Communications of the ACM
Human-computer interaction
The effect of domain and text type on text prediction quality
EACL '12 Proceedings of the 13th Conference of the European Chapter of the Association for Computational Linguistics
| Hi-index | 0.00 |
There is a fast growing need for companies to handle full text interaction via internet with chat-like functionality, think of online helpdesk, demand aggregation, etc. However, even state-of-the-art language technology is not able to automatically handle consumer requests in a reliable and sensible manner. Semantic text understanding systems are brittle and expensive to maintain, and simple pattern matching text-bots cannot handle the complexity of natural language and often give useless or wrong answers. RE-phrase meets this need by starting from threads of real human textual interaction (chat, helpdesk) and then storing and re-cycling it for fast multiple-choice conversation with automated replies. It is a new technology for users to interact via the Internet, which combines browsing and chatting. It makes use of an adaptive and collaboratively growing database of conversation phrases. It is fast, adaptive, allows full text entry, and is well suited for different types of browsers and platforms.