Improving probabilistic automata learning with additional knowledge
(2004) Structural, Syntactic, and Statistical Pattern Recognition, Joint IAPR International Workshops, SSPR 2004 and SPR 2004 — ISBN: [3-540-22570-6], p. 260-268, published
Files
improving.pdf
Restricted Access - Adobe PDF
- 187.45 KB
Details
- Authors
- Abstract
- In this paper, we propose a way of incorporating additional knowledge in probabilistic automata inference, by using typed automata. We compare two kinds of knowledge that are introduced into the learning algorithms. A statistical clustering algorithm and a part-of-speech tagger are used to label the data according to statistical or syntactic information automatically obtained from the data. The labeled data is then used to infer correctly typed automata. The inference of typed automata with statistically labeled data provides language models competitive with state-of-the-art n-grams on the Air Travel Information System (ATIS) task.
- Affiliations
Université de Montreal, Canada Departement IRO Université Jean Monnet, Saint-Etienne, France Eurise
Citations
Kermorvant, C., de la Higuera, C., & Dupont, P. (2004). Improving probabilistic automata learning with additional knowledge. In Fred, A.; Caelli, T.; Duin, R.P.W.; Campilho, A.; Ridder, D.d. (ed.), Structural, Syntactic, and Statistical Pattern Recognition, Joint IAPR International Workshops, SSPR 2004 and SPR 2004 (p. p. 260-268). Springer-Verlag. https://hdl.handle.net/2078.5/253985