• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.7-10
• /
• 2000

Studies of prosody and sentence processing have demonstrated that prosodic phrasing can exhibit strong effects on processing decisions in English. In this paper, we tested Korean sentence fragments containing syntactically ambiguous Adj-N1-N2 strings in a cross-modal naming task. Four accentual phrasing patterns were tested: (a) the default phrasing pattern, in which each word forms an accentual phrase; (b) a phrasing biased toward N1 modification; (c) a phrasing biased toward complex-NP modification; and (d) a phrasing used with adjective focus. Patterns (b) and (c) are disambiguating phrasings; the other two are commonly found with both interpretations and are thus ambiguous. The results showed that the naming time of items produced in the prosody contradicting the semantic grouping is significantly longer than that produced in either default or supporting prosody, We claim that, as in English, prosodic information in Korean is parsed into a well-formed prosodic representation during the early stages of processing. The partially constructed prosodic representation produces incremental effects on syntactic and semantic processing decisions and is retained in memory to influence reanalysis decisions.

• Journal of Information Processing Systems
• /
• v.11 no.2
• /
• pp.184-204
• /
• 2015

Fuzzy Formal Concept Analysis (FCA) is a mathematical tool for the effective representation of imprecise and vague knowledge. However, with a large number of formal concepts from a fuzzy context, the task of knowledge representation becomes complex. Hence, knowledge reduction is an important issue in FCA with a fuzzy setting. The purpose of this current study is to address this issue by proposing a method that computes the corresponding crisp order for the fuzzy relation in a given fuzzy formal context. The obtained formal context using the proposed method provides a fewer number of concepts when compared to original fuzzy context. The resultant lattice structure is a reduced form of its corresponding fuzzy concept lattice and preserves the specialized and generalized concepts, as well as stability. This study also shows a step-by-step demonstration of the proposed method and its application.

• 전기의세계
• /
• v.28 no.8
• /
• pp.66-71
• /
• 1979

This paper describes electrical analysis of the information processing of the nervous system. A general-purpose electrical neuronal model for simulating the electrical activity in a single nerve cell and in small groups of nerve cell has constructed. This model consists of two basic electronic modules to represent respectively a "cell body" and an "axon (with synapses)", together with various related appurtenances. The primary advantages of this method are; holistic view, actual physical representation of various electrical activities in a single nerve cell, display of the activity of all nerve cells flexibility with respect to network parameters. Moreover, this model can effectively help push forward our general ability to explore and conceptualize the electrical activity of interconnected networks of nerve cell behaving in concert. Also, this electronic module technique is the best of various means for this task of realistic representation of aggregates of neurons.gregates of neurons.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1996.04a
• /
• pp.709-712
• /
• 1996

Generating a project network of a specific construction project is very time consuming and difficult task in the field. To effectiviely automate and support the planning process, we design a case-based project planning expert system inspired by the fact a human expert project planner uses previous cases for planning a new project. A construction project case consist of its specific characteristics and the corresponding project network (i.e. project plan). Using frame based representation. we represent the project features affecting the progress network and the entities composing the project plan such as the buildings, construction methods, WBS (work breakdown structure), activities, and resources. The project planning process runs through most similar case retrieval, case adaptation, and user requirement satisfaction. We represent the construction domain knowledge for each procedure using constraints and rules. We develop the methodology for constraint-based case adaption. Case adaptation process mainly consist of activity generation/deletion and predecence constraint satisfaction, for which we develop the dynamic constraint generation method and connect user-level requirement representation the system-level network modification knowledge. The methodology is being applied to the prototype for apartment construction project planning.

• Proceedings of the KSPS conference
• /
• 2006.05a
• /
• pp.113-116
• /
• 2006

This research was investigate to understand how moderate bilingual subjects represent the lexicon in second language. Although most researches have focused only on high proficient bilinguals, we analysed how moderate bilinguals who have learned English mostly in school represent the prototype of verb and its inflected form of verb. Results of lexical decision task showed that moderate bilingual subjects used different mental representation depending on whether the verb have regular or irregular conjugation. With regular verbs, the identification of an inflected form was affected by both the frequency of its prototype and that of inflected form, but with irregular verbs, it is affected only by the frequency of inflected form.

• Proceedings of the KSPS conference
• /
• 2006.05a
• /
• pp.71-74
• /
• 2006

The purpose of this study is to observe how Korean listeners detect a target phoneme with 'Focus' represented by prosodic prominence and question-induced semantic emphasis. According to the automated phoneme detection task using E-Prime, Korean listeners detected phoneme targets more rapidly when the target-bearing words were in prominence position and in question-induced position. However, when phoneme targets were in prominence position, response time was much faster than in question-induced position. The results suggest that the prosodic prominence which is explicit method of focus representation be more effective than question-inducing, implicit method of it, in phoneme detecting.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.21 no.2
• /
• pp.71-93
• /
• 1996

Selection of appropriate knowledge acquisition and representation techniques is regarded as a major task in the development of expert systems. This depends on the characteristics of problem domains. Expert system builders have often adopted a technique without a formal analysis of application domains. The purpose of this paper is to provide the best knowledge aquisition and representation technique for use in human resource management problem domains. In an attempt to meet this purpose, the conceptual contingency model that suggests the best technique according to managerial decision categories is used as a guidance. In order to determine the priority of managerial decision categories, the Analytic Hierarchy Process and an extended method are proposed.

• Phonetics and Speech Sciences
• /
• v.7 no.3
• /
• pp.27-35
• /
• 2015

The purpose of this study is to examine the lexical representation of phonological variants derived from optional vs. obligatory phonological processes. Given that place assimilation is optionally processed, whereas nasal assimilation is obligatory in Korean, a long-term repetition priming experiment was conducted, using a shadowing task. Korean speakers shadowed words containing either assimilated or unassimilated consonants in three priming conditions and their shadow responses were evaluated. It was shown that in both place and nasal assimilations, shadowing latencies for unassimilated stimuli were longer than those for assimilated stimuli in the mismatched condition. These results suggest that even in the optional assimilation, assimilated variants were processed more easily and faster than the canonical variants. The present results argue against the frequency-based account of multiple lexical representation (Connine, 2004; Connine & Pinnow, 2006; Ranbom & Connine, 2007; $B{\ddot{u}rki$, Ernestus, & Frauenfelder, 2010; $B{\ddot{u}rki$, Alario, & Frauenfelder, 2011).

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.613-616
• /
• 2018

Sequence tagging is an important task in Natural Language Processing (NLP), in which the Named Entity Recognition (NER) is the key issue. So far the most widely adopted model for NER in NLP is that of combining the neural network of bidirectional long short-term memory (BiLSTM) and the statistical sequence prediction method of Conditional Random Field (CRF). In this work, we improve the prediction accuracy of the BiLSTM by supporting an aligned word representation mechanism. We have performed experiments on multilingual (English, Spanish and Dutch) datasets and confirmed that our proposed model outperformed the existing state-of-the-art models.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.7
• /
• pp.2407-2424
• /
• 2022

Recently, Transformer has made great progress in video retrieval tasks due to its high representation capability. For the structure of a Transformer, the cascaded self-attention modules are capable of capturing long-distance feature dependencies. However, the local feature details are likely to have deteriorated. In addition, increasing the depth of the structure is likely to produce learning bias in the learned features. In this paper, an improved Transformer structure named TransDCS (Transformer with Dynamic Convolution and Shortcut) is proposed. A Multi-head Conv-Self-Attention module is introduced to model the local dependencies and improve the efficiency of local features extraction. Meanwhile, the augmented shortcuts module based on a dual identity matrix is applied to enhance the conduction of input features, and mitigate the learning bias. The proposed model is tested on MSRVTT, LSMDC and Activity-Net benchmarks, and it surpasses all previous solutions for the video-text retrieval task. For example, on the LSMDC benchmark, a gain of about 2.3% MdR and 6.1% MnR is obtained over recently proposed multimodal-based methods.