A Deep Learning Approach for Dengue Fever Prediction in Malaysia Using LSTM with Spatial Attention.

This research aims to predict dengue fever cases in Malaysia using machine learning techniques. A dataset consisting of weekly dengue cases at the state level in Malaysia from 2010 to 2016 was obtained from the Malaysia Open Data website and includes variables such as climate, geography, and demographics. Six different long short-term memory (LSTM) models were developed and compared for dengue prediction in Malaysia: LSTM, stacked LSTM (S-LSTM), LSTM with temporal attention (TA-LSTM), S-LSTM with temporal attention (STA-LSTM), LSTM with spatial attention (SA-LSTM), and S-LSTM with spatial attention (SSA-LSTM). The models were trained and evaluated on a dataset of monthly dengue cases in Malaysia from 2010 to 2016, with the task of predicting the number of dengue cases based on various climate, topographic, demographic, and land-use variables. The SSA-LSTM model, which used both stacked LSTM layers and spatial attention, performed the best, with an average root mean squared error (RMSE) of 3.17 across all lookback periods. When compared to three benchmark models (SVM, DT, ANN), the SSA-LSTM model had a significantly lower average RMSE. The SSA-LSTM model also performed well in different states in Malaysia, with RMSE values ranging from 2.91 to 4.55. When comparing temporal and spatial attention models, the spatial models generally performed better at predicting dengue cases. The SSA-LSTM model was also found to perform well at different prediction horizons, with the lowest RMSE at 4- and 5-month lookback periods. Overall, the results suggest that the SSA-LSTM model is effective at predicting dengue cases in Malaysia.

Desertification prediction with an integrated 3D convolutional neural network and cellular automata in Al-Muthanna, Iraq.

Desertification is a major environmental issue all over the world, and Al-Khidhir district, Al-Muthanna, in the south of Iraq is no exception. In mapping, assessing, and predicting desertification, remote sensing and geospatial solutions (spatial analysis, machine learning) are crucial. During 1998-2018, this study employed satellite images from Landsat TM, ETM + , and OLI to map and predict desertification in the Al-Khidhir district. The year 2028 was chosen as the target date. Prediction models were constructed using a 3D convolutional neural network (3D CNN) and cellular automata (CA) techniques. In addition to the historical land cover maps, the model incorporated desertification indicators identified as important in the study, including geology, soil type, distance from waterways, elevation, population density, and Normalized Difference Vegetation Index (NDVI). Several accuracy metrics were used to evaluate the models, including overall accuracy (OA), average accuracy (AA), and the Kappa index (K). The simulated and actual land cover maps from 1998 and 2008 were used to evaluate the desertification prediction models. The 3D CNN model outperforms the typical 2D CNN for both the 2008 and 2018 images, according to the results. For the 2008 image, the 3D CNN model achieved 89.675 OA, 69.946 AA, and 0.781 K, while the 2018 image achieved 91.494 OA, 75.138 AA, and 0.770 K. The 2D CNN model performed a little worse than the 3D CNN model. The results of the change assessment showed that between 1998 and 2008, agricultural land was the dominant class (39%, 47.4%, respectively). The bare land, however, was the most dominant class in 2018, accounting for 46.6% of the total, compared to 26.2% for agricultural land. The spatial distribution characteristics of desertification in the Al-Khidhir, in the year 1998, were prevalent in the area's south (25.9%). In the following 10 years, desertification has spread to the surrounding territories. In the year 2008, desertification increased in the north of the study area (50.8%). Unless the local administration of Al-Khidhir district establishes desertification control strategies, this study suggests that the extent of bare land could expand in 2028 (54.1%).

Land use/cover change detection and urban sprawl analysis in Bandar Abbas city, Iran.

The process of land use change and urban sprawl has been considered as a prominent characteristic of urban development. This study aims to investigate urban growth process in Bandar Abbas city, Iran, focusing on urban sprawl and land use change during 1956-2012. To calculate urban sprawl and land use changes, aerial photos and satellite images are utilized in different time spans. The results demonstrate that urban region area has changed from 403.77 to 4959.59 hectares between 1956 and 2012. Moreover, the population has increased more than 30 times in last six decades. The major part of population growth is related to migration from other parts the country to Bandar Abbas city. Considering the speed of urban sprawl growth rate, the scale and the role of the city have changed from medium and regional to large scale and transregional. Due to natural and structural limitations, more than 80% of barren lands, stone cliffs, beach zone, and agricultural lands are occupied by built-up areas. Our results revealed that the irregular expansion of Bandar Abbas city must be controlled so that sustainable development could be achieved.

Capability of integrated MODIS imagery and ALOS for oil palm, rubber and forest areas mapping in tropical forest regions.

Various classification methods have been applied for low resolution of the entire Earth's surface from recorded satellite images, but insufficient study has determined which method, for which satellite data, is economically viable for tropical forest land use mapping. This study employed Iterative Self Organizing Data Analysis Techniques (ISODATA) and K-Means classification techniques to classified Moderate Resolution Imaging Spectroradiometer (MODIS) Surface Reflectance satellite image into forests, oil palm groves, rubber plantations, mixed horticulture, mixed oil palm and rubber and mixed forest and rubber. Even though frequent cloud cover has been a challenge for mapping tropical forests, our MODIS land use classification map found that 2008 ISODATA-1 performed well with overall accuracy of 94%, with the highest Producer's Accuracy of Forest with 86%, and were consistent with MODIS Land Cover 2008 (MOD12Q1), respectively. The MODIS land use classification was able to distinguish young oil palm groves from open areas, rubber and mature oil palm plantations, on the Advanced Land Observing Satellite (ALOS) map, whereas rubber was more easily distinguished from an open area than from mixed rubber and forest. This study provides insight on the potential for integrating regional databases and temporal MODIS data, in order to map land use in tropical forest regions.