Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
26
result(s) for
"Riesen, Kaspar"
Sort by:
Graph classification and clustering based on vector space embedding
This book is concerned with a fundamentally novel approach to graph-based pattern recognition based on vector space embedding of graphs. It aims at condensing the high representational power of graphs into a computationally efficient and mathematically convenient feature vector.
eBook
Graph Classification And Clustering Based On Vector Space Embedding
This book is concerned with a fundamentally novel approach to graph-based pattern recognition based on vector space embedding of graphs. It aims at condensing the high representational power of graphs into a computationally efficient and mathematically convenient feature vector.This volume utilizes the dissimilarity space representation originally proposed by Duin and Pekalska to embed graphs in real vector spaces. Such an embedding gives one access to all algorithms developed in the past for feature vectors, which has been the predominant representation formalism in pattern recognition and related areas for a long time.
eBook
On the Impact of Using Utilities Rather than Costs for Graph Matching
The concept of graph edit distance constitutes one of the most flexible graph matching paradigms available. The major drawback of graph edit distance, viz. the exponential time complexity, has been recently overcome by means of a reformulation of the edit distance problem to a linear sum assignment problem. However, the substantial speed up of the matching is also accompanied by an approximation error on the distances. Major contribution of this paper is the introduction of a transformation process in order to convert the underlying cost model into a utility model. The benefit of this transformation is that it enables the integration of additional information in the assignment process. We empirically confirm the positive effects of this transformation on five benchmark graph sets with respect to the accuracy and run time of a distance based classifier.
Journal Article
A comparative study of pattern recognition algorithms for predicting the inpatient mortality risk using routine laboratory measurements
Laboratory tests are a common and relatively cheap way to assess the general health status of patients. Various publications showed the potential of laboratory measurements for predicting inpatient mortality using statistical methodologies. However, these efforts are basically limited to the use of logistic regression models. In the present paper we use anonymized data from about 40,000 inpatient admissions to the Inselspital in Bern (Switzerland) to evaluate the potential of powerful pattern recognition algorithms employed for this particular risk prediction. In addition to the age and sex of the inpatients, a set of 33 laboratory measurements, frequently available at the Inselspital, are used as basic variables. In a large empirical evaluation we demonstrate that recent pattern recognition algorithms (such as random forests, gradient boosted trees or neural networks) outperform the more traditional approaches based on logistic regression. Moreover, we show how the predictions of the pattern recognition algorithms, which cannot be directly interpreted in general, can be calibrated to output a meaningful probabilistic risk score.
Journal Article
Graph-Based Offline Signature Verification
Graphs provide a powerful representation formalism that offers great promise to benefit tasks like handwritten signature verification. While most state-of-the-art approaches to signature verification rely on fixed-size representations, graphs are flexible in size and allow modeling local features as well as the global structure of the handwriting. In this article, we present two recent graph-based approaches to offline signature verification: keypoint graphs with approximated graph edit distance and inkball models. We provide a comprehensive description of the methods, propose improvements both in terms of computational time and accuracy, and report experimental results for four benchmark datasets. The proposed methods achieve top results for several benchmarks, highlighting the potential of graph-based signature verification.
Paper