Learning management systems are widely used as a support of distance learning. Recently, these systems successfully help in present education as well. Learning management systems store large amount of data based on the history of users’ interactions with the system. Obtained information is commonly used for further course optimization, finding e-tutors in collaboration learning, analysis of students’ behavior, or for other purposes. The partial goal of the paper is an analysis of students’ behavior in a learning management system. Students’ behavior is defined using selected methods from sequential and process mining with the focus to the reduction of large amount of extracted sequences. The main goal of the paper is description of our Left-Right Oscillate algorithm for community detection. The usage of this algorithm is presented on the extracted sequences from the learning management system. The core of this work is based on spectral ordering. Spectral ordering is the first part of an algorithm used to seek out communities within selected, evaluated networks. More precise designations for communities are then monitored using modularity.