• The Korean Journal of Applied Statistics
• /
• v.7 no.2
• /
• pp.53-77
• /
• 1994

In a certain stochastic process, Cox's regression model is used to analyze multistate survival data. From this model, the regression parameter vectors, survival functions, and the probability of being in response function are estimated based on multistate Cox's partial likelihood and nonparametric likelihood methods. The asymptotic properties of these estimators are described informally through the counting process approach. An example is given to likelihood the results in this paper.

• IE interfaces
• /
• v.24 no.4
• /
• pp.323-329
• /
• 2011

We develop a stochastic model to predict the score of a soccer match. We describe the scoring process of the soccer match as a markovian arrival process (MAP). To do this, we define a two-state underlying Markov chain, in which the two states represent the offense and defense states of the two teams to play. Then, we derive the probability vector generating function of the final scores. Numerically inverting this generating function, we obtain the desired probability distribution of the scores. Sample numerical examples are given at the end to demonstrate how to utilize this result to predict the final score of the match.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.12
• /
• pp.43-48
• /
• 2018

In this paper, more efficient classification result could be obtained by applying the combination of the Hidden Markov Model and SVM Model to HMSV algorithm gene expression data which simulated the stochastic flow of gene data and clustering it. In this paper, we verified the HMSV algorithm that combines independently learned algorithms. To prove that this paper is superior to other papers, we tested the sensitivity and specificity of the most commonly used classification criteria. As a result, the K-means is 71% and the SOM is 68%. The proposed HMSV algorithm is 85%. These results are stable and high. It can be seen that this is better classified than using a general classification algorithm. The algorithm proposed in this paper is a stochastic modeling of the generation process of the characteristics included in the signal, and a good recognition rate can be obtained with a small amount of calculation, so it will be useful to study the relationship with diseases by showing fast and effective performance improvement with an algorithm that clusters nodes by simulating the stochastic flow of Gene Data through data mining of BigData.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.10
• /
• pp.4034-4053
• /
• 2015

In order to meet various requirements for transmission quality of both primary users (PUs) and secondary users (SUs) in cognitive radio networks, we introduce a channel bonding mechanism for PUs and a channel reservation mechanism for SUs, then we propose a novel spectrum allocation strategy. Taking into account the mistake detection and false alarm due to imperfect channel sensing, we establish a three-dimensional Markov chain to model the stochastic process of the proposed strategy. Using the method of matrix geometric solution, we derive the performance measures in terms of interference rate of PU packets, average delay and throughput of SU packets. Moreover, we investigate the influence of the number of the reserved (resp. licensed) channels on the system performance with numerical experiments. Finally, to optimize the proposed strategy socially, we provide a charging policy for SU packets.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.2B
• /
• pp.215-224
• /
• 2008

The effect of streamflow considering future land use change and vegetation index information by climate change scenario was assessed using SLURP (Semi-distributed Land-Use Runoff Process) model. The model was calibrated and verified using 4 years (1999-2002) daily observed streamflow data for the upstream watershed ($260.4km^2$) of Gyeongan water level gauging station. By applying CA-Markov technique, the future land uses (2030, 2060, 2090) were predicted after test the comparison of 2004 Landsat land use and 2004 CA-Markov land use by 1996 and 2000 land use data. The future land use showed a tendency that the forest and paddy decreased while urban, grassland and bareground increased. The future vegetation indices (2030, 2060, 2090) were estimated by the equation of linear regression between monthly NDVI of NOAA AVHRR images and monthly mean temperature of 5 years (1998-2002). Using CCCma CGCM2 simulation result based on SRES A2 and B2 scenario (2030s, 2060s, 2090s) of IPCC and data were downscaled by Stochastic Spatio-Temporal Random Cascade Model (SST-RCM) technique, the model showed that the future runoff ratio was predicted from 13% to 34% while the runoff ratio of 1999-2002 was 59%. On the other hand, the impact on runoff ratio by land use change showed about 0.1% to 1% increase.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.40 no.3
• /
• pp.63-72
• /
• 2015

This paper presents the issue that is predicting the movement of an agent in an enclosed space by using the MDP (Markov Decision Process). Recent researches on the optimal path finding are confined to derive the shortest path with the use of deterministic algorithm such as $A^*$ or Dijkstra. On the other hand, this study focuses in predicting the path that the agent chooses to escape the limited space as time passes, with the stochastic method. The MDP reward structure from GIS (Geographic Information System) data contributed this model to a feasible model. This model has been approved to have the high predictability after applied to the route of previous armed red guerilla.

• Journal of the military operations research society of Korea
• /
• v.29 no.2
• /
• pp.100-110
• /
• 2003

Equipment replacement policy may not be defined with certainty, because physical states of any technological system may not be determined with foresight. This paper presents Markov Decision Process(MDP) model for army equipment which is subject to the uncertainty of deterioration and ultimately to failure. The components of the MDP model is defined as follows: ⅰ) state is identified as the age of the equipment, ⅱ) actions are classified as 'keep' and 'replace', ⅲ) cost is defined as the expected cost per unit time associated with 'keep' and 'replace' actions, ⅳ) transition probability is derived from Weibull distribution. Using the MDP model, we can determine the optimal replacement policy for an army equipment replacement problem.

• Proceedings of the KSPS conference
• /
• 2006.11a
• /
• pp.14-18
• /
• 2006

The role of dialogue manager is to select proper actions based on observed environment and inferred user intention. This paper presents stochastic model for dialogue manager based on Markov decision process. To build a mixed initiative dialogue manager, we used accumulated user utterance, previous act of dialogue manager, and domain dependent knowledge as the input to the MDP. We also used dialogue corpus to train the automatically optimized policy of MDP with reinforcement learning algorithm. The states which have unique and intuitive actions were removed from the design of MDP by using the domain knowledge. The design of dialogue manager included the usage of natural language understanding and response generator to build short message based remote control of home networked appliances.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.2
• /
• pp.579-603
• /
• 2018

In this paper we propose a QoS-enhanced intelligent stochastic optimal fair real-time packet scheduler, QUEST, for 4G LTE traffic in routers. The objective of this research is to maximize the system QoS subject to the constraint that the processor utilization is kept nearly at 100 percent. The QUEST has following unique advantages. First, it solves the challenging problem of starvation for low priority process - buffered streaming video and TCP based; second, it solves the major bottleneck of the scheduler Earliest Deadline First's failure at heavy loads. Finally, QUEST offers the benefit of arbitrarily pre-programming the process utilization ratio.Three classes of multimedia 4G LTE QCI traffic, conversational voice, live streaming video, buffered streaming video and TCP based applications have been considered. We analyse two most important QoS metrics, packet loss rate (PLR) and mean waiting time. All claims are supported by discrete event and Monte Carlo simulations. The simulation results show that the QUEST scheduler outperforms current state-of-the-art benchmark schedulers. The proposed scheduler offers 37 percent improvement in PLR and 23 percent improvement in mean waiting time over the best competing current scheduler Accuracy-aware EDF.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.45 no.5
• /
• pp.125-133
• /
• 2008

In this paper, we propose active queue management mechanism (Active-WRED) to guarantee quality of the high priority service class in multi-class traffic service environment. In congestion situation, this mechanism increases drop probability of low priority traffic and reduces the drop probability of the high priority traffic, therefore it can improve the quality of the high priority service. In order to analyze the performance of our mechanism we introduce the stochastic analysis of a discrete-time queueing systems for the performance evaluation of the Active Queue Management (AQM) based congestion control mechanism called Weighted Random Early Detection (WRED) using a two-state Markov-Modulated Bernoulli arrival process (MMBP-2) as the traffic source. A two-dimensional discrete-time Harkov chain is introduced to model the Active-WRED mechanism for two traffic classes (Guaranteed Service and Best Effort Service) where each dimension corresponds to a traffic class with its own parameters.