Journal of Applied Information Science

Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

Granular Region-Oriented Fuzzy-Rough Based KNN Improvization for Activity Recognition Modeling


Affiliations
1 Department of Information Technology and Engineering, Goa University, Goa, India
     

   Subscribe/Renew Journal


Activity recognition is a complex task of the Human Computer Interaction (HCI) domain with ever-increasing research interest. Human activity recognition has been specially addressed by the advances in pattern recognition. k-Nearest Neighbors (kNN) is a non-parametric classifier from pattern recognition theory, that mimics human decision making by taking previous experiences into consideration for segregating unknown objects. A novel fuzzy-rough model, based on granular computing for improvization of the kNN classifier is proposed herewith. In this model, feature-wise fuzzy memberships are generated to fuzzify the feature space of the nearest neighbours of the test object. These neighbor's fuzzified feature space are then aggregated into granules, based on their class-belongingness. From these, lower and upper approximation granules are generated using rough set theory to classify the test object. It is shown experimentally that this model outperforms the traditional kNN by 16.43% and Fuzzy-kNN by 10.25%, in the human activity recognition domain. Another novelty is in the efficient use of the fuzzy similarity relations in class-dependent granulated feature space, and, fuzzy-rough lower/upper approximations in the hybridization of the kNN classifier.

Keywords

K-Nearest Neighbors, Human Activity Recognition, Smart Environments, Pervasive Computing, Fuzzy Rough Sets, Fuzzy-Rough Granular Computing.
Subscription Login to verify subscription
User
Notifications
Font Size


  • Borges, V., & Jeberson, W. (2014). Survey of Context Information Fusion for Sensor Networks Based Ubiquitous Systems, Computer Science and Information Technology 2(3) 165–178. doi:10.13189/ csit.2014.020306. URL http://www.hrpub.org/journals/ jour archive.php?id=35

  • Wimo, A., Winblad, B., Jo¨nsson, L. (2010). The worldwide societal costs of dementia: Estimates for 2009, Alzheimer’s & Dementia 6(2), 98-103.

  • Borges, V., & Jeberson, W. Modeling Activity Recognition Using Text Categorization Paradigm, Elsevier Expert System with Applications (under consideration).

  • Bargiela, A., & Pedrycz, W. (2003). Granular computing: An introduction, Springer.

  • Pedrycz, W., Skowron, A., & Kreinovich, V. (2008). Handbook of granular computing. John Wiley & Sons.

  • Pal, S. K., & Dutta-Majumder, D. K. (1986). Fuzzy mathematical ap proach to pattern recognition. Halsted Press.

  • Bottou, L., & Vapnik, V. (1992). Local learning algorithms. Neural Computation, 4(6), 888-900.Z.

  • Pawlak, Imprecise Categories, Approximations and Rough Sets, Springer, 1991.

  • Pawlak, Z. (1991). Imprecise categories. Approximations and rough sets. Springer.

  • Mundici, D. (2011). Advanced Lukasiewicz calculus and MV-algebras. Springer.

  • Keller, J. M., Gray, M. R., & Givens, J. A. (1985). A fuzzy k-nearest neighbor algorithm. IEEE Transactions on Systems, Man and Cybernetics, (4), 580-585.

  • Sarkar, M. (2007). Fuzzy-rough nearest neighbors algorithm.Fuzzy Sets and Systems, 158, 2123-2152.

  • Shen, H. B., Yang, Y., Chou, K. C. (2006). Fuzzy KNN for pre dicting membrane protein types from pseudoamino acid composition. Journal of Theoretical Biology, 240(1), 9-13.

  • Jensen, R., & Cornelis, C. (2011). Fuzzy-rough nearest neighbour classification and prediction. Theoretical Computer Science, 5871-5884.

  • WSU, CASAS Project (2014). Retrieved from http:// ailab.wsu.edu/casas/Borges, V., & Jeberson, W. Fortune at the bottom of the Classifier Pyramid: A Novel approach to Human Activity Recognition. Elsevier Procedia Computer Science.

  • Borges, V., & Jeberson, W. (2014). Modification of fuzzy kNN by learning from data in activity recognition modeling. International Conference on Mathematics and Computational Sciences (In Press), Inderscience Publishers.

  • Pal, S. K., & Mitra, P. (2004). Case generation using rough sets with fuzzy representation. IEEE Transactions on Knowledge and Data Engineering, 16(3), 293-300.

  • Li, Y., Shiu, S. C. K., & Pal, S. K. (2006). Combining feature re duction and case selection in building CBR classi fiers. IEEE Transactions on Knowledge and Data Engineering, 18(3), 415-429.

  • Li, Y., Shiu, S. C. K., Pal, S. K., & Liu, J. N. K. (2006).A rough set-based case-based reasoner for text categorization.International Journal of Approximate Reasoning, 41(2), 229-255.


Abstract Views: 264

PDF Views: 0




  • Granular Region-Oriented Fuzzy-Rough Based KNN Improvization for Activity Recognition Modeling

Abstract Views: 264  |  PDF Views: 0

Authors

Vijay Borges
Department of Information Technology and Engineering, Goa University, Goa, India
Wilson Jeberson
Department of Information Technology and Engineering, Goa University, Goa, India

Abstract


Activity recognition is a complex task of the Human Computer Interaction (HCI) domain with ever-increasing research interest. Human activity recognition has been specially addressed by the advances in pattern recognition. k-Nearest Neighbors (kNN) is a non-parametric classifier from pattern recognition theory, that mimics human decision making by taking previous experiences into consideration for segregating unknown objects. A novel fuzzy-rough model, based on granular computing for improvization of the kNN classifier is proposed herewith. In this model, feature-wise fuzzy memberships are generated to fuzzify the feature space of the nearest neighbours of the test object. These neighbor's fuzzified feature space are then aggregated into granules, based on their class-belongingness. From these, lower and upper approximation granules are generated using rough set theory to classify the test object. It is shown experimentally that this model outperforms the traditional kNN by 16.43% and Fuzzy-kNN by 10.25%, in the human activity recognition domain. Another novelty is in the efficient use of the fuzzy similarity relations in class-dependent granulated feature space, and, fuzzy-rough lower/upper approximations in the hybridization of the kNN classifier.

Keywords


K-Nearest Neighbors, Human Activity Recognition, Smart Environments, Pervasive Computing, Fuzzy Rough Sets, Fuzzy-Rough Granular Computing.

References