Papers by Chatklaw Jareanpon
Indonesian Journal of Electrical Engineering and Computer Science
Online deception has become a major problem affecting people, society, the economy, and national ... more Online deception has become a major problem affecting people, society, the economy, and national security. It is mostly done by spreading deceptive messages because message are quickly spread on social networks and are easily accessed by anyone. Detecting deceptive messages is challenging as the messages are unstructured, informal, and complex; this extends into Thai language messages. In this paper, various deep learning models are proposed to detect deceptive messages under two feature extraction trials. A balanced two-class dataset of deceptive and truthful Thai messages (n=2378) is collected from Facebook pages. Instance features are encoded using word embeddings (Thai2Fit) and one-hot encoding techniques. Five classification models, convolutional neural network (CNN), bidirectional long short-term memory (BiLSTM), bidirectional gated recurrent units (BiGRU), CNN-BiLSTM, and CNN-BiGRU, are proposed and evaluated upon the dataset with each feature extraction technique. The experi...
Current Applied Science and Technology
The Twitter platform is a popular tool that is widely used by researchers to collect data on user... more The Twitter platform is a popular tool that is widely used by researchers to collect data on users’ personal lives, feelings and emotions. These data sets can be further analyzed using text mining techniques to predict the disorder of depression. There are nine symptoms of depression that are classified by American Psychiatric Association using DSM-5 criteria. The symptoms can be difficult to identify effectively. The unweighted vote ensemble is not practical for multi-class data. Therefore, this research proposes the multi-objective optimization algorithms for depressive symptom prediction modeling (MOADSP) for the weighted voting ensemble, which can improve its effectiveness compared to the singer model. The objectives of this research were 1) to find the appropriate number of features; 2) to improve the weights of the prediction models based on the recall of the class for the ensemble; and 3) to compare the performance of the single, unweighted, and weighted voting ensemble model...
Indonesian Journal of Electrical Engineering and Computer Science, 2022
This work presents a technique for classifying X-ray images of the chest (CXR) by applying deep l... more This work presents a technique for classifying X-ray images of the chest (CXR) by applying deep learning-based techniques. The CXR will be classified into three different types, i.e. (i) normal, (ii) COVID-19, and (iii) pneumonia. The classification challenge is raised when the X-ray images of COVID-19 and pneumonia are subtle. The CXR images of the chest are first proceeded to be standardized and to improve the visual contrast of the images. Then, the classification is performed by applying a deep learningbased technique that binds two deep learning network architectures, i.e., convolution neural network (CNN) and long short-term memory (LSTM), to generate a hybrid model for the classification problem. The deep features of the images are extracted by CNN before the final classification is performed using LSTM. In addition to the hybrid models, this work explores the validity of image pre-processing methods that improve the quality of the images before the classification is performed. The experiments were conducted on a public image dataset. The experimental results demonstrate that the proposed technique provides promising results and is superior to the baseline techniques.
2018 International Workshop on Advanced Image Technology (IWAIT), 2018
The human listens to the music for the entertainment and emotional expression. Nowadays, the musi... more The human listens to the music for the entertainment and emotional expression. Nowadays, the music has many emotion, such as the love-music etc. The listener's playlist is made by manual. This paper proposes the automatic lyrics classification system using text mining technique by Naïve Bayes and Random forest based on the various weighting technique. The best of the accuracy is Naïve Bayes by the longest matching algorithm.
Lecture Notes in Computer Science, 2016
This paper proposed the development of an effective algorithm for marker detection from products ... more This paper proposed the development of an effective algorithm for marker detection from products for augmented reality by Speeded-Up Robust Features (SURF) algorithm that provided the efficiency in term of speed and accuracy. The SURF alorithm is consisted of 3 processes that are (1) feature extraction calculates the interested point and interested descriptions, (2) feature matching is that the correlation of all points is calculated from the distance of similarity of featuers, and (3) logo indentification is used to find the four corner point of the logo. This experiment is conducted from the recording video at 100 frames with resolution of 640 × 360 pixels and logo appeared all frames. Objects used in the experiment are consists of 3 shapes, cylindrical (can), rectangular (bag), and bottle. The logo template is divided into 5 sizes. The result of experiment found that the best detection accuracy of logo detection is from the size of 100 × 100 pixels. The accuracy of the region of marker detection compared with ground truth shows that the bag is equal to 94.96 %, can is equal to 93.99 %, and bottle is equal to 91.01 %, respectively. The difference of the logo is not affected with the computational time. However, the fast moving camera creates the blurred image and the reflection on the packaging creats a shiny surface which affects with the accuracy.
The “Kuu” is the pagoda located in Mahasarakham province, Thailand. This pagoda is the cultural t... more The “Kuu” is the pagoda located in Mahasarakham province, Thailand. This pagoda is the cultural travel place. The most well-known pagoda named “Santaluk” is located in Nadun district, Mahasarakham province Thailand built with the laterite stone in Cambodia art for collecting the image of Buddha and functioning as the hospital. The interactive tour representation system for pagoda is the database together with the reality application based on Web application that is able to represent the in novel way. User is able to feel like at in the physical travel place. This system is divided into 2 parts that are 1) Pagoda database developed with PHP, HTML together with MYSQL database, and 2) Interactive tour representation system that is collected the 360 degrees picture with in one shot from camera and special lens. The image processing technique is used for transform the picture to the 360 degrees dome where the user located in the center of the dome feels like in the physical travel place....
Received Jul 14, 2021 Revised Oct 20, 2021 Accepted Oct 27, 2021 A proactive cyber security plan ... more Received Jul 14, 2021 Revised Oct 20, 2021 Accepted Oct 27, 2021 A proactive cyber security plan to safeguard confidential information and privacy still lacks initiatives to avoid frequent harmful attacks. Cybersecurity professionals must possess ethical competence and prove worthy of overseeing valuable information for efficient decision‐making since ethical competence is fundamental for daily practice. There is a need to define what it means to be ethically competent in the era of IR4.0. The previous competence models still lack consideration of both artificial intelligence (AI) and emotional intelligence (EI) skills. AI brings new opportunities to cyber security organizations that focus on AI skills related to cognitive Intelligence or intelligent quotient (IQ). EI, which refers to emotional quotient (EQ), is a good predictor of ethical competence as it can perceive and express emotions precisely to facilitate thought to understand and manage emotions. However, practically, most ...
Lecture Notes in Computer Science, 2021
Lecture Notes in Computer Science, 2019
In 2021, Thailand will become an ageing society. The policy of the health of older people is a ch... more In 2021, Thailand will become an ageing society. The policy of the health of older people is a challenging task for the Thai government that has to be carefully planned. Stroke is the first leading cause of death of older people in Thailand. Knowing the risk factors for stroke will help people to prevent stroke. In this paper, features affecting stroke are studied based on machine learning. Factors and diseases occurring before stroke are studied as features to detect stroke and find affective factors of stroke. The detection of stroke is investigated based on learning classifiers, SVM, Naive Bayes, KNN, and decision tree. Moreover, Chi2 is adopted to find affective factors of stroke. The four most affective factors of stroke are focused to know the risk of stroke. From the study, we can see that the factors are more affective than the diseases for detecting stroke and decision tree is the best classifier. Decision tree gives 72.10% of accuracy and 74.29% of F-measure. The factors affecting stroke are smoking, alcohol, cholesterol, blood pressure, sex, exercise, and occupation. Moreover, we found that no smoking can avoid stroke. Drinking alcohol, abnormal cholesterol, and abnormal blood pressure raise the risk of a stroke.
Advances in Science, Technology and Engineering Systems Journal, 2020
This paper proposes a closed association rule generation technique to investigate the association... more This paper proposes a closed association rule generation technique to investigate the association patterns of diseases that are frequent co-occurrence. Diseases records of 5,000 patients are studied to find the association patterns of disease co-occurrence. The CHARM algorithm is adapted to find frequent diseases that can cover all-important patterns with a small number. Then the association patterns of disease co-occurrence are created in a form of association rules from the frequent diseases. The rules represent diseases associated with other diseases. Accuracy and prediction ratio are defined to evaluate the generated association patterns. From the experimental results, the generated association patterns give 79.76% of accuracy and 84.03% of prediction ratio although the number of generated association patterns is small. Moreover, the top-10 association patterns of disease cooccurrence are investigated. Besides, the 5 most frequent diseases are found to deeply study the other related diseases of them. From the investigation, we found that diabetes mellitus, metabolic disorders, and renal failure are highly related to hypertensive diseases with 88.81% of confidence. In addition, we found that influenza and pneumonia, plastic and other anemias are highly related to metabolic disorders.
Proceedings of The 3rd International Conference on Intelligent Systems and Image Processing 2015, 2015
Currently, social networking websites are widely used for communication among people. A multitude... more Currently, social networking websites are widely used for communication among people. A multitude of websites are available for sharing ideas and knowledge. Facebook is the most popular social networking website in the world. Thai people widely use status updates on Facebook for all types of discussion topics: political issues, religious issue, technology, education, etc. The discussion topics express their opinion in the form of text. The opinion is very important part of making any decision. This paper is proposed to mine opinion of Thai people about the current government revolution. The opinion was extracted from the Facebook statuses updates which were written in Thai language. Two feature extraction methods were implemented. First, the traditional pre-process of text mining was used for extracting the features. Second, positive and negative words were collected to construct a sentiment lexicon to extract features. Comparative experiments were performed among Naïve Bayes, SVM (Support Vector Machines), KNN (K-Nearest-Neighbor) and decision trees. From the experimental results, they were shown that KNN gave highest accuracy when using the lexicon for feature extraction.
Proceedings of The 3rd International Conference on Intelligent Systems and Image Processing 2015, 2015
This paper proposes the fast logo classification on product. The search space for logo classifica... more This paper proposes the fast logo classification on product. The search space for logo classification is reduced by production using Histogram of Oriented Gradients (HOG). The template matching and Speeded-Up Robust Features (SURF) algorithm are used to detect the logo, and in term of the computational time. The experimental results found that the HOG can be detected the area and product at 88.00% accuracy rate. The logo detection found that the advantage of template matching is simple, but the advantage of SURF algorithm is speed.
2013 IEEE International Conference on Computational Intelligence and Cybernetics (CYBERNETICSCOM), 2013
Currently pornographic image on the online world, teenagers and kids can visit easily. Which stim... more Currently pornographic image on the online world, teenagers and kids can visit easily. Which stimulate sexual desire. Resulting behavior of sexual abuse, enticing a child under the age of 15 years increased, cause problems getting pregnant and sexually transmitted diseases. Pornographic detection is essential to prevent to access through analyzing image content. Many researchers are interested in pornographic detection of nipple using extended Haar-like for extracting the features, color, texture and shape that are used for classification using various algorithms cascaded AdaBoost. However, this disadvantage is the templates for nipple which require a lot of training set and it consumes the time to detect a multiple possible position similar to nipples such as eyes and navel. This research proposed the novel algorithm without using templates for detecting the nipple. Our proposed creates the novel model based on ideal proportion detection. The result of this algorithm shows the high accuracy and reducing the computational time when compares with the existing method.
This paper is proposing the method for Thai handwritten character recognition. The methods are Ro... more This paper is proposing the method for Thai handwritten character recognition. The methods are Robust C-Prototype and Back-Propagation Neural Network. The objective of experimental is recognition on Thai handwritten character. This is the result of both methods to be appearing accuracy more than 85%.
Studies in Computational Intelligence, 2015
The Google street View (GSV) is used for many applications such as using for driver decision supp... more The Google street View (GSV) is used for many applications such as using for driver decision support system, collecting and merging the image for special purpose. The interactive system can be created by GSV. However, the most difficulty is the great image volumes. Moreover, for the reality, the map of interactive system ought to like with the physical place. This research proposed the management system created by the simple concept of database and designed by the simple color menu concept. Additionally, the map is constructed by the adjacent in the directed graph theory. This concept will demonstrate with the interactive system of the small pagoda in Thailand. The administrator management is easy and simple to manage this system, and the user or visitor feels like stay in the physical place and attracts to go to that place with more satisfied contentment.
2013 International Conference on Machine Learning and Cybernetics, 2013
Recently, several binarisation techniques have been proposed to process different kinds of ancien... more Recently, several binarisation techniques have been proposed to process different kinds of ancient document images. While many well-known binarisation techniques are particularly suitable for certain types of document images, there is no specific guidelines on the determination of the appropriate type of image degradation, or characteristics of the image. In this paper, a novel method has been proposed to generate the optimal binary image from different binarised outputs from a document image. This approach is based on weight majority vote, and uncertain pixels are then determined based on local areas of the binarised images, by applying iteration of weight majority vote. Experiment over benchmark data set of the Document Image Binarization Contest (DIBCO) 2011 shows that the proposed method provided better performance than most well-known techniques. The proposed method has also been applied to ancient manuscripts on palm leaves from Thailand and this approach provided better results than binarised outputs from original binarisation techniques.
The 2013 10th International Joint Conference on Computer Science and Software Engineering (JCSSE), 2013
ABSTRACT
Rainfall prediction is a challenging task especially in a nwdern world facing the major environme... more Rainfall prediction is a challenging task especially in a nwdern world facing the major environmental problem of global warming. The proposed method uses an Adaptive Radial Basis Function neural network mode with a specially designed gerietic algoruhm (CA) to obtain the optimal model parameters. A significant feature of the Adaptive Radinl Basis Function network is that it is able creak new hidden units and solve the spread factor problem using a genetic algorithm. I t is shown that the evolved parameter values improved performance.
Advanced Robotics, 2011
This paper discusses a method for controlling a hyper-redundant arm to manipulate an object on a ... more This paper discusses a method for controlling a hyper-redundant arm to manipulate an object on a plane. The hyper-redundant arm can perform simple whole-arm manipulation by coiling or wrapping around the object and then pulling the object toward the goal position. The process of object manipulation can be separated into two steps: encircling the object and transporting the object. In the process of encircling the object, the arm is controlled by a set of virtual constraints that guide the arm to reach around the object and encircle it, keeping the arm within a specified bound to ensure the circular shape around the object. In the process of transporting the object, a simplified desired shape is generated from a Bézier curve according to a given goal position and the arm geometry. Then, the gradient descent method is used to update the joint angles of the arm at each step to move the arm toward the desired shape until the object reaches its target position. The proposed method has been tested in both simulation and real experiments.
Uploads
Papers by Chatklaw Jareanpon