Automated Ham-Spam Lexicon Generation Based on Semantic Relations Extraction

Abstract

One of the current essential methods for communication is electronic email (e-mail). It is currently considered the official method for different business activities such as conducting agreements, the setup of official meetings, and team collaboration. This continuous interest in e-mails as a communication channel has drawn the attention to the need for eliminating spam which have a vital effect on both network resources and business activities. This research focuses on generating a ham-spam lexicon based on text analysis which is aimed to be one of the main resources for detecting personal spam e-mails. The lexicon generation is a key step to efficiently and economically successful spam elimination. The proposed framework has proven its applicability on a dataset of six groups and the classification algorithms have been examined to prove the efficient classification. The research is a step in a wider view for general intelligent business communication and collaboration framework.

A Collaborative Mining-Based Decision Support Model for Granting Personal Loans in the Banking Sector

“…Another research by (Baixauli, Alvarez, & Módica, 2012) highlighted that each model could be superior to the others with respect to the data source and the problem nature. Data mining also contributed to social networks as different researches targeted manipulating the data in social networks such as in (Khedr, Idrees, & Shabaan, 2020). The research aimed at detecting the fraudlent emails.…”

The Impact of Applying ISO Standards Systems on Improving the Quality of the Performance in Higher Educational Institutions in Egypt

“…Predicting a qualified generation that can interact with the requirements and changes of the times could be achieved. Solving problems and keeping pace with advanced countries in the fields of medicine, industry, engineering, and innovation-based science is also one of the main objectives [8]. The successive events of this century have brought about many changes in economic, social, and cultural systems.…”

Facebook Profile Credibility Detection using Machine and Deep Learning Techniques based on User’s Sentiment Response on Status Message

“…After that, we analyzed users' sentiment analysis and performed LDA on the status corpus (Topic Modeling algorithm, Latent Dirichlet Allocation) to cluster topicfeatures distribution and Principal Component Analysis (PCA) method to visualize and classify each status topic distribution to compute a credibility score. Machine learning techniques contribute efficiently to detect semantic relations [3] in general and frauds [4] in specific. According to the revolution in Artificial Intelligence (AI),…”