Data classification using the Dempster–Shafer method.

Journal article


Chen, Qi, Whitbrook, Amanda, Aickelin, Uwe and Roadknight, Chris 2014. Data classification using the Dempster–Shafer method. Journal of Experimental & Theoretical Artificial Intelligence. https://doi.org/10.1080/0952813X.2014.886301
AuthorsChen, Qi, Whitbrook, Amanda, Aickelin, Uwe and Roadknight, Chris
Abstract

In this paper, the Dempster–Shafer (D–S) method is used as the theoretical basis for creating data classification systems. Testing is carried out using three popular multiple attribute benchmark data-sets that have two, three and four classes. In each case, a subset of the available data is used for training to establish thresholds, limits or likelihoods of class membership for each attribute, and hence create mass functions that establish probability of class membership for each attribute of the test data. Classification of each data item is achieved by combination of these probabilities via Dempster’s rule of combination. Results for the first two data-sets show extremely high classification accuracy that is competitive with other popular methods. The third data-set is non-numerical and difficult to classify, but good results can be achieved provided the system and mass functions are designed carefully and the right attributes are chosen for combination. In all cases, the D–S method provides comparable performance to other more popular algorithms, but the overhead of generating accurate mass functions increases the complexity with the addition of new attributes. Overall, the results suggest that the D–S approach provides a suitable framework for the design of classification systems and that automating the mass function design and calculation would increase the viability of the algorithm for complex classification problems.

In this paper, the Dempster–Shafer (D–S) method is used as the theoretical basis for
creating data classification systems. Testing is carried out using three popular multiple
attribute benchmark data-sets that have two, three and four classes. In each case, a
subset of the available data is used for training to establish thresholds, limits or
likelihoods of class membership for each attribute, and hence create mass functions
that establish probability of class membership for each attribute of the test data.
Classification of each data item is achieved by combination of these probabilities
via Dempster’s rule of combination. Results for the first two data-sets show extremely
high classification accuracy that is competitive with other popular methods. The third
data-set is non-numerical and difficult to classify, but good results can be achieved
provided the system and mass functions are designed carefully and the right attributes
are chosen for combination. In all cases, the D–S method provides comparable
performance to other more popular algorithms, but the overhead of generating
accurate mass functions increases the complexity with the addition of new attributes.
Overall, the results suggest that the D–S approach provides a suitable framework for
the design of classification systems and that automating the mass function design and
calculation would increase the viability of the algorithm for complex classification
problems.

KeywordsDempster–Shafer theory; Data classification
Year2014
JournalJournal of Experimental & Theoretical Artificial Intelligence
PublisherTaylor & Francis
ISSN0952813X
13623079
Digital Object Identifier (DOI)https://doi.org/10.1080/0952813X.2014.886301
Web address (URL)http://hdl.handle.net/10545/622385
hdl:10545/622385
Publication dates26 Feb 2014
Publication process dates
Deposited19 Mar 2018, 16:20
Rights

Archived with thanks to Journal of Experimental & Theoretical Artificial Intelligence

ContributorsUniversity of Nottingham
File
File Access Level
Open
Permalink -

https://repository.derby.ac.uk/item/94578/data-classification-using-the-dempster-shafer-method

Download files

  • 61
    total views
  • 0
    total downloads
  • 1
    views this month
  • 0
    downloads this month

Export as

Related outputs

Addressing robustness in time-critical, distributed, task allocation algorithms.
Whitbrook, Amanda, Meng, Qinggang and Chung, Paul W. H. 2018. Addressing robustness in time-critical, distributed, task allocation algorithms. Applied Intelligence. https://doi.org/10.1007/s10489-018-1169-3
A novel distributed scheduling algorithm for time-critical multi-agent systems.
Whitbrook, Amanda, Meng, Qinggang and Chung, Paul W. H. 2015. A novel distributed scheduling algorithm for time-critical multi-agent systems. IEEE. https://doi.org/10.1109/IROS.2015.7354299
Juxtaposition of system dynamics and agent-based simulation for a case study in immunosenescence.
Figueredo, Grazziela P., Siebers, Peer-Olaf, Aickelin, Uwe, Whitbrook, Amanda and Garibaldi, Jonathan M. 2015. Juxtaposition of system dynamics and agent-based simulation for a case study in immunosenescence. PLos ONE. https://doi.org/10.1371/journal.pone.0118359
A conceptual framework for combining artificial neural networks with computational aeroacoustics for design development.
McKee, Claire, Harmanto, Dani and Whitbrook, Amanda 2018. A conceptual framework for combining artificial neural networks with computational aeroacoustics for design development. Industrial Engineering and Operations Management Society (IEOM).
A cloud-based path-finding framework: Improving the performance of real-time navigation in games
Rowe, Jordan, Whitbrook, Amanda and Chen, Minsi 2017. A cloud-based path-finding framework: Improving the performance of real-time navigation in games. Association of Computing Machinery. https://doi.org/10.1145/3147234.3148097
Model building
Lowdnes, Val, Berry, Stuart, Trovati, Marcello and Whitbrook, Amanda 2017. Model building. in: Springer.
Distributed task rescheduling with time constraints for the optimization of total task allocations in a multirobot system
Turner, Joanna, Meng, Qinggang, Schaefer, Gerald, Whitbrook, Amanda and Soltoggio, Andrea 2017. Distributed task rescheduling with time constraints for the optimization of total task allocations in a multirobot system. IEEE Transactions on Cybernetics. https://doi.org/10.1109/TCYB.2017.2743164
A robust, distributed task allocation algorithm for time-critical, multi agent systems operating in uncertain environments
Whitbrook, Amanda, Meng, Qinggang and Chung, Paul W. H. 2017. A robust, distributed task allocation algorithm for time-critical, multi agent systems operating in uncertain environments.
Reliable, distributed scheduling and rescheduling for time-critical, multiagent systems
Whitbrook, Amanda, Meng, Qinggang and Chung, Paul W. H. 2017. Reliable, distributed scheduling and rescheduling for time-critical, multiagent systems. IEEE Transactions on Automation Science and Engineering. https://doi.org/10.1109/TASE.2017.2679278