• 한국멀티미디어학회논문지
• /
• 제25권2호
• /
• pp.215-220
• /
• 2022

Emotion recognition is one of the most important and challenging areas of computer vision. Nowadays, many studies on emotion recognition were conducted and the performance of models is also improving. but, more research is needed on emotion recognition and sentiment analysis of video viewers. In this paper, we propose an emotion analysis system the includes a sentiment analysis model and an interest prediction model. We analyzed the emotional patterns of people watching popular and unpopular videos and predicted the level of interest using the emotion analysis system. Experimental results showed that certain emotions were strongly related to the popularity of videos and the interest prediction model had high accuracy in predicting the level of interest.

• 한국컴퓨터정보학회논문지
• /
• 제27권8호
• /
• pp.61-68
• /
• 2022

본 논문에서는 딥러닝 모델 기반 보행자의 GPS 경로를 예측하는 시스템을 제안한다. 보행자 경로 예측은 보행자의 위험 및 충돌 상황들을 알림을 통해 방지할 수 있으며, 다양한 마케팅 등 비즈니스 면에서도 영향을 끼치는 연구이다. 또한 보행자 뿐 아니라 많은 각광을 받고 있는 무인 이동수단의 경로 예측에도 활용될 수 있다. 다양한 경로 예측 방식들 중 본 논문은 GPS 데이터를 활용하여 경로를 예측하는 연구이다. 시계열 데이터인 보행자의 GPS 경로를 학습하여 다음 경로를 예측하도록 하는 딥러닝 모델 기반 연구이다. 본 논문에서는 보행자의 GPS 경로를 딥러닝 모델이 학습할 수 있도록하는 데이터 셋 구성 방식을 제시하였으며, 예측 범위에 큰 제약이 없는 경로 예측 딥러닝 모델을 제안한다. 본 연구의 경로 예측 딥러닝 모델에 적합한 파라메터들을 제시하였으며, 우수한 예측 성능을 보이는 결과를 제시한다.

• 한국컴퓨터정보학회논문지
• /
• 제27권5호
• /
• pp.29-36
• /
• 2022

본 논문에서는 DACON에서 제공하는 데이터셋을 기반으로 한 효과적인 농산물 가격 예측 모델을 제안한다. 이 모델은 XGBoost와 CatBoost 이며 Gradient Boosting 계열의 알고리즘으로써 기존의 Logistic Regression과 Random Forest보다 평균정확도 및 수행시간이 우수하다. 이러한 장점들을 기반으로 농산물의 이전 가격들을 기반으로 1주, 2주, 4주뒤 가격을 예측하는 머신러닝 모델을 설계한다. XGBoost 모델은 회귀 방식의 모델링인 XGBoost Regressor 라이브러리를 사용하여 하이퍼 파라미터를 조정함으로써 가장 우수한 성능을 도출할 수 있다. CatBoost 모델은 CatBoost Regressor를 사용하여 모델을 구현한다. 구현한 모델은 DACON에서 제공하는 API를 이용하여 검증하고, 모델 별 성능평가를 실시한다. XGBoost는 자체적인 과적합 규제를 진행하기 때문에 적은 데이터셋에도 불구하고 우수한 성능을 도출하지만, 학습시간, 예측시간 등 시간적인 성능 면에서는 LGBM보다 성능이 낮다는 것을 알 수 있었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권5호
• /
• pp.1431-1445
• /
• 2022

We construct a deep neural network model named ECGResNet. This model can diagnosis diseases based on 12-lead ECG data of eight common cardiovascular diseases with a high accuracy. We chose the 16 Blocks of ResNet50 as the main body of the model and added the Squeeze-and-Excitation module to learn the data information between channels adaptively. We modified the first convolutional layer of ResNet50 which has a convolutional kernel of 7 to a superposition of convolutional kernels of 8 and 16 as our feature extraction method. This way allows the model to focus on the overall trend of the ECG signal while also noticing subtle changes. The model further improves the accuracy of cardiovascular and cerebrovascular disease classification by using a fully connected layer that integrates factors such as gender and age. The ECGResNet model adds Dropout layers to both the residual block and SE module of ResNet50, further avoiding the phenomenon of model overfitting. The model was eventually trained using a five-fold cross-validation and Flooding training method, with an accuracy of 95% on the test set and an F1-score of 0.841.We design a new deep neural network, innovate a multi-scale feature extraction method, and apply the SE module to extract features of ECG data.

• 대한치과교정학회지
• /
• 제52권2호
• /
• pp.112-122
• /
• 2022

Objective: This study aimed to present and evaluate a new deep learning model for determining cervical vertebral maturation (CVM) degree and growth spurts by analyzing lateral cephalometric radiographs. Methods: The study sample included 890 cephalograms. The images were classified into six cervical stages independently by two orthodontists. The images were also categorized into three degrees on the basis of the growth spurt: pre-pubertal, growth spurt, and post-pubertal. Subsequently, the samples were fed to a transfer learning model implemented using the Python programming language and PyTorch library. In the last step, the test set of cephalograms was randomly coded and provided to two new orthodontists in order to compare their diagnosis to the artificial intelligence (AI) model's performance using weighted kappa and Cohen's kappa statistical analyses. Results: The model's validation and test accuracy for the six-class CVM diagnosis were 62.63% and 61.62%, respectively. Moreover, the model's validation and test accuracy for the three-class classification were 75.76% and 82.83%, respectively. Furthermore, substantial agreements were observed between the two orthodontists as well as one of them and the AI model. Conclusions: The newly developed AI model had reasonable accuracy in detecting the CVM stage and high reliability in detecting the pubertal stage. However, its accuracy was still less than that of human observers. With further improvements in data quality, this model should be able to provide practical assistance to practicing dentists in the future.

• 대한토목학회논문집
• /
• 제29권2C호
• /
• pp.59-69
• /
• 2009

본 연구에서는 화강암, 편마암 등 결정질암 지역에서의 급경사지재해 발생지역 및 미발생지역에 대한 현장조사자료 및 토질시험자료를 토대로 의사결정나무모형을 이용한 급경사지재해 예측모델을 개발하였다. 선정된 급경사지재해 예측모델의 분리기준은 최상위부터 사면경사, 투수계수 및 간극비로 선정되었다. 그리고 이를 토대로 GIS기법을 이용한 국가 주요시설물 주변 급경사지 재해 예측프로그램 SHAPP ver 1.0을 개발하였다. 개발된 예측모델 및 예측프로그램을 검증하기 위하여 강릉시 주문진읍 일대의 현장조사결과와 대상현장에 대한 예측결과를 비교 검토하였다. 검토결과 실제 급경사지 재해가 발생된 구간과 급경사지재해 예측구간이 유사하게 일치하고 있는 것으로 나타났다. 추후 지속적인 연구를 통하여 급경사지재해 예측 결과에 대한 정확도를 높이고, 이를 실용화하여 범용적으로 사용이 가능하도록 할 예정이다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권8호
• /
• pp.3926-3945
• /
• 2018

In Jøsang's trust model, trust evaluation is obtained through operators, but there are problems with the mutuality and asymmetry of trust and the impact of event weight on trust evaluation. Trust evaluation is updated dynamically and continuously with time and the process of interactions, but it has not been reflected in Jøsang's model. Therefore, final trust evaluation is not accurate, and malicious fraud cannot be prevented effectively. This causes the success rate of interaction to be low. To solve these problems, a new dynamic trust model is proposed based on extended subjective logic (DTM-ESL). In DTM-ESL, the event weight and the mutuality of trust are fully considered, the original one-way trust relationship is extended to a two-way trust relationship, discounting and consensus operators are improved, and trust renewal is designed based on event weight. The viability and effectiveness of this new model are verified by simulation experiments.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권10호
• /
• pp.3668-3684
• /
• 2021

Video action recognition is widely used in video surveillance, behavior detection, human-computer interaction, medically assisted diagnosis and motion analysis. However, video action recognition can be disturbed by many factors, such as background, illumination and so on. Two-stream convolutional neural network uses the video spatial and temporal models to train separately, and performs fusion at the output end. The multi segment Two-Stream convolutional neural network model trains temporal and spatial information from the video to extract their feature and fuse them, then determine the category of video action. Google Xception model and the transfer learning is adopted in this paper, and the Xception model which trained on ImageNet is used as the initial weight. It greatly overcomes the problem of model underfitting caused by insufficient video behavior dataset, and it can effectively reduce the influence of various factors in the video. This way also greatly improves the accuracy and reduces the training time. What's more, to make up for the shortage of dataset, the kinetics400 dataset was used for pre-training, which greatly improved the accuracy of the model. In this applied research, through continuous efforts, the expected goal is basically achieved, and according to the study and research, the design of the original dual-flow model is improved.

• 스마트미디어저널
• /
• 제10권1호
• /
• pp.79-87
• /
• 2021

In recent years research on automatic font generation with machine learning mainly focus on using transformation-based methods, in comparison, generative model-based methods of font generation have received less attention. Transformation-based methods learn a mapping of the transformations from an existing input to a target. This makes them ambiguous because in some cases a single input reference may correspond to multiple possible outputs. In this work, we focus on font generation using the generative model-based methods which learn the buildup of the characters from noise-to-image. We propose a novel way to train a conditional generative deep neural model so that we can achieve font style control on the generated font images. Our research demonstrates how to generate new font images conditioned on both character class labels and character style labels when using the generative model-based methods. We achieve this by introducing a modified generator network which is given inputs noise, character class, and style, which help us to calculate losses separately for the character class labels and character style labels. We show that adding the character style vector on top of the character class vector separately gives the model rich information about the font and enables us to explicitly specify not only the character class but also the character style that we want the model to generate.

• Journal of Information Science Theory and Practice
• /
• 제12권2호
• /
• pp.64-78
• /
• 2024

Sentiment analysis is one of the promising approaches for developing a point of interest (POI) recommendation system. It uses natural language processing techniques that deploy expert insights from user-generated content such as reviews and feedback. By applying sentiment polarities (positive, negative, or neutral) associated with each POI, the recommendation system can suggest the most suitable POIs for specific users. The proposed study combines two models for POI recommendation. The first model uses bidirectional long short-term memory (BiLSTM) to predict sentiments and is trained on an election dataset. It is observed that the proposed model outperforms existing models in terms of accuracy (99.52%), precision (99.53%), recall (99.51%), and F1-score (99.52%). Then, this model is used on the Foursquare dataset to predict the class labels. Following this, user and POI embeddings are generated. The next model recommends the top POIs and corresponding coordinates to the user using the LSTM model. Filtered user interest and locations are used to recommend POIs from the Foursquare dataset. The results of our proposed model for the POI recommendation system using sentiment analysis are compared to several state-of-the-art approaches and are found quite affirmative regarding recall (48.5%) and precision (85%). The proposed system can be used for trip advice, group recommendations, and interesting place recommendations to specific users.