Combining Machine Learning and Rule-based Approaches in Spanish and Japanese Sentence Realization
نویسندگان
چکیده
In this paper we describe two parallel experiments on the integration of machine learning (ML) methods into the Spanish and Japanese rule-based sentence realization modules developed at Microsoft Research. The paper explores the use of decision trees (DT) for the lexical selection of the copula in Spanish and the insertion of a locative postposition in Japanese. We show that it is possible to machine-learn the contexts for these two non-trivial linguistic phenomena with high accuracy.
منابع مشابه
A Hybrid Machine Translation System Based on a Monotone Decoder
In this paper, a hybrid Machine Translation (MT) system is proposed by combining the result of a rule-based machine translation (RBMT) system with a statistical approach. The RBMT uses a set of linguistic rules for translation, which leads to better translation results in terms of word ordering and syntactic structure. On the other hand, SMT works better in lexical choice. Therefore, in our sys...
متن کاملPhonological Awareness Impact on Articulatory Accuracy of the Spanish Liquid [r] in Japanese FL Learners of Spanish
Foreign language learners tend to avoid phonological difficulties and simply transfer sounds whether from their L1 or any pre-existing L2. Phonological awareness (PA) gives students an active role in understanding their own potential in improving pronunciation through several methods. However, such methods are likely to be restricted to only passive learning methods, such as repetition, reading...
متن کاملPhonological Awareness Impact on Articulatory Accuracy of the Spanish Liquid [r] in Japanese FL Learners of Spanish
Foreign language learners tend to avoid phonological difficulties and simply transfer sounds whether from their L1 or any pre-existing L2. Phonological awareness (PA) gives students an active role in understanding their own potential in improving pronunciation through several methods. However, such methods are likely to be restricted to only passive learning methods, such as repetition, reading...
متن کاملCombining decision trees and transformation-based learning to correct transferred linguistic representations
We present a hybrid machine learning approach to correcting features in transferred linguistic representations in machine translation. The hybrid approach combines decision trees and transformation-based learning. Decision trees serve as a filter on the intractably large search space of possible interrelations among features. Transformation-based learning results in a simple set of ordered rule...
متن کاملThe Effect of Isolated vs. Combined Processing Instruction and Output-Based Instruction on the Learning of English Passives
The studies on the merits of processing instruction (PI) and output-based instruction (OI) have mostly treated the two approaches as mutually exclusive. To address the potentials of combining interpretation and production activities, this research compared the two isolated approaches of PI and OI with two combined approaches in which processing and output tasks were used in two opposite orders ...
متن کامل