Pattern Classification Model Design and Performance Comparison for Data Mining of Time Series Data

Lee, Soo-Yong; Lee, Kyoung-Joung

doi:10.5391/jkiis.2011.21.6.730

Cited by 2 publications

(2 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, rules were defined using the sequential pattern of mining techniques through the pattern analysis of exams conducted on each outpatient [ 15 ]. Second, a one & nonstop environment through master management of exams and a multi-reservation viewer were designed to manage difficulties and exam constraints [ 16 ]. Recommendation algorithms consist of content-based algorithms and rule based algorithms.…”

Section: Introductionmentioning

confidence: 99%

Implementation of Hospital Examination Reservation System Using Data Mining Technique

et al. 2015

View full text Add to dashboard Cite

ObjectivesNew methods for obtaining appropriate information for users have been attempted with the development of information technology and the Internet. Among such methods, the demand for systems and services that can improve patient satisfaction has increased in hospital care environments.MethodsIn this paper, we proposed the Hospital Exam Reservation System (HERS), which uses the data mining method. First, we focused on carrying clinical exam data and finding the optimal schedule for generating rules using the multi-examination pattern-mining algorithm. Then, HERS was applied by a rule master and recommending system with an exam log. Finally, HERS was designed as a user-friendly interface.ResultsHERS has been applied at the National Cancer Center in Korea since June 2014. As the number of scheduled exams increased, the time required to schedule more than a single condition decreased (from 398.67% to 168.67% and from 448.49% to 188.49%; p < 0.0001). As the number of tests increased, the difference between HERS and non-HERS increased (from 0.18 days to 0.81 days).ConclusionsIt was possible to expand the efficiency of HERS studies using mining technology in not only exam reservations, but also the medical environment. The proposed system based on doctor prescription removes exams that were not executed in order to improve recommendation accuracy. In addition, we expect HERS to become an effective system in various medical environments.

show abstract

Section: Introductionmentioning

confidence: 99%

Implementation of Hospital Examination Reservation System Using Data Mining Technique

et al. 2015

View full text Add to dashboard Cite

show abstract

“…While such models are implicitly employed for joint density estimation, for the last few decades they have gained significant attention as classifiers. A model of this class, the Bayesian network classifier (BNC) [3], has been used in a wide range of applications subsuming speech recognition and motion time-series classification [4][5][6][7][8][9] and has been shown | 186…”

Section: Introductionmentioning

confidence: 99%

Semi-Supervised Recursive Learning of Discriminative Mixture Models for Time-Series Classification

Kim

2013

International Journal of Fuzzy Logic and Intelligent Systems

View full text Add to dashboard Cite

We pose pattern classification as a density estimation problem where we consider mixtures of generative models under partially labeled data setups. Unlike traditional approaches that estimate density everywhere in data space, we focus on the density along the decision boundary that can yield more discriminative models with superior classification performance.We extend our earlier work on the recursive estimation method for discriminative mixture models to semi-supervised learning setups where some of the data points lack class labels.Our model exploits the mixture structure in the functional gradient framework: it searches for the base mixture component model in a greedy fashion, maximizing the conditional class likelihoods for the labeled data and at the same time minimizing the uncertainty of class label prediction for unlabeled data points. The objective can be effectively imposed as individual mixture component learning on weighted data, hence our mixture learning typically becomes highly efficient for popular base generative models like Gaussians or hidden Markov models. Moreover, apart from the expectation-maximization algorithm, the proposed recursive estimation has several advantages including the lack of need for a pre-determined mixture order and robustness to the choice of initial parameters. We demonstrate the benefits of the proposed approach on a comprehensive set of evaluations consisting of diverse time-series classification problems in semi-supervised scenarios.

show abstract

Pattern Classification Model Design and Performance Comparison for Data Mining of Time Series Data

Cited by 2 publications

References 12 publications

Implementation of Hospital Examination Reservation System Using Data Mining Technique

Implementation of Hospital Examination Reservation System Using Data Mining Technique

Semi-Supervised Recursive Learning of Discriminative Mixture Models for Time-Series Classification

Contact Info

Product

Resources

About