• 한국컴퓨터정보학회논문지
• /
• 제25권9호
• /
• pp.53-62
• /
• 2020

최근 연구된 상품 추천 서비스들은 고객들의 구매 이력을 통해서만 추천이 이루어졌다. 본 논문에서는 구매 이력을 통해 추천뿐만 아니라 고객이 상품을 고를 때 취하는 행동 패턴을 분석하여 관심도가 높은 광고를 노출하는 행동 패턴 분석 기반 디지털 사이니지 서비스를 제안한다. 이 서비스는 고객행동 패턴을 분석하여 실질적으로 관심을 가지는 상품에 대해 관심도를 추출한다. 추출된 관심도와 고객들의 구매 이력을 Wide & Deep 모델을 통해 학습하고, 이를 바탕으로 MF(Matrix Factorization) 모델을 통해 다른 상품들의 희소 벡터를 예측한다. 예측된 상품 관심도에 대한 순위를 도출하고, 적합한 광고를 노출하기 위해 고객과 상호 작용할 수 있는 인도어 사이니지를 활용한다. 본 논문의 서비스를 통해 온라인뿐만이 아닌 오프라인 환경에서도 고객의 관심 정보를 파악하고 단순히 무작위로 노출하는 광고가 아닌 고객에게 적합한 광고를 제공하여 만족도 높은 구매 환경이 조성될 것이다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권2호
• /
• pp.538-561
• /
• 2020

Recommender Systems (RecSys) have a major role in e-commerce for recommending products, which they may like for every user and thus improve their business aspects. Although many types of RecSyss are there in the research field, the state of the art RecSys has focused on finding the user similarity based on sequence (e.g. purchase history, movie-watching history) analyzing and prediction techniques like Recurrent Neural Network in Deep learning. That is RecSys has considered as a sequence prediction problem. However, evaluation of similarities among the customers is challenging while considering temporal aspects, context and multi-component ratings of the item-records in the customer sequences. For addressing this issue, we are proposing a Deep Learning based model which learns customer similarity directly from the sequence to sequence similarity as well as item to item similarity by considering all features of the item, contexts, and rating components using Dynamic Temporal Warping(DTW) distance measure for dynamic temporal matching and 2D-GRU (Two Dimensional-Gated Recurrent Unit) architecture. This will overcome the limitation of non-linearity in the time dimension while measuring the similarity, and the find patterns more accurately and speedily from temporal and spatial contexts. Experiment on the real world movie data set LDOS-CoMoDa demonstrates the efficacy and promising utility of the proposed personalized RecSys architecture.

• 한국재난정보학회 논문집
• /
• 제17권2호
• /
• pp.254-265
• /
• 2021

연구목적: 국내 항만시설의 진단을 통한 성능 및 안전에 대한 점검과 진단을 20년 넘게 진행되었지만 그 진단 이력과 결과를 활용한 중장기적인 시설개선과 성능개선을 위한 발전전략이나 방향이 현실적으로 작동하지 않고 있다. 특히, 사용년수가 오래된 항만구조물의 경우, 선박의 대형화와 사용빈도 증가, 기후변화로 인한 자연재해의 영향 등으로 안전성능과 기능적 면에서 상당히 많은 문제점을 내포하고 있다. 연구방법: 본 연구에서는 중력식 안벽에 대한 부재수준의 유지관리 이력 데이터를 수집하여 이를 빅데이터로써 정의하고 해당 데이터를 바탕으로 프로젝트 수준의 시설물의 노후화 패턴 및 열화를 추정하기 위한 예측근사모델을 도출하였다. 특히 GP 및 SGP 기법의 머신러닝 알고리즘을 통하여 생성된 상태기반 노후도 패턴 및 열화 근사모델에 대한 유효성 검토를 통해 빅데이터 활용에 적합한 모델을 상호비교하고 제안하였다. 연구결과: 제안된 기법의 적합성을 검토한 결과 GP기법은 RMSE 및 R²는 0.9854와 0.0721, SGP기법은 0.7246과 0.2518로 GP기법을 적용한 예측모델이 적합한 것으로 검토 되었다. 결론: 머신러닝 기법을 통해 이러한 연구는 향후 항만시설 데이터취합이 지속적으로 이루어진다면 향후 항만시설 투자의사결정에 중요한 역할을 할 것으로 기대한다.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제7권12호
• /
• pp.461-468
• /
• 2018

개발자들은 소프트웨어 개발과 유지보수 작업 중 하나의 코드를 수정하는데 들이는 시간보다 이를 위해 코드를 탐색하고 이해하는데 더 많은 시간을 소모한다. 코드를 탐색하는 시간을 줄이기 위하여 기존 연구들은 데이터 마이닝과 통계적 언어모델 기법을 이용하여 수정할 코드를 추천하여 왔다. 그러나 이 경우 모델의 학습 데이터와 입력되는 데이터가 정확하게 일치하지 않으면 추천이 발생하지 않는다. 이 논문에서 우리는 딥러닝의 기법 중 하나인 Recurrent Neural Networks에 동작기록을 학습시켜 기존 연구의 상기 문제점 없이 수정할 코드의 위치를 추천하는 방법을 제안한다. 제안 방법은 RNN과 동작기록을 활용한 추천 기법으로 평균 약 91%의 정확도와 71%의 재현율을 달성함으로써 기존의 추천방법보다 코드 탐색 시간을 더욱 줄일 수 있게 해 준다.

• 한국빅데이터학회지
• /
• 제5권2호
• /
• pp.53-67
• /
• 2020

본 연구는 가공식품의 제조·가공 업소를 대상으로 기계학습 분야의 지도학습(Supervised Learning) 예측 모형을 적용하여 부적합이 예상되는 업체를 사전에 적발하는 단속 선별시스템을 마련하여 단속 활동의 효율성을 높이고자 하였다. 본 연구에서는 머신러닝의 예측 모델링을 위한 목적 정의, 데이터의 기초 분석과 시각화, 특성 변수 도출 및 예측 모형의 선정 및 예측 등으로 기계학습 수행의 표준적인 절차에 따라 연구를 수행하였다. 종속변수는 2014년도부터 2018년까지 과거 5년 동안 지도점검 적발 건수로 설정하였고, 목적함수는 실제 부적합업체를 사전에 판정하여 단속활동이 이루어지는 것을 최대화하는 것으로 하였다. 제조가공업소의 매출액, 영업일수, 종업원 수 등 기본속성뿐만 아니라 과거 지도점검 단속 이력 정보를 반영하여 자료를 재구성하였다. 특성 변수 추출 방법을 적용하여 부적합 판정에 영향을 미치는 업체 위험, 품목 위험, 환경 위험 및 과거 위반 이력 등을 특성 변수로 도출하여 머신러닝 알고리즘을 데이터에 적용하였다. 랜덤포레스트 모형이 식품의약품안전처 지도점검 업무 목적에 가장 적합한 것으로 나타났다. 본 연구결과를 바탕으로 식품안전 관리 국가 사무가 데이터기반의 과학적인 행정 체계로 발전할 수 있는 기반이 되기를 기대한다.

• Smart Structures and Systems
• /
• 제23권5호
• /
• pp.507-520
• /
• 2019

In the structural health monitoring field, damage detection has been commonly carried out based on the structural model and the engineering features related to the model. However, the extracted features are often subjected to various errors, which makes the pattern recognition for damage detection still challenging. In this study, an automated damage identification method is presented for hanger cables in a tied-arch bridge using a convolutional neural network (CNN). Raw measurement data for Fourier amplitude spectra (FAS) of acceleration responses are used without a complex data pre-processing for modal identification. A CNN is a kind of deep neural network that typically consists of convolution, pooling, and fully-connected layers. A numerical simulation study was performed for multiple damage detection in the hangers using ambient wind vibration data on the bridge deck. The results show that the current CNN using FAS data performs better under various damage states than the CNN using time-history data and the traditional neural network using FAS. Robustness of the present CNN has been proven under various observational noise levels and wind speeds.

• Asia pacific journal of information systems
• /
• 제29권4호
• /
• pp.752-770
• /
• 2019

The purpose of this study is to incorporate telemarketing processes to improve telemarketing performance. For this application, we have attempted to mix the model of machine learning to extract potential customers with personalisation techniques to derive recommended products from actual contact. Most of traditional recommendation systems were mainly in ways such as collaborative filtering, which predicts items with a high likelihood of future purchase, based on existing purchase transactions or preferences for products. But, under these systems, new users or items added to the system do not have sufficient information, and generally cause problems such as a cold start that can not obtain satisfactory recommendation items. Also, indiscriminate telemarketing attempts can backfire as they increase the dissatisfaction and fatigue of customers who do not want to be contacted. To this purpose, this study presented a multi-purpose hybrid recommendation algorithm to achieve two goals: to select customers with high possibility of contact, and to recommend products to selected customers. In addition, we used subscription data from telemarketing agency that handles insurance products to derive realistic applicability of the proposed recommendation system. Our proposed recommendation system would certainly solve the cold start and scarcity problem of existing recommendation algorithm by using contents information such as customer master information and telemarketing history. Also. the model could show excellent performance not only in terms of overall performance but also in terms of the recommendation success rate of the unpopular product.

• 대한정형도수물리치료학회지
• /
• 제18권2호
• /
• pp.9-21
• /
• 2012

Background: The purpose of this study was to analysis and to give information about the type of orthopedic manual physical therapy research by KAOMPT Journal from 1995 (Volume 1) to 2011 (Volume 17). The number of research paper were total 212 studies. Method: The studies were analysed with retrospective descriptive study. The results of data analysis was as follow; Results & Conclusion: 1. The studies was analysed 6 study field; manual therapy for disease field (130 studies), the therapeutic (16 studies), the effects of manual therapy education method, learning model (7 studies), medical insurance, history, politics of manual therapy (3 studies), and others (19 studies). 2. In the manual therapy of disease field, there were 4 groups of disease; bone disease (43 studies), soft tissue and muscles disease (57 studies), nerve disease(24 studies), and the others (6 studies). The main disease were low back pain, stroke, HIVD. 3. In the assessment, and intervention field, there were only 5 studies during from 1995 to 2000, but the studies from 2001 to 2011 were more increased 21 studies. 4. In the effects of manual therapy education method, learning model, there were very few studies (7 studies), so it will be needed more studies in this field. 5. In the other fields such as more political issue, history, medical insurance for the manual therapy there were few studies, so it will be needed more advanced studies in this field.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• 제49권3호
• /
• pp.135-141
• /
• 2023

Objectives: This study aimed to develop and validate machine learning (ML) models using H₂O-AutoML, an automated ML program, for predicting medication-related osteonecrosis of the jaw (MRONJ) in patients with osteoporosis undergoing tooth extraction or implantation. Patients and Methods: We conducted a retrospective chart review of 340 patients who visited Dankook University Dental Hospital between January 2019 and June 2022 who met the following inclusion criteria: female, age ≥55 years, osteoporosis treated with antiresorptive therapy, and recent dental extraction or implantation. We considered medication administration and duration, demographics, and systemic factors (age and medical history). Local factors, such as surgical method, number of operated teeth, and operation area, were also included. Six algorithms were used to generate the MRONJ prediction model. Results: Gradient boosting demonstrated the best diagnostic accuracy, with an area under the receiver operating characteristic curve (AUC) of 0.8283. Validation with the test dataset yielded a stable AUC of 0.7526. Variable importance analysis identified duration of medication as the most important variable, followed by age, number of teeth operated, and operation site. Conclusion: ML models can help predict MRONJ occurrence in patients with osteoporosis undergoing tooth extraction or implantation based on questionnaire data acquired at the first visit.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2023년도 학술발표회
• /
• pp.175-175
• /
• 2023

Detection and monitoring of wildfires are essential for limiting their harmful effects on ecosystems, human lives, and property. In this research, we propose a novel method running in the Google Earth Engine platform for identifying and characterizing burnt regions using a hybrid of Sentinel-1 (C-band synthetic aperture radar) and Sentinel-2 (multispectral photography) images. The 2022 Uljin wildfire, the severest event in South Korean history, is the primary area of our investigation. Given its documented success in remote sensing and land cover categorization applications, we select the Random Forest (RF) method as our primary classifier. Next, we evaluate the performance of our model using multiple accuracy measures, including overall accuracy (OA), Kappa coefficient, and area under the curve (AUC). The proposed method shows the accuracy and resilience of wildfire identification compared to traditional methods that depend on survey data. These results have significant implications for the development of efficient and dependable wildfire monitoring systems and add to our knowledge of how machine learning and remote sensing-based approaches may be combined to improve environmental monitoring and management applications.