JOURNAL OF INFORMATION SCIENCE AND TECHNOLOGY

Dengue Fever Risk Prediction System Using Data Mining Techniques

Sopee Kaewchada, Sunisa Kidjaideaw, Wichit Sungton, Chaimongkon Chuaynukoon — Sun, 30 Jun 2024 00:00:00 +0700

This research aims to 1) study and measure the effectiveness of a classification model using decision tree-based methods, 2) to develop a dengue fever risk prediction system, and 3) To study the effectiveness of the dengue fever risk prediction system using a sample group created from patient data with dengue fever cases in Nakhon Si Thammarat province over a period of 5 years (B.E. 2558 - B.E. 2563), utilizing data mining techniques for classification using decision trees. The results of the research showed that 1) The model used for predicting dengue fever risk, based on the decision tree classification technique, achieved an accuracy of 83.5%, 2) The dengue fever risk prediction system consists of sub-systems for user authentication, dengue fever incidence data management, dengue fever reporting, management of dengue fever datasets after data mining, and dengue fever risk analysis, and 3) The satisfaction level with the dengue fever risk prediction system was found to be high. ( = 4.06).

Efficiency of plant diseases classification by convolutional neural network with optimization algorithm and activation function

A-nanya Promkot — Sun, 30 Jun 2024 00:00:00 +0700

Plant diseases are a problem that has a huge impact on farmers. Detection of plant diseases at an early stage to effectively control the spread of germs Therefore, it plays an important role in the agricultural industry. However, traditional approach requires extensive knowledge of the expert. It is expensive and requires a lot of labor. Nowadays, with the advancement of information technology, machine learning and deep learning has been applied to automatic identification of plant disease. Currently, convolutional neural network methods is a method that has been recognized for its efficiency in image classification. The objective of this research is to find appropriate values for the ResNet50 method with optimization algorithms, including AdaDelta, AdaGrad, Adam, RMSprop, and SGD, and activation functions including ReLU, Sigmoid, and Tanh ,for plant disease classification by using plant leave image. Evaluated performance of plant disease classification by using the PlantVillage dataset. The results showed that the ResNet50 method with RMSprop optimizer and Sigmoid activation function gave the highest Accuracy value of 0.94, Precision value of 0.94, Recall 0.93, and F -measure is equal to 0.93. Therefore, it can be concluded that in selecting a model for learning to achieve the most efficient results should consider additional factors including algorithms to increase the efficiency of the model and stimulation functions

Thai Text Compression Algorithm Employing Word-Formation Creation

Prayat Le-wan, Chouvalit Khancome — Sun, 30 Jun 2024 00:00:00 +0700

The compression of text without data loss is a fundamental aspect of computer science, crucial for minimizing the storage space required for large datasets. This principle has been continuously developed and has consistently attracted the interest of researchers. This research article presents a highly efficient design for a new text compression method specifically tailored for compressing Thai language text. The procedural mechanism involves the creation of a new dictionary-like structure termed the "Pre-Processing Section" based on the patterns of word formation in the Thai language. This structure is utilized for referencing terms during compression and decompression processes. The data compression is executed by storing information in a binary file using the newly developed Word-Formation Thai Text Compression Algorithm (WFTTCA). The compression process following this newly developed method can achieve compression rates in theoretical terms, represented by ASCII- TIS620 encoding, ranging from 37.50% to 79.17%, with a maximum average of 63.75%. For Unicode encoding, compression rates range from 68.75% to 89.58%, with a maximum average of 81.88%. In the case of UTF-8 encoding, compression rates range from 79.17% to 93.06%, with a maximum average of 87.92%. These compression rates correspond to a range of 3.51 to 10.50 times the original data size. The experimental results from the development of the program based on the new method, using actual Thai language data randomly sampled from 1Kb-100Kb and imported from news websites, reveal that the program is capable of compressing data encoded with ASCII-TIS620 by percentages ranging from 78.09% to 84.55%. For Unicode encoding, the compression rates range from 81.50% to 86.62%. Similarly, for UTF-8 encoding, the compression rates range from 88.09% to 91.11%. When comparing the compression efficiency achieved with popular current compression software, it is found that the program developed from the new method can achieve significantly higher compression rates, both in terms of percentage compression and compression ratios.

The Development of Calories Tracker Application on Mobile Device

Weerawut Chaiyasomboon , Waraporn Jirapanthong — Sun, 30 Jun 2024 00:00:00 +0700

According to the study on health of Thai people [1], it is found
that one third of Thai people are overweight and one tenth of Thais are considered obese. Moreover, the research [1] claimed that during 1991-2009, the rate of Thais aged 15 years and over was overweight double (from 17.2 percent to 34.7 percent of all Thais) and considered obese increased almost three times (from 3.2 percent to 1 percent). The problem of overnutrition, especially in early childhood and school-aged children, is an important health issue that all parties must be aware of and cooperate to solve. This research aims to initiate and develop a tool that can support Thais to take care of their health sustainably and effectively. In this research, the researchers presented the development of the application named “HealthMe” The objectives are to monitor the nutrition and calories which a user consumes at each meal and assist the user to control the amount of consuming calories each day. Consumption data is recorded and analyzed along with other health data. The data can be then analyzed along with exercise data to analyze in-depth and broad health information. In this article, the design and development of the application are presented. It can also provide recommendation about healthy food and exercise methods that are appropriate to the user's health information. There is also a community of users using this application to take care of their health and food consumption. They can communicate and share information about nutrients intake to guide one's own diet. The application is developed using the SwiftUI framework connected to the Firebase database (using Authentication, Cloud Firestore, and Cloud Storage), which is only supported on iPhones running on iOS 16.4 or later. The authors propose the development of App on iOS platform. and Firebase. Additionally, we collected and analyzed functional and quality requirements from 78 Thai people aged 18-65 years. Our work include the design of user experience and user interface (UX/UI). The experimental results are claimed that the accuracy in term of functions was 100 percent and the user satisfaction was 93 percent.

Development of Web Application to Support Special Investigations

Puwanai Sangphon , Waraporn Jirapanthong — Sun, 30 Jun 2024 00:00:00 +0700

This article introduces the development of a web application designed and developed to support the investigation of special cases, particularly those with a high number of victims. The users are categorized into three groups: the general public, who are victims; special case investigators; and administrators. The system is equipped with the capability to manage case data and investigators, as well as to register victims through online forms. Additionally, it can record testimonies, generate reports, and compile relevant documents efficiently. The system also provides efficient statistical analysis and various reports.