During COVID-19, universities started to use mobile learning applications as one of the solutions to support distance learning. The readiness of universities to apply new systems, such as mobile learning applications, is considered one of the critical issues to ensure the system’s success. Determining the most important aspects of readiness to use mobile learning is a key step to adopt mobile learning in an effective way. To address this issue, this research aims to determine the most important determinants influencing mobile learning readiness by employing the Technology Acceptance Model (TAM). The Structural Equation Modelling (SEM) method was used to test the hypotheses in the proposed model. The results showed that the relationship between mobile learning readiness and awareness, IT infrastructure and top management support was positively significant. In conclusion, the findings will be of value to decision makers and mobile learning developers in universities to enhance the development of mobile learning applications. In addition, it may help facilitate and promote the usage of mobile learning applications among users.

With the advent of online data, sentiment analysis has received growing attention in recent years. Sentiment analysis aims to determine the overall sentiment orientation of a speaker or writer towards a specific entity or towards a specific feature of a specific entity. A fundamental task of sentiment analysis is sentiment classification, which aims to automatically classify opinionated text as being positive, negative, or neutral. Although the literature on sentiment classification is quite extensive, only a few endeavors to classify opinionated text written in the Arabic language can be found. This paper provides a comprehensive survey of existing lexicon, machine learning, and hybrid sentiment classification techniques for Arabic language.

Activation functions are essential components in any neural network model; they play a crucial role in determining the network’s expressive power through their introduced non-linearity. Rectified Linear Unit (ReLU) has been the famous and default choice for most deep neural network models because of its simplicity and ability to tackle the vanishing gradient problem that faces backpropagation optimization. However, ReLU introduces other challenges that hinder its performance; bias shift and dying neurons in the negative region. To address these problems, this paper introduces a novel composite monotonic, zero-centered, semi-saturated activation function called Hyperbolic cosine Linearized SquasHing function (HcLSH) with partial gradient-based sparsity HcLSH owns many desirable properties, such as considering the contribution of the negative values of neurons while having a smooth output landscape to enhance the gradient flow during training. Furthermore, the regularization effect resulting from the self-gating property of the positive region of HcLSH reduces the risk of model overfitting and ensures learning more robust expressive representations. An extensive set of experiments and comparisons is conducted that includes four popular image classification datasets, seven deep network architectures, and ten state-of-the-art activation functions. HcLSH exhibited the Top-1 and Top-3 testing accuracy results in 20 and 25 out of 28 conducted experiments, respectively, suppressing the widely used ReLU that achieved 2 and 5, and the reputable Mish that achieved 0 and 5 Top-1 and Top-3 testing accuracy results, respectively. HcLSH attained improvements over ReLU, ranging from 0.2% to 96.4% in different models and datasets. Statistical results demonstrate the significance of the enhanced performance achieved by our proposed HcLSH activation function compared to the competitive activation functions in various datasets and models regarding the testing loss Furthermore, the ablation study further verifies the proposed activation function’s robustness, stability, and adaptability for the different model parameter.