Efficient frequent subtree mining beyond forests / Pascal Welke.

Format:

Book

Author/Creator:

Welke, Pascal, author.

Series:

Frontiers in artificial intelligence and applications. Dissertations in artificial intelligence ; Volume 348.

Dissertations in Artificial Intelligence ; Volume 348

Language:

English

Subjects (All):

Data mining.

Physical Description:

1 online resource (190 pages).

Edition:

1st ed.

Place of Publication:

Amsterdam, Netherlands : IOS Press, 2020.

Summary:

A common paradigm in distance-based learning is to embed the instance space into a feature space equipped with a metric and define the dissimilarity between instances by the distance of their images in the feature space.

Contents:

Intro

Title Page

Contents

1 Introduction

1.1 A Motivating Experiment

1.2 Contributions

1.3 Outline

1.4 Previously Published Work

2 Preliminaries

2.1 Notions and Notation

2.2 Frequent Connected Subgraph Mining

2.3 Embedding Computation

2.4 Datasets

3 Related Work

3.1 Algorithms for the SubgraphIsomorphism Problem

3.2 Algorithms for the FCSM Problem

4 Probabilistic Frequent Subtrees

4.1 Mining Probabilistic Frequent Subtrees

4.2 Experimental Evaluation

4.3 Summary

5 Boosted Probabilistic Frequent Subtrees

5.1 An Efficient Embedding Operator for Trees

5.2 Mining Boosted Probabilistic Frequent Subtrees

5.3 Exact Frequent Subtree Mining

5.4 Summary and Open Questions

6 Fast Computation

6.1 Complete Embeddings into Subtree Feature Spaces

6.2 Min-Hashing in Subtree Feature Spaces

6.3 Experimental Evaluation

6.4 Summary and Open Questions

7 Conclusion

7.1 Discussion

7.2 Outlook

A HamiltonianPath for Cactus Graphs

A.1 Three Necessary Conditions

A.2 A Linear Time Algorithm for Cactus Graphs

A.3 Some Statistics for Real-World Datasets

A.4 Summary

B Poissons Binomial Distribution

Bibliography.

Notes:

Description based on print version record.

Description based on publisher supplied metadata and other sources.

ISBN:

1-64368-079-X

OCLC:

1162846576