Machine learning (ML) provides novel and powerful ways of accurately and efficiently recognizing complex patterns, emulating nonlinear dynamics, and predicting the spatio-temporal evolution of weather and climate processes. Off-the-shelf ML models, however, do not necessarily obey the fundamental governing laws of physical systems, nor do they generalize well to scenarios on which they have not been trained. We survey systematic approaches to incorporating physics and domain knowledge into ML models and distill these approaches into broad categories. Through 10 case studies, we show how these approaches have been used successfully for emulating, downscaling, and forecasting weather and climate processes. The accomplishments of these studies include greater physical consistency, reduced training time, improved data efficiency, and better generalization. Finally, we synthesize the lessons learned and identify scientific, diagnostic, computational, and resource challenges for developing truly robust and reliable physics-informed ML models for weather and climate processes. This article is part of the theme issue ‘Machine learning for weather and climate modelling’.

Optical Character recognition (OCR) has enabled many applications as it has attained high accuracy for all printing documents and also for handwriting of many languages. However, the state-of-the-art accuracy of Arabic handwritten word recognition is far behind. Arabic script is cursive (both printed and handwritten). Therefore, traditionally Arabic recognition systems segment a word to characters first before recognizing its characters. Arabic word segmentation is very difficult because Arabic letters contain many dots. Moreover, Arabic letters are context sensitive and some letters overlapped vertically. A holistic recognizer that recognizes common words directly (without segmentation) seems the plausible model for recognizing Arabic common words. This paper presents the result of training a Conventional Neural Network (CNN), holistically, to recognize Arabic names. Experiments result shows that the proposed CNN is distinct and significantly superior to other recognizers that were used with the same dataset.

In this paper, we presented a Convolutional Neural Network (CNN) model for off-line Arabic handwritten character recognition. The proposed CNN model used the dataset which prepared by Sudan University of Science and Technology-Arabic Language Technology group. The dataset is pre-processed before feeding it to the CNN model. In the pre-processing, all the characters images are size normalized to fit in a 20 by 20 pixel and then centred in a scaled images of size 28×28 pixel using the centre of mass then all the images are converted to be having a black background and white foreground colours. The pre-processed images are fed to the CNN model, which is constructed using the sequential model of the Keras library under tensorflow environment. The accuracy obtained varied from 93.5% as test accuracy to 97.5% as training accuracy showing better results than other methods that used the same dataset.

This paper presents a novel system approach for online Arabic handwriting recognition. The approach segments the word using new character boundaries detection based algorithm. Moreover it employs HMM-based classification method for the recognition. A dataset: Sudan University of Science and Technology Online Arabic Handwriting (SUSTOLAH) is used in testing the proposed approach. Promising experimental results of testing the approach with the dataset are provided

No abstract