Jarunee Saelee

Tourists frequently use online search engines for travel planning, making search data a valuable predictor of future tourism volume. This study employs machine learning to analyse the predictive power of keyword search data for forecasting tourist arrivals, incorporating a lag time between searches and arrivals. The dataset is collected and prepared from two sources: a search engine and government agencies, covering the years 2014-2019, to be analysed by machine learning. The SARIMA model effectively forecasts trends in keyword searches and tourist numbers, while SVM (Support Vector Machine) and Random Forest outperform other methods in predicting arrivals. This research supports tourism operators and stakeholders in planning for future tourists, utilising the obtained keywords to enhance visibility in tourist searches through SEO.

Stillbirths continue to be a significant global issue, reflecting the quality of obstetric care. In Thailand, stillbirth remains a concern, especially in southern provinces like Pattani. Traditional risk identification focuses on historical obstetric complications, but there is a need for advanced technology to improve early detection and prevention. This study investigates the factors influencing stillbirth risk by analyzing perinatal data from 2018 to 2021 in Pattani Province. A machine learning-based predictive model was developed using various algorithms, including Naïve Bayes, logistic regression, deep learning, decision tree, random forest, and gradient-boosted trees, to handle imbalanced datasets. Performance comparisons of these models were conducted, and the best-performing model was implemented into a web application to provide personalized recommendations to pregnant women. The study highlights the importance of advanced data analytics in stillbirth prevention and offers an innovative, accessible solution for pregnant women to manage and monitor their pregnancies. The system not only improves the prediction of stillbirth risks but also provides an interactive, low-cost platform for delivering real-time suggestions, aiming to reduce stillbirth occurrences and improve maternal health outcomes.

The objectives of feature selection include simplifying modeling and making the results more understandable, improving data mining efficiency, and providing clean and understandable data preparation. With big data, it also allows us to reduce computational time, improve prediction performance, and better understand the data in machine learning or pattern recognition applications. In this study, we present a new feature selection approach based on hierarchical concept models using formal concept analysis (FCA) and a decision tree (DT) for selecting a subset of attributes. The presented methods are evaluated based on all learned attributes with 10 datasets from the UCI Machine Learning Repository by using three classification algorithms, namely decision trees, support vector machines (SVM), and artificial neural networks (ANN). The hierarchical concept model is built from a dataset, and it is selected by top-down considering features (attributes) node for each level of structure. Moreover, this study is considered to provide a mathematical feature selection approach with optimization based on a paired-samples t-test. To compare the identified models in order to evaluate feature selection effects, the indicators used were information gain (IG) and chi-squared (CS), while both forward selection (FS) and backward elimination (BS) were tested with the datasets to assess whether the presented model was effective in reducing the number of features used. The results show clearly that the proposed models when using DT or using FCA, needed fewer features than the other methods for similar classification performance.

The objectives of feature selection include simplifying modeling and making the results more understandable, improving data mining efficiency, and providing clean and understandable data preparation. With big data it also allows us to reduce computational time, improve prediction performance, and better understand the data in machine learning or pattern recognition applications. In this study, we present a new feature selection approach based on hierarchical concept models using formal concept analysis (FCA) and decision tree (DT) for selecting a subset of attributes. The presented methods are evaluated based on all learned attributes with 10 datasets from the UCI Machine Learning Repository by using three classification algorithms, namely decision trees, support vector machines (SVM), and artificial neural networks (ANN). The hierarchical concept model is built from a dataset, and it is selected by top-down considering features (attributes) node for each level of structure. Moreover, this study is considered to provide a mathematical feature selection approach with optimization based on paired-samples t-test. To compare the identified models in order to evaluate feature selection effects, the indicators used were information gain (IG) and chi-squared (CS), while both forward selection (FS) and backward elimination (BS) were tested with the datasets to assess whether the presented model was effective in reducing the number of features used. The results show clearly that the proposed models when using DT or using FCA, needed fewer features than the other methods for similar classification performance.