In this paper, we introduce an algorithm for grouping Arabic
documents for building an ontology and its words. We execute
the algorithm on five ontologies using Java. We manage the
documents by getting 338667 words with its weights
corresponding
to each ontology. The algorithm had proved its
efficiency in optimizing classifiers (SVM, NB) performance, which
we tested in this study, comparing with former classifiers results
for Arabic language.
With the increase in social networks, people have started to share information via different types of social media. Among themwere sites for exchanging people's opinions and others to exchange stories about real life and stories for children. In this
work we made use of children's stories and employed them to teach children with Down syndrome the correct feelings by reading a story for them, converting it into text, processing the text using natural languages and extracting feelings automatically from This story, and to achieve this, we used several techniques, combined them, and compared their results on a number of short stories dedicated to children, where each of the different techniques that were unsupervised, such as Dictionary Based or supervised, such as data-dependent neural networks, were used to analyze feelings, where we used multiple classifiers. They are Support Vector Machine, Stochastic Gradient Descent, Decision Tree, Random Forest, Naïve Bayes, K-Nearest Neighbor, and Nearest Centroid We also used deep neural networks as the example of RNN. Finally, the correct sentiment for the story was reached through Dictionary Based which gave the best accuracy and then showed a photo that shows the child the expression they want to start with The events of this story to interact with him and learn the correct expression
