作者: Rebecca Deneckère , Charlotte Hug , Ghazaleh Khodabandelou , Camille Salinesi
DOI: 10.4018/IJISMD.2014100102
关键词: Process mining 、 Data science 、 Process (engineering) 、 Action (philosophy) 、 Goal modeling 、 Machine learning 、 Artificial intelligence 、 Originality 、 Supervised learning 、 TRACE (psycholinguistics) 、 Computer science 、 Reverse engineering
摘要: Understanding people's goals is a challenging issue that met in many different areas such as security, sales, information retrieval, etc. Intention Mining aims at uncovering intentions from observations of actual activities. While most techniques proposed so far focus on mining individual to analyze web engine queries, this paper proposes generic technique mine activity traces. The relies supervised learning and generates intentional models specified with the Map formalism. originality contribution lies demonstration it actually possible reverse engineer underlying plans built by people when action, specify them e.g. levels, dependencies, links other concepts, After an introduction intention mining, presents Supervised Miner Method reports two controlled experiments were undertaken evaluate precision, recall F-Score. results are promising since authors able find activities well corresponding map process model satisfying accuracy, efficiency performance.