There has been an unsettling rise in the intensity and frequency of natural disasters due to climate change and anthropogenic activities. Artificial intelligence (AI) models have shown remarkable success and superiority to handle huge and nonlinear data owing to their higher accuracy and efficiency, making them perfect tools for disaster monitoring and management. Accordingly, natural disaster management (NDM) with the usage of AI models has received increasing attention in recent years, but there has been no systematic review so far. This paper presents a systematic review on how AI models are applied in different NDM stages based on 278 studies retrieved from Elsevier Science, Springer LINK and Web of Science. The review: (1) enables increased visibility into various disaster types in different NDM stages from the methodological and content perspective, (2) obtains many general results including the practicality and gaps of extant studies and (3) provides several recommendations to develop innovative AI models and improve the quality of modeling. Overall, a comprehensive assessment and evaluation for the reviewed studies are performed, which tracked all stages of NDM research with the applications of AI models.

No abstract

Aim/Purpose: The objective of this research is to investigate the effectiveness of educational games on learning computer programming. In particular, we are examining whether allowing students to manipulate the underlying code of the educational games will increase their intrinsic motivation. Background: Young students are fond of playing digital games. Moreover, they are also interested in creating game applications. We try to make use of both of these facts. Methodology: A prototype was created to teach the fundamentals of conditional structures. A number of errors were intentionally included in the game at different stages. Whenever an error is encountered, students have to stop the game and fix the bug before proceeding. A pilot study was conducted to evaluate this approach. Contribution: This research investigates a novel approach to teach programming using educational games. This study is at the initial stage. Findings: Allowing the programming students to manipulate the underlying code of the educational game they play will increase their intrinsic motivation. Recommendations for Practitioners: Creating educational games to teach programming, and systematically allowing the players to manipulate the gaming logic, will be beneficial to the students. Recommendation for Researchers: This research can be extended to investigate how various artificial intelligence techniques can be used to model the gamers, for example, skill level. Impact on Society: The future generations of students should be able to use digital technologies proficiently. In addition, they should also be able to understand and modify the underlying code in the digital things (like Internet of Things).This research attempts to alleviate the disenchantment associated with learning coding. Future Research: A full scale evaluation – including objective evaluation using game scores – will be conducted. One-way MANOVA will be used to analyze the efficacy of the proposed intervention on the students’ performance, and their intrinsic motivation and flow experience.

From a macro-perspective, based on machine learning and data-driven approach, this paper utilizes multi-featured data from 31 provinces and regions in China to build a Bayesian network (BN) analysis model for predicting air quality index and warning the air pollution risk at the city level. Further, a two-layer BN for analyzing influencing factors of various air pollutants is developed. Subsequently, the model is applied to forecast the trends of temporal and spatial changes in the form of probabilistic inference and to investigate the degree of impact incurred from individual influencing factors. From the comparisons with the results obtained from other machine learning approaches and algorithms such as neural networks, it is concluded that by comprehensively using the established BN, one can not only reach a monitoring and early warning accuracy rate of 90% but also scrutinize and diagnose the main cause of air pollution risk changes from the perspective of probability.

Aim/Purpose: The key objective of this research is to examine whether fix-and-play educational games improve students' performance in learning programming languages. We also quantified the flow experiences of the students and analyzed how the flow contributes to their academic performances. Background: Traditionally, learning the first computer programming language is considered challenging, In this study, we propose the fix-and-play gaming approach that utilizes the following three facts to alleviate certain difficulties associated with learning programming: 1. digital games are computer programs, 2. young students are fond of playing digital games, and 3. students are interested in creating their own games. Methodology: A simple casual game Shoot2Learn was created for learning the fundamentals of branching. A number of errors were intentionally implanted in the game at different levels, and the students were challenged to fix the bugs before continuing the game. During the play, the program keeps records of the student’s academic progress and the time logs at different stages to measure the flow experience of the students. The proposed approach was systematically evaluated using a quasi-experimental design in real classroom settings in two countries, Sri Lanka, and USA. Contribution: The results derived from this research provide empirical evidence that the fix-and-play educational games ease some challenges in learning programming and motivate the students to play and learn. Findings: The results show that the first-year programming students who play the fix-and-play game gain statistically significant improvement in their academic performance. However, the result fails to suggest a significant positive correlation between the flow experience and academic performance. Recommendations for Practitioners: Empowering the students to fix the bugs in the educational games they play will motivate them to stay in the game and learn continuously. However, we have to make sure that the types and timing of bugs do not hinder the flow experience of the players, Recommendation for Researchers: Students normally play industry-level high-quality games. Experience and interest in game-playing differ significantly between students. Gender difference also plays an important role in selecting game genres. We need to identify how to address these issues when resources are not sufficient to provide an individualized gaming experience. Impact on Society: Programming is an essential skill for computer science students. The outcome of this research shows that the proposed approach helps to reduce the disenchantment associated with learning the first programming language. Future Research: Further investigation is necessary to verify whether the AI techniques such as user modeling can be used in educational games to reduce the effects of uncertainty associated with the variations in students' gaming skills and other factors.