Training RBF neural networks on unbalanced data
... the overlaps between different classes and the overlaps between clusters of the same class. The overlaps between different classes have been considered in RBF training algorithms. For example, overlappedreceptive fields of different clusters can improve the performance of the RBF classifier when dea ...
... the overlaps between different classes and the overlaps between clusters of the same class. The overlaps between different classes have been considered in RBF training algorithms. For example, overlappedreceptive fields of different clusters can improve the performance of the RBF classifier when dea ...
Outlier Detection for High Dimensional Data
... Many algorithms have been proposed in recent years for outlier detection [7, 8, 10, 22, 23, 25, 26], but they are not methods which are speci cally designed in order to deal with the curse of high dimensionality. The statistics community has studied the concept of outliers quite extensively [8]. In ...
... Many algorithms have been proposed in recent years for outlier detection [7, 8, 10, 22, 23, 25, 26], but they are not methods which are speci cally designed in order to deal with the curse of high dimensionality. The statistics community has studied the concept of outliers quite extensively [8]. In ...
Comparison of information retrieval techniques: Latent
... SVD like in LSI Concept decomposition was introduced in 2001 ...
... SVD like in LSI Concept decomposition was introduced in 2001 ...
Introduction
... Data mining is often associated with quantitative methods but it differs from standard statistical approaches. ...
... Data mining is often associated with quantitative methods but it differs from standard statistical approaches. ...
Retail Marketing Segmentation and Customer Profiling for
... X ⇒ Y is the percentage of transactions in the database that contain X ∪ Y. That is, support (X ⇒ Y ) = P (X ∪ Y ), P is the probability. Definition 3: The confidence or strength ( Φ ) for an association rule (X ⇒ Y) is the ratio of the number of transactions that contain X ∪ Y to the number of tran ...
... X ⇒ Y is the percentage of transactions in the database that contain X ∪ Y. That is, support (X ⇒ Y ) = P (X ∪ Y ), P is the probability. Definition 3: The confidence or strength ( Φ ) for an association rule (X ⇒ Y) is the ratio of the number of transactions that contain X ∪ Y to the number of tran ...
WSARE: What`s Strange About Recent Events
... traditional anomaly detection systems, shortcomings in these systems, which we will illustrate, limit their usefulness in early disease outbreak detection. In our database of emergency department (ED) cases from several hospitals in a city, each record contains information about the individual who w ...
... traditional anomaly detection systems, shortcomings in these systems, which we will illustrate, limit their usefulness in early disease outbreak detection. In our database of emergency department (ED) cases from several hospitals in a city, each record contains information about the individual who w ...
Association Rule Mining using Apriori Algorithm: A Survey
... multiple processors and databases to speed up the execution of data mining and enable data distribution. The main aim of grid computing is to give organizations and application developers the ability to create distributed computing environments that can utilize computing resources on demand. Therefo ...
... multiple processors and databases to speed up the execution of data mining and enable data distribution. The main aim of grid computing is to give organizations and application developers the ability to create distributed computing environments that can utilize computing resources on demand. Therefo ...
Effective framework for prediction of disease outcome using medical
... Cardiac disorders diagnosis is based on SPECT (Single Photon Emission Computed Tomography) images. Bakirci and Yildirim (2004) used feed-forward ANN and achieved an accuracy of 90.04%. Polat et al. (2007c) proposed a method ensemble classifier system based on different feature subsets and AIRS class ...
... Cardiac disorders diagnosis is based on SPECT (Single Photon Emission Computed Tomography) images. Bakirci and Yildirim (2004) used feed-forward ANN and achieved an accuracy of 90.04%. Polat et al. (2007c) proposed a method ensemble classifier system based on different feature subsets and AIRS class ...
View/Download-PDF - International Journal of Computer Science
... instance to a particular class with the aim of achieving least classification error. It is used to extract models that correctly define important data classes within the given dataset. It is a two-step process. In first step the model is created by applying classification algorithm on training data ...
... instance to a particular class with the aim of achieving least classification error. It is used to extract models that correctly define important data classes within the given dataset. It is a two-step process. In first step the model is created by applying classification algorithm on training data ...
Deductive and inductive reasoning on spatio-temporal data
... According to the local interpolation method, although there is not a global function describing the whole trajectory, objects are assumed to move between the observed points following some rule. For instance, a linear interpolation function models a straight movement with constant speed, while other ...
... According to the local interpolation method, although there is not a global function describing the whole trajectory, objects are assumed to move between the observed points following some rule. For instance, a linear interpolation function models a straight movement with constant speed, while other ...
Web Search Result Optimization using Association Rule Mining
... higher than support values. Step 4: In next pass, algorithm creates item sets of three members. Repeat this process until all frequent item sets are accounted. Step 5: These item sets are then used to generate association rules which have threshold values less than or equal to confidence values. Ste ...
... higher than support values. Step 4: In next pass, algorithm creates item sets of three members. Repeat this process until all frequent item sets are accounted. Step 5: These item sets are then used to generate association rules which have threshold values less than or equal to confidence values. Ste ...
Analysis of Hepatitis Dataset using Multirelational Association Rules
... period when measurements were made of the degree of fibrosis for the same patient. To properly analyze the time period involved, the exam date was divided into two attributes: year and month. The period of time considered for the analysis was one month. The exam results of patients with more than on ...
... period when measurements were made of the degree of fibrosis for the same patient. To properly analyze the time period involved, the exam date was divided into two attributes: year and month. The period of time considered for the analysis was one month. The exam results of patients with more than on ...
A Parallel Clustering Method Study Based on MapReduce
... large scale data is an important issue. It is the development intention of big data science. Many scholars have done lots work on this topic. Some clustering methods based on MapReduce were proposed, such as k-means, EM, Dirichlet Process Clustering and so on. Though the clustering method based on I ...
... large scale data is an important issue. It is the development intention of big data science. Many scholars have done lots work on this topic. Some clustering methods based on MapReduce were proposed, such as k-means, EM, Dirichlet Process Clustering and so on. Though the clustering method based on I ...
Ranking Interesting Subspaces for Clustering High Dimensional Data*
... the whole feature space onto a lower-dimensional subspace of relevant attributes, using e.g. principal component analysis (PCA) and singular value decomposition (SVD). However, the transformed attributes often have no intuitive meaning any more and thus the resulting clusters are hard to interpret. ...
... the whole feature space onto a lower-dimensional subspace of relevant attributes, using e.g. principal component analysis (PCA) and singular value decomposition (SVD). However, the transformed attributes often have no intuitive meaning any more and thus the resulting clusters are hard to interpret. ...