الفهرس يوجد فقط 14 صفحة متاحة للعرض العام
 |  | from | 10 |  |  |
| | | from | 10 | | |
المستخلص
Human behavior is an important topic in computational research {uFB01}elds. Nowadays, human beha-vior analysis based on modern communication technologies is enabled social scientists to invest- igate human behavior beyond traditional survey methodologies. Reality mining is a science that studies objects behaviors using modern communication devices. Reality mining started mainly with studying and analyzing human behavior. In this thesis, reality mining model is proposed for human behavior modeling. The contributions of this thesis are enhancing previous models, introducing new ideas, and new methodologies in human behavior modeling. Human behavior has several branches such as personality, and relationships. We focused on speci{uFB01}c branches that have the most studies and important for physiology and economic studies. Personality, social relationships, interpersonal relationships, and human activities are the core of this thesis. Personality research has been raised as an emerging topic in computational research {uFB01}elds. The main purpose of studying personality is to identify big {uFB01}ve personality traits from smart- phone data sources. In this thesis, two di{uFB00}erent approaches are proposed in personality. The {uFB01}rst one is a system to de{uFB01}ne the user personality by observing behavioral characteristics derived from smartphone logs and the language used in text messages. Hence, the supervised machine learning methods (K-nearest neighbor (KNN), support vector machine, and Naive Bayes) and text mining techniques are used in studying the textual matter messages. The second approach is an analytical mehtodology was built based on association rules and alternative clustering techniques to predict personality traits based on SMS text