...
a clustering-based approach for enriching trajectories with
... Trajectory data, representing movement data or mobility data, is usually generated as sequences of id; x; y; t points through mobile devices (Bogorny et al., 2011). This data is required to be processed into more human-perceptible structures in order to facilitate further analysis. In a conceptual m ...
... Trajectory data, representing movement data or mobility data, is usually generated as sequences of id; x; y; t points through mobile devices (Bogorny et al., 2011). This data is required to be processed into more human-perceptible structures in order to facilitate further analysis. In a conceptual m ...
Data Mining and Visualization of Android Usage Data
... (Xu et al., 2010) to a Usage Mining context to discover the hidden set of important tasks (use cases4 ) from the World Wide Web (Web) usage. In this dissertation the LDA algorithm was adapted to the context of an Android application usage following the principles described in (Xu et al., 2010). To h ...
... (Xu et al., 2010) to a Usage Mining context to discover the hidden set of important tasks (use cases4 ) from the World Wide Web (Web) usage. In this dissertation the LDA algorithm was adapted to the context of an Android application usage following the principles described in (Xu et al., 2010). To h ...
An Explorative Parameter Sweep: Spatial-temporal Data
... does the localization and copy number of species change through time) features from individual simulations. This will be a challenge because of the high dimensionality associated with the simulations output. For the purpose of extracting features, it is not a straightforward task to analyze time ser ...
... does the localization and copy number of species change through time) features from individual simulations. This will be a challenge because of the high dimensionality associated with the simulations output. For the purpose of extracting features, it is not a straightforward task to analyze time ser ...
SRDA: An Efficient Algorithm for Large-Scale
... which can be solved by the GSVD algorithm. One limitation of this method is the high computational cost of GSVD, especially for large and high-dimensional data sets. In [25], Ye extended such approach by solving the optimization problem using simultaneous diagonalization of the scatter matrices. Ano ...
... which can be solved by the GSVD algorithm. One limitation of this method is the high computational cost of GSVD, especially for large and high-dimensional data sets. In [25], Ye extended such approach by solving the optimization problem using simultaneous diagonalization of the scatter matrices. Ano ...
Document clustering using character N
... Density-based methods can build clusters of arbitrary shape and filter out outliers as noise. One of the widely used clustering methods is K-Means [18], a partitioning clustering method. Given a cluster number k, K-Means randomly selects k positions in the same vector space as the objet representati ...
... Density-based methods can build clusters of arbitrary shape and filter out outliers as noise. One of the widely used clustering methods is K-Means [18], a partitioning clustering method. Given a cluster number k, K-Means randomly selects k positions in the same vector space as the objet representati ...
![[PDF]](http://s1.studyres.com/store/data/008775338_1-f1f4c50506d1a377a17adee51c563247-300x300.png)
[PDF]
... Even though several algorithms are available in the literature for association rule mining, [12-20] a good number of them deal with efficient implementations rather than the production of effective rules [11, 16, 18]. The techniques that aid in the extraction of suitable and genuine association patt ...
... Even though several algorithms are available in the literature for association rule mining, [12-20] a good number of them deal with efficient implementations rather than the production of effective rules [11, 16, 18]. The techniques that aid in the extraction of suitable and genuine association patt ...
BORDER: Efficient Computation of Boundary Points
... Utilizing RkNN in data mining tasks will require the execution of a RkNN query for each point in the dataset (the set-oriented RkNN query). However, this is very expensive and the complexity will be O(N 3 ) since the complexity of a single RkNN query is O(N 2 ) time using sequential scan for non-ind ...
... Utilizing RkNN in data mining tasks will require the execution of a RkNN query for each point in the dataset (the set-oriented RkNN query). However, this is very expensive and the complexity will be O(N 3 ) since the complexity of a single RkNN query is O(N 2 ) time using sequential scan for non-ind ...
Evolving Temporal Association Rules with Genetic Algorithms
... mining quantitative association rules since these are present in many real-world applications. These are different from boolean association rules because they include a quantitative value describing the amount of each item. A method of mining quantitative data requires the values to be discretised i ...
... mining quantitative association rules since these are present in many real-world applications. These are different from boolean association rules because they include a quantitative value describing the amount of each item. A method of mining quantitative data requires the values to be discretised i ...
Cluster Analysis: Basic Concepts and Methods
... each object must belong to exactly one group. This requirement may be relaxed, for example, in fuzzy partitioning techniques. References to such techniques are given in the bibliographic notes (Section 10.9). Most partitioning methods are distance-based. Given k, the number of partitions to construc ...
... each object must belong to exactly one group. This requirement may be relaxed, for example, in fuzzy partitioning techniques. References to such techniques are given in the bibliographic notes (Section 10.9). Most partitioning methods are distance-based. Given k, the number of partitions to construc ...
Hierarchical Convex NMF for Clustering Massive Data
... thereby select k convex hull data points for solving Eq. (4). Typically, I results in unary representations. If this is not the case, we simply map SI to their nearest neighboring data point in S. Given X, the computation of the coefficients H is straight forward. For smaller datasets it is possible ...
... thereby select k convex hull data points for solving Eq. (4). Typically, I results in unary representations. If this is not the case, we simply map SI to their nearest neighboring data point in S. Given X, the computation of the coefficients H is straight forward. For smaller datasets it is possible ...