• Text Analytics: Motivation – Two Challenges • Representation • Labels • Text Categorization via HIN – HIN construction from texts – From HIN similarity to clustering and classification – World knowledge indirect supervision • Conclusions and future work