In recent disease advancement and with increasing rate of cases of Pneumonia for the past decades, this has led to the overwhelming challenges within several health systems worldwide. This problem has necessitated the need for efficient and effective measures to mediate and alleviate these challenging issues. Pneumonia is largely caused by inflammation of the lungs because of cold temperature or associated with cold weather. Any delay in treating this disease could be dangerous and could develop to lung cancer. Medical diagnoses are carried out mostly at the beginning with a chest X-ray image to ascertain the presence of pneumonia within a patient. However, this process can be laborious and time sensitive, which might produce inaccurate outcomes or results. In this study, we proposed a ResNet50 architecture to extract the characteristic features from the chest X-ray images. This architecture helps to classify the features to allow us to predict or detect the presence of pneumonia in a patient X-ray image or not. In this study, we utilized dataset extracted from Kaggle which was made up of two distinctive files (1) for training (2) for testing. The dataset was preprocessed to balance the data, and the training data was divided into two sections or parts, one for the training set and the other for the validation set. To enhance our model performance, we performed data augmentation to increase the image variations and data generator was created to reduce the memory usage during the training phase of the model. The ResNet50 architecture was customized for the task with several hyperparameter tuning and the validation data was used to experiment on the model performance validation set. The model was evaluated and measured using classification reports and confusion matrix. We performed the experiment with the following optimization algorithms, SGD, Adam, Adamax and Adagrad, each achieving the following accuracy 93%, 90%, 92% and 89% respectively. We implement the following activation functions , ReLU and ULU within the research. Different validation split ratios were implemented to ascertain the best result. The best performance was observed when 90% or 80% of the training set data was used and 10% or 20% of the validation set data was used within the training set image files.

Speech Emotion Recognition (SER) is a fascinating area of research in machine learning. Researchers have been exploring different techniques to improve this field including using deep learning models, feature extraction methods and transfer strategies to improve the accuracy and robustness of SER models. The advancements in SER not only contribute to the field of artificial intelligence but also have the potential to enhance our understanding of human emotions and improve communication between humans and machines.

With the advent and popularity of big data mining and huge text analysis in modern times, automated text summarization became prominent for extracting and retrieving important information from documents. This research investigates aspects of automatic text summarization from the perspectives of single and multiple documents. Summarization is a task of condensing huge text articles into short, summarized versions. The text is reduced in size for summarization purpose but preserving key vital information and retaining the meaning of the original document. This study presents the Latent Dirichlet Allocation (LDA) approach used to perform topic modelling from summarised medical science journal articles with topics related to genes and diseases. In this study, PyLDAvis web-based interactive visualization tool was used to visualise the selected topics. The visualisation provides an overarching view of the main topics while allowing and attributing deep meaning to the prevalence individual topic. This study presents a novel approach to summarization of single and multiple documents. The results suggest the terms ranked purely by considering their probability of the topic prevalence within the processed document using extractive summarization technique. PyLDAvis visualization describes the flexibility of exploring the terms of the topics’ association to the fitted LDA model. The topic modelling result shows prevalence within topics 1 and 2. This association reveals that there is similarity between the terms in topic 1 and 2 in this study. The efficacy of the LDA and the extractive summarization methods were measured using Latent Semantic Analysis (LSA) and Recall-Oriented Understudy for Gisting Evaluation (ROUGE) metrics to evaluate the reliability and validity of the model.