A neural network that learns from fuzzy data for language acquisition

Chin Teng Lin*, Ming Chih Kan, I. Fang Chung

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

1 Scopus citations


This paper proposes a four-layered fuzzy language acquisition network (FLAN) for acquiring fuzzy language. It can catch the intended information from a sentence (command) spoken in natural language with fuzzy terms. The intended information includes a meaningful semantic action and the fuzzy linguistic information of that action (for example, the phrase "move forward" represents the meaningful semantic action and the phrase "very high speed" represents the linguistic information in the fuzzy command "Move forward in a very high speed."). The proposed FLAN has two important features. First, we can make no restrictions whatever on the fuzzy language input which is used to specify the desired information, and the network requires no acoustic, prosodic, grammar and syntactic structure. Second, the linguistic information of an action is learned automatically and it is represented by fuzzy numbers based on α-level sets. A supervised learning scheme is proposed to train the FLAN on fuzzy training data. This learning scheme consists of the mutual-information (MI) supervised learning algorithm for learning meaningful semantic actions, and the fuzzy backpropagation (FBP) learning algorithm for learning linguistic information. An experimental system is constructed to illustrate the performance and applicability of the proposed FLAN.

Original languageEnglish
Pages (from-to)581-603
Number of pages23
JournalInternational Journal of Uncertainty, Fuzziness and Knowlege-Based Systems
Issue number6
StatePublished - Dec 1996


Dive into the research topics of 'A neural network that learns from fuzzy data for language acquisition'. Together they form a unique fingerprint.

Cite this