Title: The Development of a Unified Predictive Model to Predict Closed Price for a Variety of Cryptocurrencies
Cover Date: 2024-01-01
Cover Display Date: 2024
DOI: 10.1109/RI2C64012.2024.10784451
Description: Predicting the closing price of a cryptocurrency typically involves utilizing an individual predictive model for each particular cryptocurrency. However, a unified model has the potential to predict the closing price of many cryptocurrencies, providing convenience and facilitating comparative research. Thus, this study utilized feature-based data fusion to integrate historical data of cryptocurrencies from the same time period using a data fusion, called as the average method. The predictive models were developed using two machine learning algorithms, i.e. Support Vector Regression (SVR) with RBF kernel and Random Forest (RF). This study used three types of cryptocurrencies: Bitcoin (BTC), Ethereum (ETH), and Litecoin (LTC). After evaluating the effectiveness of several predictive models utilizing MAE, RMSE, and MAPE for short-term (5-day) predictions, it was found that the unified predictive model yielded similar outcomes to the models specifically d eveloped f or the given particular cryptocurrency. This method offers the ability to help investors in identifying general price movements among multiple cryptocurrencies and decreasing the amount of time needed for observations. However, the unified p redictive models developed by the random forest algorithm surpass other models in successfully predicting the short-term (5-day) closing prices of cryptocurrencies.
Citations: 0
Aggregation Type: Conference Proceeding
-------------------


Title: Logistic Regression-based Sentiment Classification Approach for Identifying Undergraduate Student Sentiments in a Course Studied
Cover Date: 2024-01-01
Cover Display Date: 2024
DOI: 10.1109/RI2C64012.2024.10784409
Description: This study aimed to utilize sentiment classification to ascertain the sentiment of undergraduate students towards the course they have studied. This case study specifically examines the character design course given by the Department of Creative Media, Faculty of Informatics, Mahasarakham University. Unfortunately, our data collection exhibits an imbalance between the positive class and the negative class, with a greater likelihood for the data belong to the positive class. This issue has the potential to result in sentiment classifiers that generate subpar outcomes. Consequently, this issue was also addressed in this study. To develop the binary-based sentiment classifiers, logistic regression methods were employed, specifically traditional logistic regression and logistic regression with class weights. The term weighting scheme is tf-idf, The results were determined to be satisfactory after being evaluated using the F1 score and AUC. However, it was found that the sentiment classifiers generated by L R with class weights showed better results in terms of average F1 score and AUC compared to the sentiment classifiers developed using traditional LR. The overall improvements of F1 score and AUC were 14.51 % and 13.50%, respectively.
Citations: 0
Aggregation Type: Conference Proceeding
-------------------


Title: DYNAMIC FINGERSPELLING RECOGNITION FROM VIDEO USING DEEP LEARNING APPROACH: FROM DETECTION TO RECOGNITION
Cover Date: 2022-09-01
Cover Display Date: September 2022
DOI: 10.24507/icicelb.13.09.949
Description: The World Health Organization found that more than 34 million people suf-fer from hearing loss and these people need to use sign language to communicate. Hence, the sign language recognition system is proposed to communicate with hearing loss people and others. In this paper, we aim to propose an end-to-end system to recognize the dynamic Thai fingerspelling from video. The proposed system includes two main processes. First, we use the YOLOv5 algorithm for the human detection task. Subsequently, a uni-form distribution method is proposed to select the robust frames before applying robust frames to the detection algorithm. Second, we propose dynamic fingerspelling recognition that consists of two deep learning architectures: convolutional neural network (CNN) and long short-term memory (LSTM). We then combine CNN and LSTM, called CNN-LSTM architecture, followed by the recognition block. The recognition block comprises dropout, global average pooling, and softmax layers. For the CNN architectures, we evaluated three CNNs: MobileNetV2, ResNet50, and DenseNet201. We found that the proposed ResNet50-LSTM architecture achieved an accuracy of 88.42% on the test set of the dynamic Thai fingerspelling dataset and also prevented the overfitting problem.
Citations: 2
Aggregation Type: Journal
-------------------


Title: AN END-TO-END THAI FINGERSPELLING RECOGNITION FRAMEWORK WITH DEEP CONVOLUTIONAL NEURAL NETWORKS
Cover Date: 2022-05-01
Cover Display Date: May 2022
DOI: 10.24507/icicel.16.05.529
Description: The WHO reports that approximately 34 million people worldwide experience deafness and hearing loss. In 2050, these will increase to affect 900 million people. It is essential to communicate with the hearing impaired in hand sign language. This paper proposes an end-to-end fingerspelling recognition framework of the Thai sign language based on deep convolutional neural networks (CNNs). We divided our framework into two processes. In the first process, we focus on the detection of hands using the YOLOv3 objection detection framework. In the second process, we propose using five CNN architectures, MobileNetV2, DenseNet121, InceptionResNetV2, NASNetMobile, and EfficientNetB2, to create the most robust model that provides high recognition performance. Hence, we evaluated the proposed framework to detect and recognize three Thai fingerspelling (TFS) datasets: TFS, KKU-TFS, and Unseen-TFS. We found that YOLOv3 showed a high precision value on the TFS dataset. However, the worst performance was found with KKU-TFS and Unseen-TFS datasets. Also, our proposed framework could not detect hands from only one image on the KKU-TFS and Unseen-TFS datasets. Therefore, we also examined the CNN architectures to recognize the 1-stage Thai fingerspelling images. The experimental results showed that DenseNet121 obtained an accuracy of 93.99% on the TFS dataset and 90.40% on the KKU-TFS dataset.
Citations: 8
Aggregation Type: Journal
-------------------


Title: Plant Leaf Image Recognition using Multiple-grid Based Local Descriptor and Dimensionality Reduction Approach
Cover Date: 2020-03-19
Cover Display Date: 19 March 2020
DOI: 10.1145/3388176.3388180
Description: The identification process of plant species is one of the significant and challenging problems. In this research area, many researchers have focused on identifying the plant leaf images because the leaves of a plant are found almost all year round. The achieve method of the plant leaf image recognition is based on unique extraction features from the plant leaf and using the well-known machine learnings as a classification method. As a result, recognition accuracy was often not very high. In order to improve recognition accuracy, we proposed a multiple grids technique based on the local descriptors and dimensionality reduction. Firstly, we divided the plant leaf image according to grid size and calculated the local descriptors from each grid. Secondly, the dimensionality reduction is proposed to transform and decrease the correlated variables of the feature vector. Finally, the feature vector with a relatively low-dimensional is transferred to the machine learning techniques, which are the support vector machine and multi-layer perceptron algorithms. We have evaluated and compared the proposed algorithm with the bag of visual words method and the deep convolutional neural network (including AlexNet and GoogLeNet architectures) on the Folio leaf image dataset. The experiments show that the proposed algorithm has improved and obtained very high accuracy on plant leaf image recognition.
Citations: 5
Aggregation Type: Conference Proceeding
-------------------