• The Journal of the Acoustical Society of Korea
• /
• v.18 no.5
• /
• pp.44-51
• /
• 1999

This paper presents a massively parallel computational model for the efficient integration of speech and natural language understanding. The phoneme model is based on continuous Hidden Markov Model with context dependent phonemes, and the language model is based on a knowledge base approach. To construct the knowledge base, we adopt a hierarchically-structured semantic network and a memory-based parsing technique that employs parallel marker-passing as an inference mechanism. Our parallel speech recognition algorithm is implemented in a multi-Transputer system using distributed-memory MIMD multiprocessors. Experimental results show that the parallel speech recognition system performs better in recognition accuracy than a word network-based speech recognition system. The recognition accuracy is further improved by applying code-phoneme statistics. Besides, speedup experiments demonstrate the possibility of constructing a realtime parallel speech recognition system.

• KIISE Transactions on Computing Practices
• /
• v.23 no.4
• /
• pp.226-231
• /
• 2017

Word representation, an important area in natural language processing(NLP) used machine learning, is a method that represents a word not by text but by distinguishable symbol. Existing word embedding employed a large number of corpora to ensure that words are positioned nearby within text. However corpus-based word embedding needs several corpora because of the frequency of word occurrence and increased number of words. In this paper word embedding is done using dictionary definitions and semantic relationship information(hypernyms and antonyms). Words are trained using the feature mirror model(FMM), a modified Skip-Gram(Word2Vec). Sense similar words have similar vector. Furthermore, it was possible to distinguish vectors of antonym words.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.5
• /
• pp.19-29
• /
• 2016

This paper suggests a method to refine a massive collective intelligence data, and visualize with multilevel sentiment network, in order to understand information in an intuitive and semantic way. For this study, we first calculated a frequency of sentiment words from each movie review. Second, we designed a Heatmap visualization to effectively discover the main emotions on each online movie review. Third, we formed a Sentiment-Movie Network combining the MDS Map and Social Network in order to fix the movie network topology, while creating a network graph to enable the clustering of similar nodes. Finally, we evaluated our progress to verify if it is actually helpful to improve user cognition for multilevel analysis experience compared to the existing network system, thus concluded that our method provides improved user experience in terms of cognition, being appropriate as an alternative method for semantic understanding.

• The KIPS Transactions:PartB
• /
• v.12B no.7 s.103
• /
• pp.829-840
• /
• 2005

Word Sense Disambiguation(WSD) is one of the most difficult problem in Korean information processing. We propose a WSD model with the capability to filter semantic information using the specific characteristics in dictionary dictions, and nth added information, useful to sense determination, such as statistical, distance and case information. we propose a model, which can resolve the issues resulting from the scarcity of semantic information data based on the word hierarchy system (thesaurus) developed by Ulsan University's UOU Word Intelligent Network, a dictionary-based toxicological database. Among the WSD models elaborated by this study, the one using statistical information, distance and case information along with the thesaurus (hereinafter referred to as 'SDJ-X model') performed the best. In an experiment conducted on the sense-tagged corpus consisting of 1,500,000 eojeols, provided by the Sejong project, the SDJ-X model recorded improvements over the maximum frequency word sense determination (maximum frequency determination, MFC, accuracy baseline) of $18.87\%$ ($21.73\%$ for nouns and inter-eojeot distance weights by $10.49\%$ ($8.84\%$ for nouns, $11.51\%$ for verbs). Finally, the accuracy level of the SDJ-X model was higher than that recorded by the model using only statistical information, distance and case information, without the thesaurus by a margin of $6.12\%$ ($5.29\%$ for nouns, $6.64\%$ for verbs).

• 중국학논총
• /
• no.71
• /
• pp.41-59
• /
• 2021

With the progress of social economy and science and technology, network media technology has developed rapidly, China has ushered in the network information age, and the network buzzwords emerged to reflect the interaction and influence between language and society. The network buzzwords of "X+ ren "indirectly show the social psychology and value orientation of modern people with their unique structural characteristics, semantic connotation and cultural deposits, and so on. Based on this, we have conducted a multi-angle investigation on the network buzzwords "X+ ren". This paper first analyzes the structure types and syntactic functions of the lexical model of "X+ ren ", then makes a semantic analysis of the lexical model of "X+ Ren ", and finally investigates the causes and influences of the popularity of "X+ ren ". Through the investigation, we believe that "X+ ren "will continue to grow, and "X+ ren" will continue to attract the attention of the academic community.

• Journal of Information Processing Systems
• /
• v.19 no.3
• /
• pp.355-369
• /
• 2023

Since the widespread adoption of deep-learning and related distributed representation, there have been substantial advancements in part-of-speech (POS) tagging for many languages. When training word representations, morphology and shape are typically ignored, as these representations rely primarily on collecting syntactic and semantic aspects of words. However, for tasks like POS tagging, notably in morphologically rich and resource-limited language environments, the intra-word information is essential. In this study, we introduce a deep neural network (DNN) for POS tagging that learns character-level word representations and combines them with general word representations. Using the proposed approach and omitting hand-crafted features, we achieve 90.47%, 80.16%, and 79.32% accuracy on our own dataset for three morphologically rich languages: Uyghur, Uzbek, and Kyrgyz. The experimental results reveal that the presented character-based strategy greatly improves POS tagging performance for several morphologically rich languages (MRL) where character information is significant. Furthermore, when compared to the previously reported state-of-the-art POS tagging results for Turkish on the METU Turkish Treebank dataset, the proposed approach improved on the prior work slightly. As a result, the experimental results indicate that character-based representations outperform word-level representations for MRL performance. Our technique is also robust towards the-out-of-vocabulary issues and performs better on manually edited text.

• Proceedings of the Korea Contents Association Conference
• /
• 2006.11a
• /
• pp.547-550
• /
• 2006

The criterion which judges an information retrieval system performance is to how many accurately retrieve an information that the user wants. The search result which uses only homograph has been appears the various documents that relates to each meaning of the word or intensively appears the documents that relates to specific meaning of it. So in this paper, we suggest semantic information retrieval technique using relation within User-Word Intelligent Network(U-WIN) to solve a disambiguation of query In our experiment, queries divide into two classes, the homograph used in terminology and the general homograph, and it sets the expansion query forms at "query + hypemym". Thus we found that only web document search's precision is average 73.5% and integrated search's precision is average 70% in two portal site. It means that U-WIN-Based semantic information retrieval technique can be used efficiently for a IR system.

• Journal of KIISE:Software and Applications
• /
• v.36 no.1
• /
• pp.92-108
• /
• 2009

The Princeton WordNet (PWN), which was developed during last 20 years since the mid 80, aimed at representing a mental lexicon inside the human mind. Its potentiality, applicability and portability were more appreciated in the fields of NLP and KE than in cognitive psychology. The semantic and knowledge processing is indispensable in order to obtain useful information using human languages, in the CMC and HCI environment. The PWN is able to provide such NLP-based systems with 'concrete' semantic units and their network. Referenced to the PWN, about 50 wordnets of different languages were developed during last 10 years and they enable a variety of multilingual processing applications. This paper aims at describing PWN-referenced Korean Wordnet, KorLex 1.5, which was developed from 2004 to 2007, and which contains currently about 130,000 synsets and 150,000 word senses for nouns, verbs, adjectives, adverbs, and classifiers.

• The Journal of the Korea Contents Association
• /
• v.15 no.7
• /
• pp.578-590
• /
• 2015

This paper investigated how high school students perceive on nuclear power applying semantic network analysis. The total number of 250 high school students in 6 metropolitan areas responded to the survey. Results indicate that the word which most frequently appeared is 'Nuclear plant' (87 times) following by 'Japan' (71 times), 'Danger'(59 times), 'Fukushima' (59 times), 'Radioactivity' (56 times), 'Energy' (47 times), 'Youngkwang' (37 times), 'Electricity' (30times), 'Chernobyl' (29 times), 'Explosion' (25 times). For students in higher acceptance level of nuclear 'Radioactivity' (25 times) was most frequently showed up following by 'Fukushima' (23 times), 'Energy' (21 times), 'Japan' (21 times), Nuclear plant' (20 times), 'Danger' (17 times), 'Youngkwang' (16 times). For student in lower acceptance level of nuclear, the word of 'Nuclear plant' (40 times) most frequently appeared following by 'Japan' (31 times)', 'Danger' (29 times), 'Fukushima' (23 times), 'Radioactivity' (17 times), 'Energy' (16 times), 'Youngkwang' (16 times), 'Chernobyl' (15 times).

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.28 no.3
• /
• pp.217-240
• /
• 2017

The purpose of this study was to investigate the semantic network analysis to understand image contents and to examine the degree to which words, word clusters contributed to the formation of semantic map within image contents. For this research, from 1993 until 2016 the field of the image contents were collected for a total of 2,624 cases papers. The word appeared in Title analyzed the social network by using the R program of Big Data. The results were as follows: First, The field of image contents is based on researches related to 'image', 'media' and 'contents'. Second, there is a three-step flow ('education' -> 'media' -> 'contents') of research in the field of image contents. Third, researches related to 'broadcasting', 'digital', 'technology', and 'production' were continuously carried out. Finally, There were new research subjects for each regime.