• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2002년도 추계정기학술대회
• /
• pp.203-212
• /
• 2002

This study examined the predictive power of data mining algorithms by comparing the performance of logistic regression and decision tree algorithm, called CHAID (Chi-squared Automatic Interaction Detection), On the contrary to the previous studies, decision tree performed better than logistic regression. We have also developed a CDSS (Clinical Decision Support System) with three modules (doctor, nurse, and patient) based on data warehouse architecture. Data warehouse collects and integrates relevant information from various databases from hospital information system (HIS ). This system can help improve decision making capability of doctors and improve accessibility of educational material for patients.

• 한국데이터정보과학회:학술대회논문집
• /
• 한국데이터정보과학회 2003년도 추계학술대회
• /
• pp.239-247
• /
• 2003

Decision tree algorithms are used extensively for data mining in many domains such as retail target marketing, fraud dection, data reduction and variable screening, etc. CHAID(Chi-square Automatic Interaction Detector), is an exploratory method used to study the relationship between a dependent variable and a series of predictor variables. In this paper we propose and CHAID algorithm by cube-based sampling and explore CHAID algorithm in view of accuracy and speed by the number of variables.

• 대한한의학회지
• /
• 제33권4호
• /
• pp.1-8
• /
• 2012

Objectives: The purpose of this study was to specify major tongue diagnostic indicators and evaluate their significance in discriminating pattern identification subtypes in stroke patients. Methods: This study used a community based multi-center observational design. Participants (n=1,502) were stroke patients admitted to 11 oriental medical university hospitals between December 2006 and February 2010. To determine which tongue indicator affected each pattern identification, a decision tree analysis of the chi-square automatic interaction detector (CHAID) algorithm was performed. The chi-squared test was used as the criterion in splitting data with a p-value less than 0.05 for division, which is the main procedure for developing a decision tree. The minimum sample size for each node was specified as n =10, and branching was limited to two levels. Results: From the 9 tongue diagnostic indicators, 6 major tongue indicators (red tongue, pale tongue, yellow fur, white fur, thick fur, and teeth-marked tongue) were identified through the decision tree analysis. Furthermore, each pattern identification was composed of specific combinations of the 6 major tongue indicators. Conclusions: This study suggests that the 6 tongue indicators identified through the decision tree analysis can be used to discriminate pattern identification subtypes in stroke patients. However, it is still necessary to re-evaluate other pattern identification indicators to further the objectivity and reliability of traditional Korean medicine.

• Journal of the Korean Data and Information Science Society
• /
• 제16권1호
• /
• pp.41-50
• /
• 2005

The various methods have been studied to develop discriminant model for pregnancy induced hypertension(PIH) as high risk pregnant. In this study, we adapt the approximate entropy which is the non-linear chaotic measuring method. Then, we develop a system to discriminant PIH pregnant using QUEST with S-PLUS.

• Journal of Communications and Networks
• /
• 제18권5호
• /
• pp.699-712
• /
• 2016

Recently, greedy algorithm has received much attention as a cost-effective means to reconstruct the sparse signals from compressed measurements. Much of previous work has focused on the investigation of a single candidate to identify the support (index set of nonzero elements) of the sparse signals. Well-known drawback of the greedy approach is that the chosen candidate is often not the optimal solution due to the myopic decision in each iteration. In this paper, we propose a tree search based sparse signal recovery algorithm referred to as the tree search matching pursuit (TSMP). Two key ingredients of the proposed TSMP algorithm to control the computational complexity are the pre-selection to put a restriction on columns of the sensing matrix to be investigated and the tree pruning to eliminate unpromising paths from the search tree. In numerical simulations of Internet of Things (IoT) environments, it is shown that TSMP outperforms conventional schemes by a large margin.

• 한국군사과학기술학회지
• /
• 제22권4호
• /
• pp.555-566
• /
• 2019

In order to prepare for the future warfare environment, which requires a faster operational tempo, it is necessary to utilize the fourth industrial revolution technology in the field of military operations. This study propose a methodology, 'machine learning based dynamic targeting', which can contribute to reduce required man-hour for dynamic targeting. Specifically, a decision tree algorithm is considered to apply to dynamic targeting process. The algorithm learns target prioritization patterns from JIPTL(Joint Integrated Prioritized Target List) which is the result of the deliberate targeting, and then learned algorithm rapidly(almost real-time) determines priorities for new targets that occur during ATO(Air Tasking Order) execution. An experiment is performed with artificially generated data to demonstrate the applicability of the methodology.

• 스마트미디어저널
• /
• 제13권2호
• /
• pp.23-31
• /
• 2024

In precision cattle farm, reliably tracking the identity of each cattle is necessary. Effective tracking of cattle within farm environments presents a unique challenge, particularly with the need to minimize the occurrence of excessive tracking trajectories. To address this, we introduce a trajectory playback decision tree algorithm that reevaluates and cleans tracking results based on spatio-temporal relationships among trajectories. This approach considers trajectory as metadata, resulting in more realistic and accurate tracking outcomes. This algorithm showcases its robustness and capability through extensive comparisons with popular tracking models, consistently demonstrating the promotion of performance across various evaluation metrics that is HOTA, AssA, and IDF1 achieve 68.81%, 79.31%, and 84.81%.

• 한국음향학회지
• /
• 제32권2호
• /
• pp.174-180
• /
• 2013

본 논문은 HMM 기반의 TTS 시스템을 위하여 상호유사도 비율을 이용한 결정트리 기반의 문맥 군집화 알고리즘을 제안한다. 기존의 알고리즘들은 유사한 통계적 특성을 가지는 문맥종속 HMM을 하나로 묶고 있다. 그러나 기존의 알고리즘들은 결정트리의 나누어진 노드간의 통계적 유사도를 고려하지 않음으로 인하여 최종 노드 사이의 통계적인 차이를 보장하지 못한다. 제안한 알고리즘은 분리된 노드들 간의 통계적 유사도를 최소화하여 모델 파라미터의 신뢰도를 향상시킨다. 실험 결과를 통해 제안한 알고리즘이 기존의 알고리즘들에 비해 우수한 성능을 나타낸다는 것을 확인할 수 있다.

• ETRI Journal
• /
• 제44권3호
• /
• pp.462-475
• /
• 2022

Although researchers have proposed numerous techniques for speech emotion recognition, its performance remains unsatisfactory in many application scenarios. In this study, we propose a speech emotion recognition model based on a genetic algorithm (GA)-decision tree (DT) fusion of deep and acoustic features. To more comprehensively express speech emotional information, first, frame-level deep and acoustic features are extracted from a speech signal. Next, five kinds of statistic variables of these features are calculated to obtain utterance-level features. The Fisher feature selection criterion is employed to select high-performance features, removing redundant information. In the feature fusion stage, the GA is is used to adaptively search for the best feature fusion weight. Finally, using the fused feature, the proposed speech emotion recognition model based on a DT support vector machine model is realized. Experimental results on the Berlin speech emotion database and the Chinese emotion speech database indicate that the proposed model outperforms an average weight fusion method.

• 융합보안논문지
• /
• 제23권5호
• /
• pp.73-79
• /
• 2023

사이버 위협은 기술의 발전에 따라 진화되고 정교해지고 있으며, DDoS 공격으로 인한 서비스 장애를 발생 이슈들이 증가하고 있다. 최근 DDoS 공격은 특정 서비스나 서버의 도메인 주소에 대량의 트래픽을 유입시켜 서비스 장애를 발생시키는 유형이 많아지고 있다. 본 논문에서는 대역폭 소진 공격의 대표적인 공격 유형인 Syn Flooding 공격의 데이터를 생성 후, 효과적인 공격 탐지를 위해 Random Forest, Decision Tree, Multi-Layer Perceptron, KNN 알고리즘을 사용하여 비교 분석하였고 최적의 알고리즘을 도출하였다. 이 결과를 토대로 Syn Flooding 공격 탐지 정책을 위한 기법으로 효과적인 활용이 가능할 것이다.