Persian socio-cultural ontology for profile extraction

This is an ontology that incorporates commonsense knowledge along with socio-cultural information in Persian. Using this asset, we have developed a hybrid profile extraction method. By profile extraction we mean extracting user-specific profiles that include general personal information, such as hobbies, occupation, or age which is a valuable asset for systems like recommendation engines or personalized chatbots. Using this ontology, we can also detect conflicts between user's personal information such as conflict between user's claimed residential city and thier already mentioned residential country. As a test set of assessing our method's conflict detection performance, we build 100 sentences consisting of conflictive vs. non-conflictive sentences presented in excel file.