• Journal of Multimedia Information System
• /
• v.9 no.2
• /
• pp.93-102
• /
• 2022

Dependency parsing is a decision problem of the syntactic relation between words in a sentence. Recently, deep learning models are used for dependency parsing based on the word representations in a continuous vector space. However, it causes a mislabeled tagging problem for the proper nouns that rarely appear in the training corpus because it is difficult to express out-of-vocabulary (OOV) words in a continuous vector space. To solve the OOV problem in dependency parsing, we explored the proper noun embedding method according to the embedding unit. Before representing words in a continuous vector space, we replace the proper nouns with a special token and train them for the contextual features by using the multi-layer bidirectional LSTM. Two models of the syllable-based and morpheme-based unit are proposed for proper noun embedding and the performance of the dependency parsing is more improved in the ensemble model than each syllable and morpheme embedding model. The experimental results showed that our ensemble model improved 1.69%p in UAS and 2.17%p in LAS than the same arc-eager approach-based Malt parser.

• Journal of KIISE:Software and Applications
• /
• v.37 no.6
• /
• pp.486-490
• /
• 2010

Incremental parsing has been researched in the intention that the parse result of the original string is reused in the parsing of a new string. This paper proposes an improved method of the previous incremental LL(1) parser with nonterminal lookahead symbols. The previous work is time-inefficient because it repeatedly performs unnecessary steps when an error occurs. This paper gives a solution for the problem.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.7
• /
• pp.291-298
• /
• 2022

In this paper, we present a method for dependency parsing with chunking in Korean. Dependency parsing is a task of determining a governor of every word in a sentence. In general, we used to determine the syntactic governor in Korean and should transform the syntactic structure into semantic structure for further processing like semantic analysis in natural language processing. There is a notorious problem to determine whether syntactic or semantic governor. For example, the syntactic governor of the word "먹고 (eat)" in the sentence "밥을 먹고 싶다 (would like to eat)" is "싶다 (would like to)", which is an auxiliary verb and therefore can not be a semantic governor. In order to mitigate this somewhat, we propose a Korean dependency parsing after chunking, which is a process of segmenting a sentence into constituents. A constituent is a word or a group of words that function as a single unit within a dependency structure and is called a chunk in this paper. Compared to traditional dependency parsing, there are some advantage of the proposed method: (1) The number of input units in parsing can be reduced and then the parsing speed could be faster. (2) The effectiveness of parsing can be improved by considering the relation between two head words in chunks. Through experiments for Sejong dependency corpus, we have shown that the USA and LAS of the proposed method are 86.48% and 84.56%, respectively and the number of input units is reduced by about 22%p.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.5 no.1
• /
• pp.44-48
• /
• 1980

This paper proposes a method of the automatic recognition of the handprinted Hanguel patterns. A certain pattern oriented basic letters is normalized to the prototype Hanguel patten by the linking compansation and nomalization algorithm. Tree grammar is used in recognition process. Compared with the previous method. automata processing is simplified and the error is reduced by the new parsing method. This method shows the effectiveness for the constrained pattern. We expect that the new parsing method is very useful for the on-line pattern recognition.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.8
• /
• pp.2113-2123
• /
• 1999

The partial free order feature of Korean makes grammar size represented by CFG too big and that's why grammar has to contain all the ordered words. There are some problems to parse spoken language, because spontaneous spoken language has special features such as meaningless words, repetitious speech, etc. So, in this paper, we define 'Core-Concept' as the necessary element for parsing and we describe grammar only using Core-Concept. And we can prevent grammar from becoming very large and reduce an additional parsing burden as we select. Core-Concept described in grammar as parsing element. Through this strategy, we present that the simplified grammar can give us more efficient method to get right results. Experiments show that our parsing strategy has an average of 98% or over success rate in correct parsing results.

• Journal of KIISE:Software and Applications
• /
• v.31 no.5
• /
• pp.651-657
• /
• 2004

This paper proposes incremental LR parsers which are constructed by using pre-determinable reduction goal. The new method requires a relatively small number of memory space and computing time compared to the previous methods.

• The KIPS Transactions:PartB
• /
• v.13B no.3 s.106
• /
• pp.345-352
• /
• 2006

The practical English-Korean machine translation system should be able to translate long sentences quickly and accurately. The intra-sentence segmentation method has been proposed and contributed to speeding up the syntactic analysis. This paper proposes the syntactic category prediction method using decision trees for getting accurate parsing results. In parsing with segmentation, the segment is separately parsed and combined to generate the sentence structure. The syntactic category prediction would facilitate to select more accurate analysis structures after the partial parsing. Thus, we could improve the parsing accuracy by the prediction. We construct features for predicting syntactic categories from the parsed corpus of Wall Street Journal and generate decision trees. In the experiments, we show the performance comparisons with the predictions by human-built rules, trigram probability and neural networks. Also, we present how much the category prediction would contribute to improving the translation quality.

• Journal of Korea Multimedia Society
• /
• v.11 no.5
• /
• pp.703-709
• /
• 2008

The methodology of the compiler construction improved by well-defined parsing techniques and developments of automatic generation tools. Through them, a variety of compilers for the special applications can be developed effectively: particularly, the compiler for embedded/mobile devices. Also, as contents industry is proliferating recently, the necessity of developing a compiler which is suitable for contents system is highly increasing. These various demands can be resolved by modular techniques and automatic construction of compilers. But, optimization of compiler itself as development tools uses heuristic methods and it needs higher cost. In this paper, we suggest the parsing method which can decrease unnecessary reduce actions by analyzing the characteristics of LR parser. The suggested parsing technique uses lookahead/ states, reachable reduction goals information in parsing process and enhances the parsing efficiency by changing continuous reduce actions to one. Actually, we applied it to the front-end of ANSI C compiler and proved the parsing performance in terms of the number of reduce actions.

• Journal of Internet Computing and Services
• /
• v.21 no.5
• /
• pp.21-29
• /
• 2020

Accurate estimation of human pose relies on backbone method in which its role is to extract feature map. Up to dated, the method of backbone feature extraction is conducted by the plain convolutional neural networks named by CNN and the residual neural networks named by Resnet, both of which have various architectures and performances. The CNN family network such as VGG which is well-known as a multiple stacked hidden layers architecture of deep learning methods, is base and simple while Resnet which is a bottleneck layers architecture yields fewer parameters and outperform. They have achieved inspired results as a backbone network in human pose estimation. However, they were used then followed by different pose estimation networks named by pose parsing module. Therefore, in this paper, we present a comparison between the plain CNN family network (VGG) and bottleneck network (Resnet) as a backbone method in the same pose parsing module. We investigate their performances such as number of parameters, loss score, precision and recall. We experiment them in the bottom-up method of human pose estimation system by adapted the pose parsing module of openpose. Our experimental results show that the backbone method using VGG network outperforms the Resent network with fewer parameter, lower loss score and higher accuracy of precision and recall.

• Journal of Digital Contents Society
• /
• v.17 no.3
• /
• pp.203-210
• /
• 2016

Korean is free of omission of sentence elements and modifying scope, so managing it on morphological analyzer is better than parser. In this paper, we propose functional expansion methods of the morphological analyzer to ease the burden of parsing. This method is a longest phrase matching method. When the series of several morpheme have one syntax category by processing of Unknown-words, Compound verbs, Compound nouns, Numbers and Symbols, our method combines them into a syntactic unit. And then, it is to treat by giving them a semantic features as syntax unit. The proposed morphological analysis method removes unnecessary morphological ambiguities and deceases results of morphological analysis, so improves accuracy of tagger and parser. By empirical results, we found that our method deceases 73.4% of Parsing tree and 52.4% of parsing time on average.