Tutorial 3

Scientists have long dreamed of creating machines humans could interact with by voice. Although one no longer believes Turing’s prophecy that machines will be able to converse like humans in the near future, real progress has been made in the voice and text-based human-machine interaction. After five decades of research, natural language understanding and prediction technology has become an essential part of many human-machine interaction systems (and even human-to-human: automated translation and speech-to-speech systems). There are now voice-based personal assistants, search and transactional systems for most smart phone platforms. The technology is pushed even further by the search engines which have evolved from simple keyword search to semantic search (they can now provide direct answers to a wide range of questions).

This tutorial is aimed at providing the machine learning and data mining community an overview of the deployed natural language technologies and their historical evolution. We review two fundamental problems involving natural language: the language prediction problem and the language understanding problem. The presentation focuses on the theory and algorithms used to build voiced/text-based human-computer interaction systems from the early automated directory assistance to today’s smart-phone virtual assistants and semantic web search.