Sentiment analysis (SA) and opinions mining (OM) are used to evaluate users' feedbacks and comments on issues related to news, products, services, etc. This topic has received increasing interests over the last decade due to the spread and expansion of social networks. SA for online reviews poses challenges to researchers and decision makers because such comments are written in unstructured formats with usually informal languages, expressions and possibly mixed languages. For Arabic, further challenges exist due to the language complexity and the limited number of research publications and datasets collected and analysed for such purpose. In SA, two approaches are generally used to determine the polarity of reviews: supervised (corpus-based) and unsupervised (lexicon-based). In this work, we follow the second approach and build a very large sentiment lexicon and a lexicon-based SA tool. The results show that the proposed tool performs very well.

Information users depend heavily on emails’ system as one of the major sources of communication. Its importance and usage are continuously growing despite the evolution of mobile applications, social networks, etc. Emails are used on both the personal and professional levels. They can be considered as official documents in communication among users. Emails’ data mining and analysis can be conducted for several purposes such as: Spam detection and classification, subject classification, etc. In this paper, a large set of personal emails is used for the purpose of folder and subject classifications. Algorithms are developed to perform clustering and classification for this large text collection. Classification based on NGram is shown to be the best for such large text collection especially as text is Bi-language (i.e. with English and Arabic content)

The continuous inventions and evolutions in all information technology fields open new channels and opportunities to enhance teaching and educational methods. On one side, these may improve the abilities of educators to present information in interactive and media-enhanced formats relative to traditional methods. This may help students or learners through offering them the information in channels and methods that can be easier to understand, deal with and retrieve. On the other hand, offering those alternative methods of teaching can be helpful particularly for children, people with special needs, or students in rural areas where they can have virtual or remote instructors, especially for majors who have shortages. The purpose of this study is to investigate the impact of utilising multimedia technologies on enhancing, or not, the effectiveness of teaching students at early stages in Jordanian primary schools. To achieve this objective, a program has been developed to test the students' ability to understand basic mathematical knowledge and skills. Two groups were selected from a local school based on their own class distribution, where one group was taught the subject in basic math using a program developed for this purpose and the second class was taught the same subject using traditional methods of teaching (i.e. direct student-to-child instruction, board, etc.). Results showed that in such math skills at this age, using programs or multimedia-enhanced methods of teaching can be effective in getting students' attention, especially when cartoon characters are used. Results also showed that there is no significant difference in learning and knowledge skills and information absorption based on gender distribution, as a comparison of the results between little boys and girls showed no significant difference in their learning skills.

The vast majority of digital currency transactions rely on a blockchain framework to ensure quick and accurate execution. As such, understanding how a blockchain works is vital to understanding the dynamics of cryptocurrency operations. One of the key benefits of this type of system is the exhaustive records captured in a given marketplace. The interwoven movement between agents can effectively be expressed as a graph via the extraction of historical data from the blockchain. By looking at a specific blockchain as an interaction of its agents, network representation learning can be leveraged to examine these relationships. Furthermore, the analysis of a graph structure can be enhanced through the application of modern and sophisticated machine learning techniques. Leveraging the automated nature of these methods can create meaningful observations of the input network. In this paper, we utilize several machine learning models to detect anomalous transactions in various digital currency markets. We find that supervised learning techniques yield encouraging results, whereas unsupervised learning techniques struggle more with the classification.