Show/Hide Menu
Hide/Show Apps
Logout
Türkçe
Türkçe
Search
Search
Login
Login
OpenMETU
OpenMETU
About
About
Open Science Policy
Open Science Policy
Open Access Guideline
Open Access Guideline
Postgraduate Thesis Guideline
Postgraduate Thesis Guideline
Communities & Collections
Communities & Collections
Help
Help
Frequently Asked Questions
Frequently Asked Questions
Guides
Guides
Thesis submission
Thesis submission
MS without thesis term project submission
MS without thesis term project submission
Publication submission with DOI
Publication submission with DOI
Publication submission
Publication submission
Supporting Information
Supporting Information
General Information
General Information
Copyright, Embargo and License
Copyright, Embargo and License
Contact us
Contact us
A Unified Model for Human Behavior Modeling using a Hierarchy with a Variable Number of States
Date
2014-08-28
Author
Alemdar, Hande
Niessen, M. E.
MERENTİTİS, ANDREAS
Ersoy, Cem
Metadata
Show full item record
This work is licensed under a
Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License
.
Item Usage Stats
58
views
0
downloads
Cite This
Human behavior modeling enables many applications for smart cities, smart homes, mobile phones and other domains. We present a hierarchical hidden Markov model for human activity recognition that uses semi-supervised learning to automatically learn the model parameters using only labeled data of the top-layer of the hierarchy. This significantly reduces the annotation requirements for such a model and simplifies the design of such a model, since the inherent structure of the activity is automatically learned from data. The design consideration that remains is the number of states used for representing the actions that an activity consists of. Using multiple real world datasets we show that the same model works both for the recognition of activities of daily living in a smart home and for recognizing office activities from audio data. We show how a variable number of action states per activity can result in a significant increase in performance over using a fixed number per activity. Finally, we show how the use of Bayesian and Akaike information criterion results in models using a sub-optimal set of action states, since a model using intuitively chosen set states is able to outperform them.
Subject Keywords
Hidden Markov model
,
Actvity recognition
URI
https://hdl.handle.net/11511/40900
DOI
https://doi.org/10.1109/icpr.2014.653
Collections
Department of Computer Engineering, Conference / Seminar
Suggestions
OpenMETU
Core
A Cloud Based Architecture for Distributed Real Time Processing of Continuous Queries
Gökalp, Mert Onuralp; Koçyiğit, Altan; Department of Information Systems (2015)
The technological advancements in Internet of Things (IoT) domain have enabled us to reshape the physical world through smart devices, sensors and actuators. The data collected by IoT devices has become a valuable asset to extract knowledge about the environment and other nearby devices. Existing IoT applications mostly store collected data in a central server and allow users to query stored data to notice and react to changes in the environment. Usually cloud and big data technologies are utilized in those...
An intelligent security architecture for sdn-assisted iot networks
Demirpolat, Ahmed; Angın, Pelin; Department of Computer Engineering (2021-1-26)
The rise of the Internet of Things (IoT) paradigm in the past decade has had a significant impact on all aspects of our lives through the many use cases it has made possible, including smart farming, smart homes, and remote healthcare services, among many others. While the number of smart devices and utilization scenarios aimed at supporting them grow exponentially, the large attack surface created by the interconnectivity of millions of these devices is a concerning aspect that needs to be addressed with i...
A Graph Based Big Data Model for Wireless Multimedia Sensor Networks
Küçükkeçeci, Cihan; Yazıcı, Adnan (2016-10-08)
Wireless multimedia sensor networks are of interest to researchers from different disciplines and many studies have been proposed in a wide variety of application domains, such as military surveillance systems, environmental monitoring, fault monitoring and distributed smart cameras in the last decade. In a wireless sensor network, a large number of sensors can be deployed to monitor target areas and autonomously collect sensor data. This produces a large amount of raw data that needs to be stored, processe...
A generic framework for optimizing performance metrics by tuning parameters of clustering protocols in WSNs
Alchihabi, Abdullah; Dervis, Ates; Ever, Enver; Al-Turjman, Fadi;( Abstracts: Wireless sensor network (WSN) is a key technology trend in emerging internet of things paradigms which are commonly used for application areas such as smart-cities, smart-grids, wearables, and connected health. There is a wealth of literature which considers various cluster-based routing protocols such as LEACH, HEED, and UHEED where these protocols are compared in terms of the network lifetime and/or the total number of packets successfully received by the base station under various operational conditions. While existing studies present various approaches to form WSN clusters in the most efficient way, various parameters are manually-assigned their values such as the radius of the cluster, the number of nodes in the cluster, and the number of clusters that should be formed to reach the base station. The choice of correct parameters is essential for reaching the most efficient configuration, however existing studies do not specify a systematic way for tuning these parameters. In other words, the optimization of cluster-based WSNs through fine tuning of related system parameters is not considered in the existing studies. We believe that presenting a generic approach to tune the parameters of clustering algorithms in order to optimize the performance metrics of WSNs is a significant contribution. In this study a systematic and an efficient method is presented to tune the parameters of clustering and routing protocols. Instead of brute force, or trial and error approaches, simulated annealing and K-beams algorithms are adopted together with discrete event system simulator OMNET++ with Castalia Framework. Results are presented comparatively with brute force approach in order to show the efficiency of the new approach in finding the optimum configuration in terms of energy efficiency as well as the rate of successfully received packets.; 2019-04-01)
Wireless sensor network (WSN) is a key technology trend in emerging internet of things paradigms which are commonly used for application areas such as smart-cities, smart-grids, wearables, and connected health. There is a wealth of literature which considers various cluster-based routing protocols such as LEACH, HEED, and UHEED where these protocols are compared in terms of the network lifetime and/or the total number of packets successfully received by the base station under various operational conditions....
An energy efficient hierarchical approach using multimedia and scalar sensors for emergency services
Kızılkaya, Burak; Ever, Enver; Sustainable Environment and Energy Systems (2019-7)
Recently, environment monitoring and detection systems became more accessible with the help of IoT applications. Furthermore, connecting smart devices makes monitoring applications more accurate and reliable. On the other hand, optimizing the energy requirement of smart sensors especially while transmitting data has always been very important, and there are different applications to create energy efficient IoT systems. Detailed analysis of lifetimes of various types of sensors (survival analysis) has theref...
Citation Formats
IEEE
ACM
APA
CHICAGO
MLA
BibTeX
H. Alemdar, M. E. Niessen, A. MERENTİTİS, and C. Ersoy, “A Unified Model for Human Behavior Modeling using a Hierarchy with a Variable Number of States,” 2014, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/40900.