Hard and soft data fusion for joint tracking and classification/intent- detection

Rafael C. Nunez, Buddhika Samarakoon, Kamal Premaratne, Manohar Murthi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

In many application scenarios such as target tracking., one is confronted with the problem of suitably fusing both hard data (e.g., radar, acoustic) and soft data (e.g., text from blogs, or witness statements). Complicating this challenge are the imperfections inherent within soft text data which possess semantic uncertainty, in addition to uncertainty regarding the source reliability and statement credibility. While natural language processing can convert text into uncertainty-augmented first order logic, the question of how to incorporate such soft data models into estimation and tracking remains an important fundamental issue. In this paper, we develop a Bayesian filtering and tracking framework for incorporating both hard and soft data, demonstrating how the probability posterior can be decomposed into a product of combining functions over subsets of the state and measurement variables. This combining function approach offers a framework for the development and incorporation of more sophisticated uncertainty modeling and tracking/estimation models.

Original languageEnglish
Title of host publicationProceedings of the 16th International Conference on Information Fusion, FUSION 2013
Pages661-668
Number of pages8
StatePublished - Dec 26 2013
Event16th International Conference of Information Fusion, FUSION 2013 - Istanbul, Turkey
Duration: Jul 9 2013Jul 12 2013

Other

Other16th International Conference of Information Fusion, FUSION 2013
CountryTurkey
CityIstanbul
Period7/9/137/12/13

Fingerprint

Data fusion
Blogs
Set theory
Target tracking
Data structures
Radar
Acoustics
Semantics
Defects
Uncertainty
Processing

ASJC Scopus subject areas

  • Information Systems

Cite this

Nunez, R. C., Samarakoon, B., Premaratne, K., & Murthi, M. (2013). Hard and soft data fusion for joint tracking and classification/intent- detection. In Proceedings of the 16th International Conference on Information Fusion, FUSION 2013 (pp. 661-668). [6641344]

Hard and soft data fusion for joint tracking and classification/intent- detection. / Nunez, Rafael C.; Samarakoon, Buddhika; Premaratne, Kamal; Murthi, Manohar.

Proceedings of the 16th International Conference on Information Fusion, FUSION 2013. 2013. p. 661-668 6641344.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Nunez, RC, Samarakoon, B, Premaratne, K & Murthi, M 2013, Hard and soft data fusion for joint tracking and classification/intent- detection. in Proceedings of the 16th International Conference on Information Fusion, FUSION 2013., 6641344, pp. 661-668, 16th International Conference of Information Fusion, FUSION 2013, Istanbul, Turkey, 7/9/13.
Nunez RC, Samarakoon B, Premaratne K, Murthi M. Hard and soft data fusion for joint tracking and classification/intent- detection. In Proceedings of the 16th International Conference on Information Fusion, FUSION 2013. 2013. p. 661-668. 6641344
Nunez, Rafael C. ; Samarakoon, Buddhika ; Premaratne, Kamal ; Murthi, Manohar. / Hard and soft data fusion for joint tracking and classification/intent- detection. Proceedings of the 16th International Conference on Information Fusion, FUSION 2013. 2013. pp. 661-668
@inproceedings{cb1e23b917fa499a931ef7ff822a38da,
title = "Hard and soft data fusion for joint tracking and classification/intent- detection",
abstract = "In many application scenarios such as target tracking., one is confronted with the problem of suitably fusing both hard data (e.g., radar, acoustic) and soft data (e.g., text from blogs, or witness statements). Complicating this challenge are the imperfections inherent within soft text data which possess semantic uncertainty, in addition to uncertainty regarding the source reliability and statement credibility. While natural language processing can convert text into uncertainty-augmented first order logic, the question of how to incorporate such soft data models into estimation and tracking remains an important fundamental issue. In this paper, we develop a Bayesian filtering and tracking framework for incorporating both hard and soft data, demonstrating how the probability posterior can be decomposed into a product of combining functions over subsets of the state and measurement variables. This combining function approach offers a framework for the development and incorporation of more sophisticated uncertainty modeling and tracking/estimation models.",
author = "Nunez, {Rafael C.} and Buddhika Samarakoon and Kamal Premaratne and Manohar Murthi",
year = "2013",
month = "12",
day = "26",
language = "English",
isbn = "9786058631113",
pages = "661--668",
booktitle = "Proceedings of the 16th International Conference on Information Fusion, FUSION 2013",

}

TY - GEN

T1 - Hard and soft data fusion for joint tracking and classification/intent- detection

AU - Nunez, Rafael C.

AU - Samarakoon, Buddhika

AU - Premaratne, Kamal

AU - Murthi, Manohar

PY - 2013/12/26

Y1 - 2013/12/26

N2 - In many application scenarios such as target tracking., one is confronted with the problem of suitably fusing both hard data (e.g., radar, acoustic) and soft data (e.g., text from blogs, or witness statements). Complicating this challenge are the imperfections inherent within soft text data which possess semantic uncertainty, in addition to uncertainty regarding the source reliability and statement credibility. While natural language processing can convert text into uncertainty-augmented first order logic, the question of how to incorporate such soft data models into estimation and tracking remains an important fundamental issue. In this paper, we develop a Bayesian filtering and tracking framework for incorporating both hard and soft data, demonstrating how the probability posterior can be decomposed into a product of combining functions over subsets of the state and measurement variables. This combining function approach offers a framework for the development and incorporation of more sophisticated uncertainty modeling and tracking/estimation models.

AB - In many application scenarios such as target tracking., one is confronted with the problem of suitably fusing both hard data (e.g., radar, acoustic) and soft data (e.g., text from blogs, or witness statements). Complicating this challenge are the imperfections inherent within soft text data which possess semantic uncertainty, in addition to uncertainty regarding the source reliability and statement credibility. While natural language processing can convert text into uncertainty-augmented first order logic, the question of how to incorporate such soft data models into estimation and tracking remains an important fundamental issue. In this paper, we develop a Bayesian filtering and tracking framework for incorporating both hard and soft data, demonstrating how the probability posterior can be decomposed into a product of combining functions over subsets of the state and measurement variables. This combining function approach offers a framework for the development and incorporation of more sophisticated uncertainty modeling and tracking/estimation models.

UR - http://www.scopus.com/inward/record.url?scp=84890821427&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84890821427&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9786058631113

SP - 661

EP - 668

BT - Proceedings of the 16th International Conference on Information Fusion, FUSION 2013

ER -