Modeling events in time using cascades of Poisson processes

From WikiLit
Revision as of 20:29, January 30, 2014 by Fnielsen (Talk | contribs) (Text replace - "|collected_datatype=" to "|data_source=")

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search
Publication (help)
Modeling events in time using cascades of Poisson processes
Authors: Aleksandr Simma [edit item]
Citation: University of California, Berkeley  : . 2010.
Publication type: Thesis
Peer-reviewed: Yes
Database(s):
DOI: Define doi.
Google Scholar cites: Citations
Link(s): Paper link
Added by Wikilit team: Added on initial load
Search
Article: Google Scholar BASE PubMed
Other scholarly wikis: AcaWiki Brede Wiki WikiPapers
Web search: Bing Google Yahoo!Google PDF
Other:
Services
Format: BibTeX
Modeling events in time using cascades of Poisson processes is a publication by Aleksandr Simma.


[edit] Abstract

For many applications, the data of interest can be best thought of as events--entities that occur at a particular moment in time, have features and may in turn trigger the occurrence of other events. This thesis presents techniques for modeling the temporal dynamics of events by making each event induce an inhomogeneous Poisson process of others following it. The collection of all events observed is taken to be a draw from the superposition of the induced Poisson processes, as well as a baseline process for some of the initial triggers. The magnitude and shape of the induced Poisson processes controls the number, timing and features of the triggered events. We provide techniques for parameterizing these processes and present efficient, scalable techniques for inference. The framework is then applied to three different domains that demonstrate the power of the approach. First, we consider the problem of identifying dependencies in a computer network through passive observation and provide a technique based on hypothesis testing for accurately discovering interactions between machines. Then, we look at the relationships between Twitter messages about stocks, using the application as a test-bed to experiment with different parameterizations of induced processes. Finally, we apply these tools to build a model of the revision history of Wikipedia, identifying how the community propagates edits from a page to its neighbors and demonstrating the scalability of our approach to very large datasets.

[edit] Research questions

"This thesis presents techniques for modeling the temporal dynamics of events by making each event induce an inhomogeneous Poisson process of others following it. The collection of all events observed is taken to be a draw from the superposition of the induced Poisson processes, as well as a baseline process for some of the initial triggers. The magnitude and shape of the induced Poisson processes controls the number, timing and features of the triggered events. We provide techniques for parameterizing these processes and present efficient, scalable techniques for inference. we apply these tools to build a model of the revision history of Wikipedia, identifying how the community propagates edits from a page to its neighbors and demonstrating the scalability of our approach to very large datasets."

Research details

Topics: Other information retrieval topics [edit item]
Domains: Computer science [edit item]
Theory type: Design and action [edit item]
Wikipedia coverage: Sample data [edit item]
Theories: "This thesis has employed poisson models and cox processes for modeling events" [edit item]
Research design: Experiment, Mathematical modeling [edit item]
Data source: Experiment responses, Wikipedia pages [edit item]
Collected data time dimension: Cross-sectional [edit item]
Unit of analysis: Edit [edit item]
Wikipedia data extraction: Live Wikipedia [edit item]
Wikipedia page type: Article [edit item]
Wikipedia language: Not specified [edit item]

[edit] Conclusion

"The model presented in this work is conceptually relatively simple but flexible, due to the wide range of delay, transition and fertility functions that can be used within the framework. It captures a key aspect of event data – one event may trigger other ones – that has been insufficiently addressed by existing tools and so, has applications in various real-life situations. The resulting models can be used to extract dependencies and properties of a process’ dynamics. Furthermore, the approaches’ scaling properties allow the analysis of very large amounts of data, enabling the analysis of massive datasets."

[edit] Comments


Further notes[edit]

Facts about "Modeling events in time using cascades of Poisson processes"RDF feed
AbstractFor many applications, the data of interesFor many applications, the data of interest can be best thought of as events--entities that occur at a particular moment in time, have features and may in turn trigger the occurrence of other events. This thesis presents techniques for modeling the temporal dynamics of events by making each event induce an inhomogeneous Poisson process of others following it. The collection of all events observed is taken to be a draw from the superposition of the induced Poisson processes, as well as a baseline process for some of the initial triggers. The magnitude and shape of the induced Poisson processes controls the number, timing and features of the triggered events. We provide techniques for parameterizing these processes and present efficient, scalable techniques for inference. The framework is then applied to three different domains that demonstrate the power of the approach. First, we consider the problem of identifying dependencies in a computer network through passive observation and provide a technique based on hypothesis testing for accurately discovering interactions between machines. Then, we look at the relationships between Twitter messages about stocks, using the application as a test-bed to experiment with different parameterizations of induced processes. Finally, we apply these tools to build a model of the revision history of Wikipedia, identifying how the community propagates edits from a page to its neighbors and demonstrating the scalability of our approach to very large datasets.ty of our approach to very large datasets.
Added by wikilit teamAdded on initial load +
Collected data time dimensionCross-sectional +
ConclusionThe model presented in this work is concepThe model presented in this work is conceptually relatively simple but flexible, due to the wide range of delay, transition and fertility functions that can be used within the framework. It captures a key aspect of event data – one event may trigger other ones – that has been insufficiently addressed by existing tools and so, has applications in various real-life situations. The resulting models can be used to extract dependencies and properties of a process’ dynamics. Furthermore, the approaches’ scaling properties allow the analysis of very large amounts of data, enabling the analysis of massive datasets.enabling the analysis of massive datasets.
Data sourceExperiment responses + and Wikipedia pages +
Google scholar urlhttp://scholar.google.com/scholar?ie=UTF-8&q=%22Modeling%2Bevents%2Bin%2Btime%2Busing%2Bcascades%2Bof%2BPoisson%2Bprocesses%22 +
Has authorAleksandr Simma +
Has domainComputer science +
Has topicOther information retrieval topics +
Peer reviewedYes +
Publication typeThesis +
Published inUniversity of California, Berkeley +
Research designExperiment + and Mathematical modeling +
Research questionsThis thesis presents techniques for modeliThis thesis presents techniques for modeling the temporal dynamics of events by making each event induce an inhomogeneous Poisson process of others following it. The collection of all events observed is taken to be a draw from the superposition of the induced Poisson processes, as well as a baseline process for some of the initial triggers. The magnitude and shape of the induced Poisson processes controls the number, timing and features of the triggered events. We provide techniques for parameterizing these processes and present efficient, scalable techniques for inference. we apply these tools to build a model of the revision history of Wikipedia, identifying how the community propagates edits from a page to its neighbors and demonstrating the scalability of our approach to very large datasets.ty of our approach to very large datasets.
Revid10,875 +
TheoriesThis thesis has employed poisson models and cox processes for modeling events
Theory typeDesign and action +
TitleModeling events in time using cascades of Poisson processes
Unit of analysisEdit +
Urlhttp://proquest.umi.com/pqdweb?did=2128789941&Fmt=7&clientId=10306&RQT=309&VName=PQD +
Wikipedia coverageSample data +
Wikipedia data extractionLive Wikipedia +
Wikipedia languageNot specified +
Wikipedia page typeArticle +
Year2010 +