• Advances in nano research
• /
• v.15 no.5
• /
• pp.451-466
• /
• 2023

Gastrointestinal cancer (GC) is a prevalent malignant tumor of the digestive system that poses a severe health risk to humans. Due to the specific organ structure of the gastrointestinal system, both endoscopic and MRI diagnoses of GIC have limited sensitivity. The primary factors influencing curative efficacy in GIC patients are drug inefficacy and high recurrence rates in surgical and pharmacological therapy. Due to its unique optical features, good biocompatibility, surface effects, and small size effects, nanotechnology is a developing and advanced area of study for the detection and treatment of cancer. Because of its deep location and complex surgery, diagnosing and treating gastrointestinal cancer is very difficult. The early diagnosis and urgent treatment of gastrointestinal illness are enabled by nanotechnology. As diagnostic and therapeutic tools, nanoparticles directly target tumor cells, allowing their detection and removal. XGBoost was used as a classification method known for achieving numerous winning solutions in data analysis competitions, to capture nonlinear relations among many input variables and outcomes using the boosting approach to machine learning. The research sample included 300 GC patients, comprising 190 males (72.2% of the sample) and 110 women (27.8%). Using convolutional neural networks (CNN) and artificial neural networks (ANN)-EXtreme Gradient Boosting (XGBoost), the patients mean± SD age was 50.42 ± 13.06. High-risk behaviors (P = 0.070), age at diagnosis (P = 0.037), distant metastasis (P = 0.004), and tumor stage (P = 0.015) were shown to have a statistically significant link with GC patient survival. AUC was 0.92, sensitivity was 81.5%, specificity was 90.5%, and accuracy was 84.7 when analyzing stomach picture.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.6
• /
• pp.2168-2187
• /
• 2021

The smart grid replaces the traditional power structure with information inventiveness that contributes to a new physical structure. In such a field, malicious information injection can potentially lead to extreme results. Incorrect, FDI attacks will never be identified by typical residual techniques for false data identification. Most of the work on the detection of FDI attacks is based on the linearized power system model DC and does not detect attacks from the AC model. Also, the overwhelming majority of current FDIA recognition approaches focus on FDIA, whilst significant injection location data cannot be achieved. Building on the continuous developments in deep learning, we propose a Deep Learning based Locational Detection technique to continuously recognize the specific areas of FDIA. In the development area solver gap happiness is a False Data Detector (FDD) that incorporates a Convolutional Neural Network (CNN). The FDD is established enough to catch the fake information. As a multi-label classifier, the following CNN is utilized to evaluate the irregularity and cooccurrence dependency of power flow calculations due to the possible attacks. There are no earlier statistical assumptions in the architecture proposed, as they are "model-free." It is also "cost-accommodating" since it does not alter the current FDD framework and it is only several microseconds on a household computer during the identification procedure. We have shown that ANN-MLP, SVM-RBF, and CNN can conduct locational detection under different noise and attack circumstances through broad experience in IEEE 14, 30, 57, and 118 bus systems. Moreover, the multi-name classification method used successfully improves the precision of the present identification.

• Annual Conference on Human and Language Technology
• /
• 2017.10a
• /
• pp.220-224
• /
• 2017

최근 은행, 보험회사 등 핀테크 관련 업체에서는 챗봇과 같은 인공지능 대화 시스템을 고객상담 업무에 도입하고 있다. 본 논문에서는 금융 도메인을 위한 고객상담 챗봇을 구현하기 위하여, 자연어 이해 기술 중 하나인 고객상담 대화의 도메인 및 화행분류 방법을 제시한다. 이 기술을 통해 자연어로 이루어지는 상담내용을 이해하고 적합한 응답을 해줄 수 있는 기술을 개발할 수 있다. TF-IDF, LDA, 문장 임베딩 등 대화 문장에 대한 자질을 추출하고, 추출된 자질을 Extreme learning machine(ELM)을 통해 도메인 및 화행 분류 모델을 학습한다.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.64 no.3
• /
• pp.164-168
• /
• 2015

Due to the uncertainty of weather, it is difficult to construct an accurate forecasting model for daily PV power generation. It is very important work to know PV power in next day to manage power system. In this paper, correlation analysis between weather and power generation was carried out and daily PV power forecasting models based on Extreme Learning Machine(ELM) was presented. Performance of district ELM model was compared with single ELM model. The proposed method has been tested using actual data set measured in 2014.

• Journal of Electrical Engineering and Technology
• /
• v.12 no.4
• /
• pp.1548-1555
• /
• 2017

This paper attempts to employ and investigate neural based approaches as interpolation tools for modeling of Switched Reluctance Motor (SRM) drive. Precise modeling of SRM is essential to analyse the performance of control strategies for variable speed drive application. In this work the suitability of Generalized Regression Neural Network (GRNN) and Extreme Learning Machine (ELM) in addition to conventional neural network are explored for improving the modeling accuracy of SRM. The neural structures are trained with the data obtained by modeling of SRM using Finite Element Analysis (FEA) and the trained neural network is incorporated in the model of SRM drive. The results signify the modeling accuracy with GRNN model. The closed loop drive simulation is performed in MATLAB/Simulink environment and the closeness of the results in comparison with the experimental prototype validates the modeling approach.

• 한국어정보학회:학술대회논문집
• /
• 2017.10a
• /
• pp.220-224
• /
• 2017

최근 은행, 보험회사 등 핀테크 관련 업체에서는 챗봇과 같은 인공지능 대화 시스템을 고객상담 업무에 도입하고 있다. 본 논문에서는 금융 도메인을 위한 고객상담 챗봇을 구현하기 위하여, 자연어 이해 기술 중 하나인 고객상담 대화의 도메인 및 화행분류 방법을 제시한다. 이 기술을 통해 자연어로 이루어지는 상담내용을 이해하고 적합한 응답을 해줄 수 있는 기술을 개발할 수 있다. TF-IDF, LDA, 문장 임베딩 등 대화 문장에 대한 자질을 추출하고, 추출된 자질을 Extreme learning machine(ELM)을 통해 도메인 및 화행 분류 모델을 학습한다.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.61 no.4
• /
• pp.228-233
• /
• 2012

In power system, substation facilities have become too complex and larger according to an extended power system. Also, customers require the high quality of electrical power system. However, some facilities become old and often break down unexpectedly. The unexpected failure may cause a break in power system and loss of profits. Therefore it is important to prevent abrupt faults by monitoring the condition of power systems. Among the various power facilities, power transformers play an important role in the transmission and distribution systems. In this research, we develop intelligent diagnosis technique for monitoring of power transformer by FCM(Fuzzy c-means) and ELM(Extreme Learning Machine). The proposed technique make it possible to diagnosis the faults occurred in transformer. To demonstrate the validity of proposed method, various experiments are performed and their results are presented.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.65 no.3
• /
• pp.204-209
• /
• 2016

It is difficult to establish accurate modeling of PV power system because of various uncertainty. However, it is important work to modeling of PV for fault diagnosis. This paper proposes modeling and fault diagnosis method using correlation analysis and ELM(Extreme Learning Machine). Rather than using total data, we select optimal time interval with higher corelation between PV power and solar irradiation. Also, we use average value during 60 minute to avoid rapid variation of PV power. To show the effectiveness of the proposed method, we performed various experiments by dataset.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.05a
• /
• pp.204-207
• /
• 2013

Music genre classification is an essential component for music information retrieval system. There are two important components to be considered for better genre classification, which are audio feature extraction and classifier. This paper incorporates two different kinds of features for genre classification, timbral texture and rhythmic content features. Timbral texture contains several spectral and Mel-frequency Cepstral Coefficient (MFCC) features. Before choosing a timbral feature we explore which feature contributes less significant role on genre discrimination. This facilitates the reduction of feature dimension. For the timbral features up to the 4-th order central moments and the covariance components of mutual features are considered to improve the overall classification result. For the rhythmic content the features extracted from beat histogram are selected. In the paper Extreme Learning Machine (ELM) with bagging is used as classifier for classifying the genres. Based on the proposed feature sets and classifier, experiment is performed with well-known datasets: GTZAN databases with ten different music genres, respectively. The proposed method acquires the better classification accuracy than the existing approaches.

• ETRI Journal
• /
• v.44 no.2
• /
• pp.327-338
• /
• 2022

Classification models for human action recognition require robust features and large training sets for good generalization. However, data augmentation methods are employed for imbalanced training sets to achieve higher accuracy. These samples generated using data augmentation only reflect existing samples within the training set, their feature representations are less diverse and hence, contribute to less precise classification. This paper presents new data augmentation and action representation approaches to grow training sets. The proposed approach is based on two fundamental concepts: virtual video generation for augmentation and representation of the action videos through robust features. Virtual videos are generated from the motion history templates of action videos, which are convolved using a convolutional neural network, to generate deep features. Furthermore, by observing an objective function of the genetic algorithm, the spatiotemporal features of different samples are combined, to generate the representations of the virtual videos and then classified through an extreme learning machine classifier on MuHAVi-Uncut, iXMAS, and IAVID-1 datasets.