Using implications from FCA to represent a two mode network data

In a world of ever-growing connectivity, full of connec- tions between people and objects, new multidisciplinary complex network analysis needs to arise. This work presents a solution to analyze an In- ternet Service Provider database using a formal concept analysis element named implications and complex network techniques. Our goal is to an- alyze access to the 25 most visited websites to find access patterns. We selected 9 time intervals in one week. Data were converted to a clarified formal context and the FindImplications algorithm was used to extract implications sets. These sets were cross-checked to look for patterns. The implications were used to explore the complex network substructures. As a result, we found access patterns that guarantee that whenever premise websites are accessed, so are conclusion websites. This result can aid in creating security policies and network configurations to help predict fu- ture accesses. Without this technique relationships between events nodes (websites) of a two mode network could not be identified.