|
Graphs are mathematical constructs for representing objects or systems which contain structural (i.e. relationship) information. Graphs have been used in many domains, from software engineering to artificial intelligence. However, the use of graphs in machine learning has been limited, compared to the more prevalent vector model which does not capture structural information. This can be attributed to the fact that until recently we have not had suitable graph-theoretic tools for dealing with graphs, and that comparing graphs for the purpose of evaluating structural matching is often of high time complexity. For these reasons, most machine learning approaches that deal with graph-based data introduce either restrictions on the type or related attributes of the graphs used, or they require a totally new mathematical foundation for handling graphs (such as probability theory). An unfortunate drawback of these types of methods is that existing machine learning algorithms either cannot be applied or require extensive modifications.
In this book, we present methods for utilizing graphs with well-known machine learning algorithms, such as fc-means clustering and fc-nearest neighbors classification, with virtually no significant modifications; the extensions for allowing these algorithms to use graphs is direct and intuitive. Further, we show how we can achieve polynomial time complexity with the restriction that the graphs contain unique node labels. This is the only restriction we place on the graphs, and we can even discard this limitation if we use a sub-optimal approximation approach for determining graph distance or if the graphs are relatively small. |
|
|
 Domain Driven Data MiningIn the present thriving global economy a need has evolved for complex data analysis to enhance an organization’s production systems, decision-making tactics, and performance. In turn, data mining has emerged as one of the most active areas in information technologies. Domain Driven Data Mining offers state-of the-art research and... | | 97 Things Every Programmer Should Know: Collective Wisdom from the ExpertsProgrammers have a lot on their minds . Programming languages, programming techniques, development environments, coding style, tools, development process, deadlines, meetings, software architecture, design patterns, team dynamics, code, requirements, bugs, code quality. And more. A lot. There is an art, craft, and science to programming that... | | Python Testing: Beginner's GuideAutomated testing moves much of the labor of testing off the developer and makes it easier as well as quicker to find bugs and fix them. Automated tests run faster, increase test coverage, and lower costs. However, testing is neither an easy process nor remotely exciting for most developers. But with the right techniques and tools, testing can... |
Managing and Mining Graph Data (Advances in Database Systems)Managing and Mining Graph Data is a comprehensive survey book in graph data analytics. It contains extensive surveys on important graph topics such as graph languages, indexing, clustering, data generation, pattern mining, classification, keyword search, pattern matching, and privacy. It also studies a number of domain-specific... | | Pro Dynamic .NET 4.0 Applications: Data-Driven Programming for the .NET FrameworkAs a developer you are likely painfully aware that not all features of an application can be anticipated when the software ships. In order to cope with these eventualities and save yourself time (and perhaps money), it makes sense to write your applications in such a way that end users can be made as independent of the developers as possible.... | | Algorithms for Next Generation Networks (Computer Communications and Networks)Data networking now plays a major role in everyday life and new applications continue to appear at a blinding pace. Yet we still do not have a sound foundation for designing, evaluating and managing these networks.
This book covers topics at the intersection of algorithms and networking. It builds a complete picture of the current state... |
|
