• Journal of Information Processing Systems
• /
• 제20권2호
• /
• pp.215-225
• /
• 2024

With the rapid development of Internet of Things (IoT) and big data technology, a large amount of data will be generated during the operation of related industries. How to classify the generated data accurately has become the core of research on data mining and processing in IoT industry chain. This study constructs a classification model of IoT industry chain based on improved random forest algorithm and text analysis, aiming to achieve efficient and accurate classification of IoT industry chain big data by improving traditional algorithms. The accuracy, precision, recall, and AUC value size of the traditional Random Forest algorithm and the algorithm used in the paper are compared on different datasets. The experimental results show that the algorithm model used in this paper has better performance on different datasets, and the accuracy and recall performance on four datasets are better than the traditional algorithm, and the accuracy performance on two datasets, P-I Diabetes and Loan Default, is better than the random forest model, and its final data classification results are better. Through the construction of this model, we can accurately classify the massive data generated in the IoT industry chain, thus providing more research value for the data mining and processing technology of the IoT industry chain.

• International Journal of Computer Science & Network Security
• /
• 제21권9호
• /
• pp.179-184
• /
• 2021

In our study we performed a sentiments analysis from the images. For this purpose, we used 153 images that contain: people, animals, buildings, landscapes, cakes and objects that we divided into two categories: images that suggesting a positive or a negative emotion. In order to classify the images using the two categories, we created two models. The SAI-G model was created with Google's AutoML Vision service. The SAI-C model was created on the Clarifai platform. The data were labeled in a preprocessing stage, and for the SAI-C model we created the concepts POSITIVE (POZITIV) AND NEGATIVE (NEGATIV). In order to evaluate the performances of the two models, we used a series of evaluation metrics such as: Precision, Recall, ROC (Receiver Operating Characteristic) curve, Precision-Recall curve, Confusion Matrix, Accuracy Score and Average precision. Precision and Recall for the SAI-G model is 0.875, at a confidence threshold of 0.5, while for the SAI-C model we obtained much lower scores, respectively Precision = 0.727 and Recall = 0.571 for the same confidence threshold. The results indicate a lower classification performance of the SAI-C model compared to the SAI-G model. The exception is the value of Precision for the POSITIVE concept, which is 1,000.

• Journal of Community Nutrition
• /
• 제4권3호
• /
• pp.143-150
• /
• 2002

The objective of this study is to establish outcome validity for three instruments that assess whether subjects met recommendations for daily servings of fruits and vegetables for“Stage of Change”research. A convenience sample of 294 college students was recruited from introductory nutrition classes at Michigan State University in the United States. To measure servings of fruits and vegetables, separately, three types of methods (self-rated intake, 24-hour recall and food frequency) were used in comparison to the servings from a three-day food record, the criterion used. The outcome validity was assessed based on whether or not at least two servings of fruits and three servings of vegetables were reported. Validity was assessed by sensitivity, to measure the ability to detect low intakes, and by specificity, to measure ability to detect adequate intakes. Cohen's kappa was used as well to examine the agreement between the three methods and a three-day food record. The results showed, for fruits, the 24-hour recall had the best agreement (recall 0.54, self-rated : 0.31, food frequency : 0.29) with a three-day food record for servings consumed by people in pre-action or post-action stages. Sensitivity for fruit was also best using a recall (0.81). For vegetables, however, all three methods had low agreement (food frequency : 0.27, recall : 0.21, self-rated : 0.17) with the results of the three-day food record. Self-rated intakes for vegetables had the best sensitivity (0.66) and the food frequency had the best specificity (0.73). Therefore, researcher can use the 24-hour recall method to identify people who consume inadequate servings of fruit. To detect adequate vegetable intake, the food frequency was best of the three methods. Accuracy may be improved by probing for vegetables in mixed dishes and on sandwiches. (J Community Nutrition 4(3)： 143∼150, 2002)

• Clinical Nutrition Research
• /
• 제12권2호
• /
• pp.138-153
• /
• 2023

The prevalence of metabolic syndrome (MetS) and its cost are increasing due to lifestyle changes and aging. This study aimed to develop a deep neural network model for prediction and classification of MetS according to nutrient intake and other MetS-related factors. This study included 17,848 individuals aged 40-69 years from the Korea National Health and Nutrition Examination Survey (2013-2018). We set MetS (3-5 risk factors present) as the dependent variable and 52 MetS-related factors and nutrient intake variables as independent variables in a regression analysis. The analysis compared and analyzed model accuracy, precision and recall by conventional logistic regression, machine learning-based logistic regression and deep learning. The accuracy of train data was 81.2089, and the accuracy of test data was 81.1485 in a MetS classification and prediction model developed in this study. These accuracies were higher than those obtained by conventional logistic regression or machine learning-based logistic regression. Precision, recall, and F1-score also showed the high accuracy in the deep learning model. Blood alanine aminotransferase (β = 12.2035) level showed the highest regression coefficient followed by blood aspartate aminotransferase (β = 11.771) level, waist circumference (β = 10.8555), body mass index (β = 10.3842), and blood glycated hemoglobin (β = 10.1802) level. Fats (cholesterol [β = -2.0545] and saturated fatty acid [β = -2.0483]) showed high regression coefficients among nutrient intakes. The deep learning model for classification and prediction on MetS showed a higher accuracy than conventional logistic regression or machine learning-based logistic regression.

• 한국정보통신학회논문지
• /
• 제21권2호
• /
• pp.277-284
• /
• 2017

사용자가 좋아할만한 콘텐츠를 정확하게 추천하는 것은 추천 시스템에서 매우 중요한 요소 중 하나이다. 원치 않는 콘텐츠를 추천하거나, 원하는 것을 추천하지 않는 것은 사용자 만족도 측면에서 안 좋은 영향을 끼친다. 본 연구에서는 콘텐츠의 정확한 추천을 위해 사용자 군집 기반 추천 시스템을 제안한다. 제안하는 알고리즘에서 사용자들의 실제 선호도 점수와 피어슨 상관 계수를 기반으로 사용자들을 여러 군집으로 나눈다. 이 후, 특정 사용자에게 어떤 콘텐츠의 추천 여부 결정은, 같은 군집 내에 있는 다른 사용자들의 해당 콘텐츠의 실제 선호도 점수를 근거로 정한다. 제안하는 알고리즘은 군집화를 사용하지 않는 아이템 기반 협력 필터링 알고리즘보다 정밀도, 재현율, F1 스코어와 같은 추천 정확도에 있어서 의미 있는 성능 향상을 보인다.

• Journal of Information Processing Systems
• /
• 제14권5호
• /
• pp.1114-1135
• /
• 2018

A multiple classification system based on a new boosting technique has been approached utilizing different biometric traits, that is, color face, iris and eye along with fingerprints of right and left hands, handwriting, palm-print, gait (silhouettes) and wrist-vein for person authentication. The images of different biometric traits were taken from different standard databases such as FEI, UTIRIS, CASIA, IAM and CIE. This system is comprised of three different super-classifiers to individually perform person identification. The individual classifiers corresponding to each super-classifier in their turn identify different biometric features and their conclusions are integrated together in their respective super-classifiers. The decisions from individual super-classifiers are integrated together through a mega-super-classifier to perform the final conclusion using programming based boosting. The mega-super-classifier system using different super-classifiers in a compact form is more reliable than single classifier or even single super-classifier system. The system has been evaluated with accuracy, precision, recall and F-score metrics through holdout method and confusion matrix for each of the single classifiers, super-classifiers and finally the mega-super-classifier. The different performance evaluations are appreciable. Also the learning and the recognition time is fairly reasonable. Thereby making the system is efficient and effective.

• 정보과학회 논문지
• /
• 제42권1호
• /
• pp.68-75
• /
• 2015

본 논문에서는 띄어쓰기가 전혀 되어 있지 않은 문자열을 입력 받아 말뭉치에서 추출한 어절 정보를 이용하여 자동 띄어쓰기를 해 주는 방법론을 제안한다. 형태소 분석기도 사용되나 오류 수정이라는 제한적인 용도로만 사용된다. 성능 평가를 위해 1,000만 어절 규모의 세종 말뭉치에서 순수 한글 585만 어절을 발췌하여 10 개의 세트로 나누고 10 배수 교차 검증을 실시한 결과 98.06%의 음절 정확도와 94.15%의 어절 재현율을 얻었다. 또한, 개인용 컴퓨터에서 초당 25만 어절, 1.8 MB의 문서를 처리할 수 있을 정도로 빠르다. 제안된 방법의 정확도나 재현율은 어절 사전의 크기에 영향을 받기 때문에 보다 큰 말뭉치로 어절 사전을 구축하면 성능이 더욱 향상될 것으로 기대된다.

• 센서학회지
• /
• 제28권2호
• /
• pp.94-100
• /
• 2019

Drivable area detection is a major task in advanced driver assistance systems. For drivable area detection, several studies have proposed vision-sensor-based approaches. However, conventional drivable area detection methods that use vision sensors are not suitable for environments with changes in road elevation. In addition, if the boundary between the road and vegetation is not clear, judging a vegetation area as a drivable area becomes a problem. Therefore, this study proposes an accurate method of detecting drivable areas in environments in which road elevations change and vegetation exists. Experimental results show that when compared to the conventional method, the proposed method improves the average accuracy and recall of drivable area detection on the KITTI vision benchmark suite by 3.42%p and 8.37%p, respectively. In addition, when the proposed vegetation area removal method is applied, the average accuracy and recall are further improved by 6.43%p and 9.68%p, respectively.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제18권2호
• /
• pp.494-510
• /
• 2024

Internet users are exposed to sophisticated cyberattacks that intrusion detection systems have difficulty detecting. Therefore, research is increasing on intrusion detection methods that use artificial intelligence technology for detecting novel cyberattacks. Unsupervised learning-based methods are being researched that learn only from normal data and detect abnormal behaviors by finding patterns. This study developed an anomaly-detection method based on unsupervised machines and deep learning for a network intrusion detection system (NIDS). We present a hybrid anomaly detection approach based on unsupervised learning techniques using the autoencoder (AE), Isolation Forest (IF), and Local Outlier Factor (LOF) algorithms. An oversampling approach that increased the detection rate was also examined. A hybrid approach that combined deep learning algorithms and traditional machine learning algorithms was highly effective in setting the thresholds for anomalies without subjective human judgment. It achieved precision and recall rates respectively of 88.2% and 92.8% when combining two AEs, IF, and LOF while using an oversampling approach to learn more unknown normal data improved the detection accuracy. This approach achieved precision and recall rates respectively of 88.2% and 94.6%, further improving the detection accuracy compared with the hybrid method. Therefore, in NIDS the proposed approach provides high reliability for detecting cyberattacks.

• 문화기술의 융합
• /
• 제10권4호
• /
• pp.575-580
• /
• 2024

소셜 미디어의 대중화로 지능화된 피싱 공격을 방어하기 위해 접근하고자 하는 사이트의 상태(정상/피싱)를 판별하는 것이 필요하다. 본 연구에서는 머신러닝 기반 분류 모델을 통해 사이트의 정상/피싱 여부를 예측하는 모델을 제안한다. 첫째, 'URL'에 대한 정보를 수집하여 수치 데이터로 변환한 후, 이상치를 제거한다. 둘째, 변수들 간의 상관관계 및 독립성을 파악하기 위해 VIF(Variance Inflation Factors)를 적용한다. 셋째, 머신러닝 기반 분류 모델을 활용하여 피싱 사이트 탐지 모델을 개발하고, 이를 통해 사이트의 상태를 예측한다. 분류 모델 중 랜덤 포레스트(Random Forest)의 성능이 가장 우수했으며, 테스트 데이터에서 정밀도(Precision) 93.74%, 재현율(Recall) 92.26%, 정확도(Accuracy) 93.14%를 보였다. 향후 이 연구는 다방면의 피싱 범죄 탐지에 적용할 수 있을 것으로 기대된다.