Local Exceptionality Detection on Social Interaction Networks

@inproceedings{Atzmller2016LocalED,
  title={Local Exceptionality Detection on Social Interaction Networks},
  author={Martin Atzm{\"u}ller},
  booktitle={ECML/PKDD},
  year={2016}
}
  • M. Atzmüller
  • Published in ECML/PKDD 19 September 2016
  • Computer Science
Local exceptionality detection on social interaction networks includes the analysis of resources created by humans (e. g., social media) as well as those generated by sensor devices in the context of (complex) interactions. This paper provides a structured overview on a line of work comprising a set of papers that focus on data-driven exploration and modeling in the context of social network analysis, community detection and pattern mining. 

Exceptional Model Mining in Ubiquitous and Social Environments

This paper provides a structured overview on a line of work comprising a set of papers that focus on data-driven exploration and modeling using exceptional model mining in ubiquitous and social environments.

Mining Exceptional Social Behaviour

The proposed method combines Subgroup Discovery and Network Science techniques for finding social behaviour that deviates from the norm and transforms movement and demographic data into attributed social interaction networks, and returns descriptive subgroups.

Descriptive Community Detection

  • M. Atzmüller
  • Computer Science
    Formal Concept Analysis of Social Networks
  • 2017
An organized picture of recent research in descriptive community (and subgroup) detection is presented, focusing on attributed graphs, i.e.,complex relational graphs that are annotated with additional information.

Summarizing Graphs at Multiple Scales: New Trends

This tutorial is to give a systematic overview of methods for summarizing and explaining graphs at different scales: the node-group level, the network level, and the multi-network level.

Sequential Modeling and Structural Anomaly Analytics in Industrial Production Environments

This demonstration paper presents an integrated approach for anomaly analytics in an industrial production scenario based on first-order Markov chain models and analyzes sequential trails relative to specific hypotheses in a industrial application context.

Benelearn 2017: Proceedings of the Twenty-Sixth Benelux Conference on Machine Learning, Technische Universiteit Eindhoven, 9-10 June 2017

An overview of the author's research in this fascinating area, including pattern mining, the analysis of influence propagation in social networks, and ethical challenges such as models that discriminate are given.

Anomaly Analytics and Structural Assessment in Process Industries

This work model sequential alarm data for anomaly detection and analysis, based on first-order Markov chain models, and outlines hypothesis-driven and description-oriented modeling and provides an interactive dashboard for exploration and visualization.

Have It Both Ways - From A/B Testing to A&B Testing with Exceptional Model Mining

Exceptional Model Mining is one such advanced data science technique, which specializes in identifying subgroups that behave differently from the overall population, using the association model class for EMM to find subpopulations that prefer variant A where the general population prefers variant B, and vice versa.

Anomaly Detection and Structural Analysis in Industrial Production Environments

Das Erkennen von anormalem Verhalten kann im Kontext industrieller Anwendung von entscheidender Bedeutung sein. Wahrend moderne Produktionsanlagen mit hochentwickelten Alarmsteuerungssytemen

Community Detection and Analysis on Attributed Social Networks

  • M. Atzmüller
  • Computer Science
    Encyclopedia of Social Network Analysis and Mining. 2nd Ed.
  • 2018

References

SHOWING 1-10 OF 24 REFERENCES

Data Mining on Social Interaction Networks

This work considers social interaction networks from a data mining perspective - also with a special focus on real-world face-to-face contact networks - and presents predictive data mining methods, i.e., for localization, recommendation and link prediction.

User-Relatedness and Community Structure in Social Interaction Networks

There are dependencies and correlations between the networks, which allow for drawing reciprocal conclusions concerning pairs of networks, based on the assessment of structural correlations and ranking interchangeability, which can be used for assessing the results of structural analysis techniques, e.g., community mining methods.

Exploratory pattern mining on social media using geo-references and social tagging information

This paper presents exploratory pattern mining techniques for describing communities of resources and for characterising locations of interest using Flickr as an exemplary case study for demonstrating the effectiveness and validity of the interactive approach.

Community Assessment Using Evidence Networks

A set of so-called evidence networks which are capturing typical interactions in social network applications are introduced which are able to apply a rich set of implicit information for the evaluation of communities.

Detecting community patterns capturing exceptional link trails

  • M. Atzmüller
  • Computer Science
    2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)
  • 2016
A new method for detecting descriptive community patterns capturing exceptional (sequential) link trails, model sequential data as first-order Markov chain models, mapped to an attributed weighted network represented as a graph and presents a comprehensive modeling approach.

The social distributional hypothesis: a pragmatic proxy for homophily in online social networks

This work conducts a series of experiments on social interaction networks from Twitter, Flickr, and BibSonomy and investigates the relatedness concerning the interactions, their frequency, and the specific interaction characteristics, supporting the social distributional hypothesis.

Mining social media: key players, sentiments, and communities

  • M. Atzmüller
  • Computer Science
    WIREs Data Mining Knowl. Discov.
  • 2012
This focus article considers mining approaches concerning social media in social networks and organizations and the analysis of such data, and describes the VIKAMINE system for mining communities and subgroups in socialMedia in the sketched application domains.

Temporal evolution of contacts and communities in networks of face-to-face human interactions

This article analyzes the evolution of contacts and communities over time to consider the stability of the respective communities and assess different factors which have an influence on the quality of community prediction.

Big Data Analytics Using Local Exceptionality Detection

This chapter presents the novel SD-MapR algorithmic framework for large-scale local exceptionality detection implemented using subgroup discovery on the Map/Reduce framework and describes the basic algorithm in detail and provides an experimental evaluation using several real-world datasets.

Is web content a good proxy for real-life interaction? A case study considering online and offline interactions of computer scientists

This paper investigates whether relations captured by online social networks can be used as a proxy for the relations in offline social networks, such as networks of human face-to-face (F2F) proximity and coauthorship networks.