• 한국융합학회논문지
• /
• 제10권9호
• /
• pp.23-29
• /
• 2019

본 연구에서는 영상의 분위기를 심층 합성곱 신경망을 통해 8 가지로 분류하고, 이에 맞는 배경 음악을 적용하여 동영상을 자동적으로 생성하였다. 수집된 이미지 데이터를 바탕으로 다층퍼셉트론을 사용하여 분류 모델을 학습한다. 이를 활용하여 다중 클래스 분류를 통해 동영상 생성에 사용할 이미지의 분위기를 예측하며, 미리 분류된 음악을 매칭시켜 동영상을 생성한다. 10겹 교차 검증의 결과, 72.4%의 정확도를 얻을 수 있었고, 실제 영상에 대한 실험에서 64%의 오차 행렬 정확도를 얻을 수 있었다. 오답의 경우, 주변의 비슷한 분위기로 분류하여 동영상에서 나오는 음악과 크게 위화감이 없음을 확인하였다.

• 한국물환경학회지
• /
• 제39권1호
• /
• pp.1-8
• /
• 2023

Monitoring and prediction of water quality are essential for effective river pollution prevention and water quality management. In this study, a multi-classification model was developed to predict chlorophyll-a (Chl-a) level in rivers. A model was developed using CatBoost, a novel ensemble machine learning algorithm. The model was developed using hourly field monitoring data collected from January 1 to December 31, 2015. For model development, chl-a was classified into class 1 (Chl-a≤10 ㎍/L), class 2 (10<Chl-a≤50 ㎍/L), and class 3 (Chl-a>50 ㎍/L), where the number of data used for the model training were 27,192, 11,031, and 511, respectively. The macro averages of precision, recall, and F1-score for the three classes were 0.58, 0.58, and 0.58, respectively, while the weighted averages were 0.89, 0.90, and 0.89, for precision, recall, and F1-score, respectively. The model showed relatively poor performance for class 3 where the number of observations was much smaller compared to the other two classes. The imbalance of data distribution among the three classes was resolved by using the synthetic minority over-sampling technique (SMOTE) algorithm, where the number of data used for model training was evenly distributed as 26,868 for each class. The model performance was improved with the macro averages of precision, rcall, and F1-score of the three classes as 0.58, 0.70, and 0.59, respectively, while the weighted averages were 0.88, 0.84, and 0.86 after SMOTE application.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제52권9호
• /
• pp.540-546
• /
• 2003

A fast pattern classification algorithm with Cellular Nonlinear Network-based dynamic programming is proposed. The Cellular Nonlinear Networks is an analog parallel processing architecture and the dynamic programing is an efficient computation algorithm for optimization problem. Combining merits of these two technologies, fast pattern classification with optimization is formed. On such CNN-based dynamic programming, if exemplars and test patterns are presented as the goals and the start positions, respectively, the optimal paths from test patterns to their closest exemplars are found. Such paths are utilized as aggregating keys for the classification. The algorithm is similar to the conventional neural network-based method in the use of the exemplar patterns but quite different in the use of the most likely path finding of the dynamic programming. The pattern classification is performed well regardless of degree of the nonlinearity in class borders.

• International Journal of Computer Science & Network Security
• /
• 제24권7호
• /
• pp.108-117
• /
• 2024

The selection and recommendation of a suitable job applicant from the pool of thousands of applications are often daunting jobs for an employer. The recommendation and selection process significantly increases the workload of the concerned department of an employer. Thus, Resume Classification System using the Natural Language Processing (NLP) and Machine Learning (ML) techniques could automate this tedious process and ease the job of an employer. Moreover, the automation of this process can significantly expedite and transparent the applicants' selection process with mere human involvement. Nevertheless, various Machine Learning approaches have been proposed to develop Resume Classification Systems. However, this study presents an automated NLP and ML-based system that classifies the Resumes according to job categories with performance guarantees. This study employs various ML algorithms and NLP techniques to measure the accuracy of Resume Classification Systems and proposes a solution with better accuracy and reliability in different settings. To demonstrate the significance of NLP & ML techniques for processing & classification of Resumes, the extracted features were tested on nine machine learning models Support Vector Machine - SVM (Linear, SGD, SVC & NuSVC), Naïve Bayes (Bernoulli, Multinomial & Gaussian), K-Nearest Neighbor (KNN) and Logistic Regression (LR). The Term-Frequency Inverse Document (TF-IDF) feature representation scheme proven suitable for Resume Classification Task. The developed models were evaluated using F-Score_M, Recall_M, Precission_M, and overall Accuracy. The experimental results indicate that using the One-Vs-Rest-Classification strategy for this multi-class Resume Classification task, the SVM class of Machine Learning algorithms performed better on the study dataset with over 96% overall accuracy. The promising results suggest that NLP & ML techniques employed in this study could be used for the Resume Classification task.

• 대한원격탐사학회지
• /
• 제29권6호
• /
• pp.611-619
• /
• 2013

Although the need of crop production has increased in Mongolia, crop cultivation is very limited because of the harsh climatic and topographic conditions. Crop lands are sparsely distributed with relatively small sizes and, therefore, it is difficult to survey the exact area of crop lands. The study aimed to find an easy and effective way of accurate classification to map crop lands in Mongolia using satellite images. To classify the crop lands over the study area in northern Mongolia, four classifications were carried out by using 1) Thematic Mapper (TM) image August 23, 2) TM image of July 6, 3) combined 12 bands of TM images of July and August, and 4) both TM images of July and August by layered classification. Wheat and potato are the major crop types and they show relatively high variation in crop conditions between July and August. On the other hands, other land cover types (forest, riparian vegetation, grassland, water and bare soil) do not show such difference between July and August. The results of four classifications clearly show that the use of multi-temporal images is essential to accurately classify the crop lands. The layered classification method, in which each class is separated by a subset of TM images, shows the highest classification accuracy (93.7%) of the crop lands. The classification accuracies are lower when we use only a single TM image of either July or August. Because of the different planting practice of potato and the growth condition of wheat, the spectral characteristics of potato and wheat cannot be fully separated from other cover types with TM image of either July or August. Further refinements on the spatial characteristics of existing crop lands may enhance the crop mapping method in Mongolia.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제10권12호
• /
• pp.555-560
• /
• 2021

최근 안면 피부 미용에 대한 사람들의 관심이 높아짐에 따라 딥 러닝을 활용한 안면 피부 미용을 위한 피부 질환 인식 연구가 진행되고 있다. 이러한 연구들은 여드름을 비롯한 다양한 피부 질환을 인식한다. 기존의 연구들은 단일 피부 질환만을 인식하지만, 안면에 발생하는 피부 질환은 더 다양하고 복합적으로 발생할 수 있다. 따라서 본 논문에서는 Inception-ResNet V2 모델을 활용하여 다중 레이블 분류 방법으로 여드름, 블랙헤드, 주근깨, 검버섯, 일반 피부, 화이트헤드에 관한 복합적인 피부 질환을 인식한다. 사용한 평가 지표 중 정확도는 98.8%, 해밍 손실은 0.003을 달성하였고, 단일 클래스별 정밀도, 재현율, F1-점수는 모두 96.6% 이상을 달성하였다.

• Structural Monitoring and Maintenance
• /
• 제2권3호
• /
• pp.181-197
• /
• 2015

Assessing the condition of paint on civil structures is an important but challenging and costly task, in particular when it comes to large and complex structures. Current practices of visual inspection are labour-intensive and time-consuming to perform. In addition, this task usually relies on the experience and subjective judgment of individual inspectors. In this study, hyperspectral imaging and classification techniques are proposed as a method to objectively assess the state of the paint on a civil or other structure. The ultimate objective of the work is to develop a technology that can provide precise and automatic grading of paint condition and assessment of degradation due to age or environmental factors. Towards this goal, we acquired hyperspectral images of steel surfaces located at long (mid-range) and short distances on the Sydney Harbour Bridge with an Acousto-Optics Tunable filter (AOTF) hyperspectral camera (consisting of 21 bands in the visible spectrum). We trained a multi-class Support Vector Machines (SVM) classifier to automatically assess the grading of the paint from hyperspectral signatures. Our results demonstrate that the classifier generates highly accurate assessment of the paint condition in comparison to the judgement of human experts.

• 센서학회지
• /
• 제24권1호
• /
• pp.47-53
• /
• 2015

Fall-related injuries in elderly people are a major health care problem. This paper introduces determination of fall direction before impact using support vector machine (SVM). Once a falling phase is detected, dynamic characteristic parameters measured by the accelerometer and gyroscope and then processed by a Kalman filter are used in the SVM to determine the fall directions, i.e., forward (F), backward (B), rightward (R), and leftward (L). This paper compares the determination sensitivities according to the selected parameters for the SVM (velocities, tilt angles, vs. accelerations) and sensor attachment locations (waist vs. chest) with regards to the binary classification (i.e., F vs. B and R vs. L) and the multi-class classification (i.e., F, B, R, vs. L). Based on the velocity of waist which was superior to other parameters, the SVM in the binary case achieved 100% sensitivities for both F vs. B and R vs. L, while the SVM in the multi-class case achieved the sensitivities of F 93.8%, B 91.3%, R 62.3%, and L 63.6%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권2호
• /
• pp.632-657
• /
• 2022

Detecting web attacks is a major challenge, and it is observed that the use of simple models leads to low sensitivity or high false positive problems. In this study, we aim to develop a robust two-stage deep learning based stacked ensemble web application firewall. Normal and abnormal classification is carried out in the first stage of the proposed WAF model. The classification process of the types of abnormal traffics is postponed to the second stage and carried out using an integrated stacked ensemble model. By this way, clients' requests can be served without time delay, and attack types can be detected with high sensitivity. In addition to the high accuracy of the proposed model, by using the statistical similarity and diversity analyses in the study, high generalization for the ensemble model is achieved. Within the study, a comprehensive, up-to-date, and robust multi-class web anomaly dataset named GAZI-HTTP is created in accordance with the real-world situations. The performance of the proposed WAF model is compared to state-of-the-art deep learning models and previous studies using the benchmark dataset. The proposed two-stage model achieved multi-class detection rates of 97.43% and 94.77% for GAZI-HTTP and ECML-PKDD, respectively.

• 제어로봇시스템학회논문지
• /
• 제15권4호
• /
• pp.372-378
• /
• 2009

For effective mobility control of unmanned ground vehicles in outdoor off-road environments, terrain cover classification technology using passive sensors is vital. This paper presents a novel method far terrain classification based on color and texture information of off-road images. It uses a neural network classifier and wavelet features. We exploit the wavelet mean and energy features extracted from multi-channel wavelet transformed images and also utilize the terrain class spatial coordinates of images to include additional features. By comparing the classification performance according to applied features, the experimental results show that the proposed algorithm has a promising result and potential possibilities for autonomous navigation.