• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권10호
• /
• pp.3498-3512
• /
• 2021

This article describes how events that make up text stories can be represented and extracted. We also address the results from our simple experiment on extracting and clustering events in terms of emotions, under the assumption that different emotional events can be associated with the classified clusters. Each emotion cluster is based on Plutchik's eight basic emotion model, and the attributes of the NLTK-VADER are used for the classification criterion. While comparisons of the results with human raters show less accuracy for certain emotion types, emotion types such as joy and sadness show relatively high accuracy. The evaluation results with NRC Word Emotion Association Lexicon (aka EmoLex) show high accuracy values (more than 90% accuracy in anger, disgust, fear, and surprise), though precision and recall values are relatively low.

• Asia pacific journal of information systems
• /
• 제28권4호
• /
• pp.308-319
• /
• 2018

This paper aims at providing valuable insights on Financial Fraud Detection on a mobile money transactional activity. We have predicted and classified the transaction as normal or fraud with a small sample and massive data set using Azure and Spark ML, which are traditional systems and Big Data respectively. Experimenting with sample dataset in Azure, we found that the Decision Forest model is the most accurate to proceed in terms of the recall value. For the massive data set using Spark ML, it is found that the Random Forest classifier algorithm of the classification model proves to be the best algorithm. It is presented that the Spark cluster gets much faster to build and evaluate models as adding more servers to the cluster with the same accuracy, which proves that the large scale data set can be predictable using Big Data platform. Finally, we reached a recall score with 0.73, which implies a satisfying prediction quality in predicting fraudulent transactions.

• 인지과학
• /
• 제35권1호
• /
• pp.1-21
• /
• 2024

본 연구는 두 가지 이론적 모델인 통합된 객체 모형과 특장 병렬-독립 저장 모형을 검증함으로써 시각작업기억 표상의 특성을 조사하였다. 실험 I에서 참가자들은 색상 사각형, 방위 막대 또는 두 가지 모두로 구성된 배열을 기억한 뒤 이를 토대로 변화탐지과제를 수행했다. 단일 특징 조건에서 기억배열은 하나의 특징(방위 또는 색상)으로만 구성된 반면, 두 가지 특징 조건은 둘 모두를 포함했다. 두 조건간 변화탐지 수행의 차이는 없었으며 이는 병렬-독립 저장 모형보다는 통합된 객체 모형을 지지한다. 실험 II에서는 이등변삼각형의 방위, 색상 사각형 또는 두 특징 모두로 구성된 기억배열을 대상으로 회상과제가 실시되었으며, 단일 특징과 두 가지 특징 조건 간 회상 수행이 비교되었다. 두 조건 간 회상 정확도에는 차이가 없었으나 표상 선명도와 추측반응에 대한 분석 결과는 강한 객체 모형보다는 약한 객체 모형을 시사했다. 본 연구의 결과는 시각작업기억의 표상 특성을 둘러싼 현시점의 논쟁에 있어서 병렬-독립 저장 모형이 아닌 통합된 객체 모형의 우세를 지지한다.

• 한국작물학회:학술대회논문집
• /
• 한국작물학회 2022년도 추계학술대회
• /
• pp.80-80
• /
• 2022

Estimation of heading date of paddy rice is laborious and time consuming. Therefore, automatic estimation of heading date of paddy rice is highly essential. In this experiment, deep learning classification models were used to classify two difference categories of rice (vegetative and reproductive stage) based on the panicle initiation of paddy field. Specifically, the dataset includes 444 slanted view images belonging to two categories and was then expanded to include 1,497 images via IMGAUG data augmentation technique. We adopt two transfer learning strategies: (First, used transferring model weights already trained on ImageNet to six classification network models: VGGNet, ResNet, DenseNet, InceptionV3, Xception and MobileNet, Second, fine-tuned some layers of the network according to our dataset). After training the CNN model, we used several evaluation metrics commonly used for classification tasks, including Accuracy, Precision, Recall, and F1-score. In addition, GradCAM was used to generate visual explanations for each image patch. Experimental results showed that the InceptionV3 is the best performing model in terms of the accuracy, average recall, precision, and F1-score. The fine-tuned InceptionV3 model achieved an overall classification accuracy of 0.95 with a high F1-score of 0.95. Our CNN model also represented the change of rice heading date under different date of transplanting. This study demonstrated that image based deep learning model can reliably be used as an automatic monitoring system to detect the heading date of rice crops using CCTV camera.

• 가정과삶의질연구
• /
• 제23권1호
• /
• pp.209-222
• /
• 2005

This study was conducted to investigate the effect of interview techniques on memory accuracy and suggestibility of preschool children. Forty-five preschool children participated in a magic show(target event) and 1 week later, all children received suggestive interview in two conditions(language condition & drawing condition). Another 1 week later, all children's recall contents of the magic show was assessed. During suggestive interview, children in drawing condition show more 'acception' response than children in language condition, and children in the question condition show less 'remember' response than children in drawing condition. In second interview children reported more words, and specially ones in language condition report more suggested words than ones in drawing condition. Finally, children's recalls were more accurate on controled informations of the event than on suggestive.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2020년도 추계학술발표대회
• /
• pp.920-923
• /
• 2020

사이버 공격이 더욱 지능화됨에 따라 기존의 침입 탐지 시스템(Intrusion Detection System)은 기존의 저장된 패턴에서 벗어난 지능형 공격을 탐지하기에 적절하지 않다. 딥러닝(Deep Learning) 기반 침입 탐지는 새로운 탐지 규칙을 생성하는데 적절하다. 그 이유는 딥러닝은 데이터 학습을 통해 새로운 침입 규칙을 자체적으로 생성하기 때문이다. 침입 탐지 시스템 데이터 세트는 가장 널리 사용되는 KDD99 데이터와 LID-DS(Leipzig Intrusion Detection-Data Set)를 사용했다. 본 논문에서는 1차원 벡터를 이미지로 변환하고 CNN(Convolutional Neural Network)을 적용하여 두 데이터 세트에 대한 성능을 실험했다. 평가를 위해 Accuracy, Precision, Recall 및 F1-Score 지표를 측정했다. 그 결과 LID-DS 데이터 세트의 Accuracy가 KDD99 데이터 세트의 Accuracy 보다 약 8% 높은 것을 확인했다. 또한, 1차원 벡터에 대한 데이터를 Kibana를 사용하여 데이터를 시각화하여 대용량 데이터를 한눈에 보기 어려운 단점을 해결하는 방법을 제안한다.

• International Journal of Computer Science & Network Security
• /
• 제24권3호
• /
• pp.107-112
• /
• 2024

Nowadays the deadly virus famous as COVID-19 spread all over the world starts from the Wuhan China in 2019. This disease COVID-19 Virus effect millions of people in very short time. There are so many symptoms of COVID19 perhaps the Identification of a person infected with COVID-19 virus is really a difficult task. Moreover it's a challenging task to identify whether a person or individual have covid test positive or negative. We are developing a framework in which we used machine learning techniques..The proposed method uses DecisionTree, KNearestNeighbors, GaussianNB, LogisticRegression, BernoulliNB , RandomForest , Machine Learning methods as the classifier for diagnosis of covid ,however, 5-fold and 10-fold cross-validations were applied through the classification process. The experimental results showed that the best accuracy obtained from Decision Tree classifiers. The data preprocessing techniques have been applied for improving the classification performance. Recall, accuracy, precision, and F-score metrics were used to evaluate the classification performance. In future we will improve model accuracy more than we achieved now that is 93 percent by applying different techniques

• Journal of Nutrition and Health
• /
• 제53권5호
• /
• pp.476-487
• /
• 2020

본 연구는 만 65세 이상의 여자 노인 23명을 대상으로 이중표식수법을 이용하여 측정한 에너지소비량을 기준으로 에너지섭취량을 조사하는 24시간 회상법의 정확도를 평가하였으며, 그 결과는 다음과 같다. 연구 대상자 평균 연령은 70.3 ±3.3세, 신장 및 체중은 각각 153.0 ± 5.9 cm와 56.0 ± 8.0 kg, 체질량지수 (BMI)는 23.9 ± 2.8 kg/㎡이었다. 24시간 회상법으로 산출된 총 에너지섭취량 (TEI)과 이중표식수법으로 측정된 총에너지소비량 (TEE_DLW) 간의 피어슨 상관계수는 r = 0.482로 두 값 간에 의미 있는 양의 상관성 (p < 0.05)을 보여주었다. 그러나 24시간 회상법으로 조사된 3일간의 평균 에너지섭취량 (1,489.6 ± 211.1 kcal/day)은 이중표식수법으로 측정된 총에너지소비량 (2,023.5 ± 234.9 kcal/day)보다 -533.9 ± 228.0 kcal/day만큼 과소보고 되었으며, 두 값 간에 유의한 차이가 있었다 (p < 0.001). 총에너지섭취량과 총에너지소비량간의 과소보고율은 -25.9% ± 10.5%로 나타났다. Bland-Altman 방법으로 총에너지섭취량과 총에너지소비량간의 일치도 평가 결과로 두 값 일치 한계의 범위가 -980.8 kcal/day에서 -86.9 kcal/day로 음의 값으로 치우쳐 나타났다. 본 연구 결과에 따르면 에너지섭취량을 조사하는 24시간 회상법을 여자노인에게 적용시, 과소보고율이 -25.9%로 높았고, 에너지섭취량을 정확하게 예측한 비율 (오차범위 ± 10% 이내)도 8.7%로 낮았다. 따라서 여자노인을 대상으로 24시간 회상법으로 에너지섭취량을 조사하고자 하는 경우, 남자와는 다른 별도의 접근법이 강구되어야 할 것이다. 성별이 24시간 회상법의 정확도에 영향을 미치는 주요한 요인으로 보고된 만큼 향후 연구 대상자 수를 증가하여 더 다양한 연령에서 성별에 따른 차이를 평가하는 지속적인 연구가 필요하다고 사료된다.

• 대한지역사회영양학회지
• /
• 제23권6호
• /
• pp.516-524
• /
• 2018

Objectives: This study assessed the accuracy of the 24-hour diet recall method for estimating the energy intake of elderly men using the doubly labeled water as a reference method. Methods: Seventeen subjects (mean age $72.5{\pm}3.9years$), who maintained the same body weight during the two weeks study period, were included in this study. Three 24-hour diet recalls (two weekdays and one weekend) were obtained over a 14 day period to estimate the mean energy intake. The total energy expenditure was measured over the same 14 days using the doubly labeled water method. The total energy intake and total energy expenditure were compared by paired t-test. Results: The total energy intake from the 24-hour diet recalls method was $2536.7{\pm}350.6kcal/day$, and the total energy expenditure from the doubly labeled water method was $2659.8{\pm}306.8kcal/day$. The total energy intake was slightly under-reported by $-123.2{\pm}260.8kcal/day$ (-4.4%). On the other hand, no significant difference was observed between the total energy intake and total energy expenditure of the subjects (p=0.069). The percentage of accurate predictions was 64.7%. The correlation between the total energy intake and total energy expenditure was statistically significant (r=0.697, p<0.005). Conclusions: The present study supports the use of the 24-hour diet recall method to estimate the mean energy intake in elderly men group. More studies are needed to assess the validity of 24-hour diet recall method in other population groups, including elderly women, adults and children.

• International journal of advanced smart convergence
• /
• 제8권3호
• /
• pp.95-101
• /
• 2019

Recently, there have been many issues related to text recognition using Tesseract. One of these issues is that the text recognition accuracy is significantly lower for smaller fonts. Tesseract extracts text by creating an outline with direction in the image. By searching the Tesseract database, template matching with characters with similar feature points is used to select the character with the lowest error. Because of the poor text extraction, the recognition accuracy is lowerd. In this paper, we compared text recognition accuracy after applying various super-resolution methods to smaller text images and experimented with how the recognition accuracy varies for various image size. In order to recognize small Korean text images, we have used super-resolution algorithms based on deep learning models such as SRCNN, ESRCNN, DSRCNN, and DCSCN. The dataset for training and testing consisted of Korean-based scanned images. The images was resized from 0.5 times to 0.8 times with 12pt font size. The experiment was performed on x0.5 resized images, and the experimental result showed that DCSCN super-resolution is the most efficient method to reduce precision error rate by 7.8%, and reduce the recall error rate by 8.4%. The experimental results have demonstrated that the accuracy of text recognition for smaller Korean fonts can be improved by adding super-resolution methods to the OCR preprocessing module.