"The InPhO analyzes over 37 million words of philosophical content from: Stanford Encyclopedia of Philosophy; Internet Encyclopedia of Philosophy; PhilPapers; HathiTrust/Google Books Collection. The InPhO is constructed by a novel 3-step semi-automated process, called dynamic ontology: 1. Data Mining — Natural language processing (NLP) techniques generate statistical hypothesis about the relationships among philosophically-salient topics. 2. Expert Feedback — These hypotheses are evaluated by domain experts through online interfaces. 3. Machine Reasoning — Our machine reasoning program uses feedback and statistical measures to populate the ontology."