• Journal of KIISE
• /
• v.44 no.8
• /
• pp.822-831
• /
• 2017

In this paper, we propose a Korean dependency parsing model using multi-task learning based pointer networks. Multi-task learning is a method that can be used to improve the performance by learning two or more problems at the same time. In this paper, we perform dependency parsing by using pointer networks based on this method and simultaneously obtaining the dependency relation and dependency label information of the words. We define five input criteria to perform pointer networks based on multi-task learning of morpheme in dependency parsing of a word. We apply a fine-tuning method to further improve the performance of the dependency parsing proposed in this paper. The results of our experiment show that the proposed model has better UAS 91.79% and LAS 89.48% than conventional Korean dependency parsing.

• Journal of KIISE:Software and Applications
• /
• v.37 no.11
• /
• pp.845-850
• /
• 2010

Incremental parsing has been developed to reuse the parse result of the original string during the parsing of a new string. The previous incremental LL(1) parsing methods precomputed the reusable point information before parsing and used it during parsing. This paper proposes an efficient reusable point computation by factoring the common part of the computation. The common symbol storing method and the distance storing method were previously suggested to find the reusable point, and by combining the methods, this paper gives the storing method of the distance to common symbols. Based on it, an efficient incremental LL(1) parser is constructed.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.2
• /
• pp.524-543
• /
• 2022

Correct facade image parsing is essential to the semantic understanding of outdoor scenes. Unfortunately, there are often various occlusions in front of buildings, which fails many existing methods. In this paper, we propose an end-to-end deep network for facade parsing with occlusions. The network learns to decompose an input image into visible and invisible parts by occlusion reasoning. Then, a context aggregation module is proposed to collect nonlocal cues for semantic segmentation of the visible part. In addition, considering the regularity of man-made buildings, a repetitive pattern completion branch is designed to infer the contents in the invisible regions by referring to the visible part. Finally, the parsing map of the input facade image is generated by fusing the results of the visible and invisible results. Experiments on both synthetic and real datasets demonstrate that the proposed method outperforms state-of-the-art methods in parsing facades with occlusions. Moreover, we applied our method in applications of image inpainting and 3D semantic modeling.

• Journal of KIISE:Software and Applications
• /
• v.41 no.7
• /
• pp.518-522
• /
• 2014

A single state parsing automaton has the characteristics of the decision of actions which do not depend on the history of the parsing paths but on the current state. The single state parsing automaton hence has the advantage of the reduced parsing time and a small memory requirement compared to those of the conventional LR automaton. However, currently, the grammar classes generating single state parsing automata have not been known. This paper deals with the grammar classes generating single state parsing automata; in addition, this paper gives the generating method of single state parsing automata of the grammar classes.

• ETRI Journal
• /
• v.28 no.6
• /
• pp.803-806
• /
• 2006

This letter addresses a structural disambiguation method for Korean adverbs based on the correlative relation constraints between adverbs and modifiees, and the morphological context information of sentences. Using the proposed method, we improved the dependency parsing accuracy of adverbs from 79.2 to 89%. The experimental result shows that the proposed method is especially expert in parsing adverbs which can modify multiple word classes or have a long distance dependency relation to their modifiees.

• Journal of Korea Society of Industrial Information Systems
• /
• v.9 no.4
• /
• pp.1-7
• /
• 2004

In this thesis, we introduce a parsing method which use information of the post position in Korean to get the exact parsing tree. In order to implement this method we classified categories of the predicates, and defined sentence-types based on these categories. We tried to make parsing using the method grasping the grammatical role of the noun phrase that have to exist in each sentence-type. In parser control mechanism, we use some heuristics based on linguistic frame. We use conditional unification to implement analysis. It is Possible to reduce ambiguous because the parsing method suggested helps to Prune the branches which are unnecessary.

• Journal of KIISE
• /
• v.44 no.3
• /
• pp.239-245
• /
• 2017

Probabilistic grammar is used in natural language processing, and the parse result of the grammar has to preserve the probability of the original grammar. As for the representative parsing method, LL parsing and LR parsing, the former preserves the probability information of the original grammar, but the latter does not. A characteristic of a probabilistic parsing automaton has been studied; but, currently, the generating model of probabilistic parsing automata has not been known. The paper provides a model of probabilistic parsing automata based on the single state parsing automata. The generated automaton preserves the probability of the original grammar, so it is not necessary to test whether or not the automaton is probabilistic parsing automaton; defining a probability function for the automaton is not required. Additionally, an efficient automaton can be constructed by choosing an appropriate parameter.

• Journal of KIISE
• /
• v.43 no.10
• /
• pp.1079-1084
• /
• 2016

Single-state parsing automata have a characteristic such that the decision of an action depends only on the current state but not on the parsing history. The memory space and the parsing time of single-state parsing automata are less than the memory space and the parsing time of LR automata. However, the applicable grammar class of single-state parsing automata is less than that of LR automata. This paper provides extended single-state parsing automata, which are applicable to LR grammars. In the prior work, the special state, referred to as the cyclic state was not treated in the construction of single-state parsing automata, and hence, the applicable grammar class was less than LR grammars. The paper solves the problem of cyclic states by processing dynamic information depending on an input string. The proposed method expands the application of grammar class of single-state parsing automata to LR grammars.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.1
• /
• pp.21-28
• /
• 2016

The performance of an English-Korean machine translation system depends heavily on its English parser. The parser in this paper is a part of the rule-based English-Korean MT system, which includes many syntactic rules and performs the chart-based parsing. The parser generates too many structures due to many syntactic rules, so much time and memory are required. The rule-based parser has difficulty in analyzing and translating the long sentences including the commas because they cause high parsing complexity. In this paper, we propose the 3-phase parsing method with sentence segmentation to efficiently translate the long sentences appearing in usual. Each phase of the syntactic analysis applies its own independent syntactic rules in order to reduce parsing complexity. For the purpose, we classify the syntactic rules into 3 classes and design the 3-phase parsing algorithm. Especially, the syntactic rules in the 3rd class are for the sentence structures composed with commas. We present the automatic rule acquisition method for 3rd class rules from the syntactic analysis of the corpus, with which we aim to continuously improve the coverage of the parsing. The experimental results shows that the proposed 3-phase parsing method is superior to the prior parsing method using only intra-sentence segmentation in terms of the parsing speed/memory efficiency with keeping the translation quality.

• Korean Journal of Cognitive Science
• /
• v.15 no.1
• /
• pp.25-33
• /
• 2004

This paper discusses two problems with LR parsing with regard to constructing parsing tables with feature-based grammars. First, we show that traditional LR parsing methods suffer from nontermination and nondeterminism problems when they are applied to feature-based grammars. We then present an LR method for feature-based grammars that avoids both nontermination and nondetermisim by making use of partial information of a feature structure. Second, we describe the problem of adapting LR parsing to feature-based grammars with schematic rules (i.e., rules that do not contain enough information to construct parsing tables). To remedy this problem, we propose a rule inference algorithm which instantiates underspecified rules into more specified ones containing enough information.