• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 추계종합학술대회 논문집(3)
• /
• pp.25-28
• /
• 2000

In this paper, an effective intention reading scheme is proposed for human-friendly interface. Soft computing techniques such as fuzzy logic and artificial neural networks are used for this. And Gabor filter based feature(GG feature) is also proposed to deal with local activity in the human face. It is based on human visual system and Gabor filter based approach is very popular in these days. The proposed scheme is adopted for human-friendly interface for rehabilitation service robotic system KARES II.

• ETRI Journal
• /
• 제39권6호
• /
• pp.832-840
• /
• 2017

Recently, deep recurrent neural networks have achieved great success in various machine learning tasks, and have also been applied for sound event detection. The detection of temporally overlapping sound events in realistic environments is much more challenging than in monophonic detection problems. In this paper, we present an approach to improve the accuracy of polyphonic sound event detection in multichannel audio based on gated recurrent neural networks in combination with auditory spectral features. In the proposed method, human hearing perception-based spatial and spectral-domain noise-reduced harmonic features are extracted from multichannel audio and used as high-resolution spectral inputs to train gated recurrent neural networks. This provides a fast and stable convergence rate compared to long short-term memory recurrent neural networks. Our evaluation reveals that the proposed method outperforms the conventional approaches.

• 대한음성학회지:말소리
• /
• 제49호
• /
• pp.123-134
• /
• 2004

An automatic pronunciation correction system provides learners with correction guidelines for each mispronunciation. In this paper we propose an HMM based speech recognizer which automatically classifies pronunciation errors when Koreans speak Japanese. We also propose two pronunciation networks for automatic detection of mispronunciation. In this paper, we evaluated performances of the networks by computing the correlation between the human ratings and the machine scores obtained from the speech recognizer.

• Journal of Information Processing Systems
• /
• 제15권1호
• /
• pp.1-6
• /
• 2019

Recently, big data and artificial intelligence (AI) based on communication systems have become one of the hottest issues in the technology sector, and methods of analyzing big data using AI approaches are now considered essential. This paper presents diverse paradigms to subjects which deal with diverse research areas, such as image segmentation, fingerprint matching, human tracking techniques, malware distribution networks, methods of intrusion detection, digital image watermarking, wireless sensor networks, probabilistic neural networks, query processing of encrypted data, the semantic web, decision-making, software engineering, and so on.

• International Journal of Computer Science & Network Security
• /
• 제24권9호
• /
• pp.111-118
• /
• 2024

The concept of social stratification and hierarchy among human dates back to the origin of human race. Presently, the growing reputation of social networks has given us with an opportunity to analyze these well-studied phenomena over different networks at different scales. Generally, a social network could be defined as a collection of actors and their interactions. In this work, we concern ourselves with a particular type of social networks, known as trust networks. In this type of networks, there is an explicit show of trust (positive interaction) or distrust (negative interaction) among the actors. In other words, an actor can designate others as friends or foes. Trust networks are typically modeled as signed networks. A signed network is a directed graph in which the edges carry an edge weight of +1 (indicating trust) or -1 (indicating distrust). Examples of signed networks include the Slashdot Zoo network, the Epinions network and the Wikipedia adminship election network. In a social network, actors tend to connect with each other on the basis of their perceived social hierarchy. The emergence of such a hierarchy within a social community shows the manner in which authority manifests in the community. In the case of signed networks, the concept of social hierarchy can be interpreted as the emergence of a tree-like structure comprising of actors in a top-down fashion in the order of their ranks, describing a specific parent-child relationship, viz. child trusts parent. However, owing to the presence of positive as well as negative interactions in signed networks, deriving such "trust hierarchies" is a non-trivial challenge. We argue that traditional notions (of unsigned networks) are insufficient to derive hierarchies that are latent within signed networks In order to build hierarchies in signed networks, we look at two interpretations of trust namely presence of trust (or "good") and lack of distrust (or "not bad"). In order to develop a hierarchy signifying both trust and distrust effectively, the above interpretations are combined together for calculating the overall trustworthiness (termed as deserve) of actors. The actors are then arranged in a hierarchical fashion based on these aggregate deserve values, according to the following hypothesis: actor v is assigned as a child of actor u if: (i) v trusts u, and (ii) u has a higher deserve value than v. We describe this hypothesis with additional qualifiers in this thesis.

• 한국생활과학회지
• /
• 제4권1호
• /
• pp.31-38
• /
• 1995

The purpose of this study was to examine the relationship between social networks and perceived self-competence. The subject were 350 third and sixth graders selected from a public elementary school in Seoul. The children were administered the Pattison Psychosocial Network Inventory and The Self Perception Profile for Children. Three major questions are addressed; l)age changes 2)sex differences, and 3) the relation between the characteristics of social networks and self-perceived competence. Data were analyzed by pearson's product moment correlation, one-way ANOVA. Proportion measures were also used since they had allowed for the observation of relationship between two subsets of people within given categories(e.g.,the number of relatives divided by the number of relatives plus nonrelatives). The effects of age and sex of child on the number and daily contact with kin, nankin, peers, adults, males, and females were found. Sixth graders had a larger number of adults than third graders. While children had a greater number of adults than peers, they had more contact with peers than with adults in their network. It was found that boys had greater number of people in their network than girls. And boys had a larger number of and more contact with males, and girls had a larger number of more contact with females. The structure of childrens' social networks structure were correlated with perceived self-competence.

• 인터넷정보학회논문지
• /
• 제20권2호
• /
• pp.101-108
• /
• 2019

Workflow management systems (WfMSs) automate and manage workflows, which are implementations of organizational processes operated in process-centric organizations. In this paper, wepropose an algorithm to discover temporal work transference networks from workflow execution logs. The temporal work transference network is a special type of enterprise social networks that consists of workflow performers, and relationships among them that are formed by work transferences between performers who are responsible in performing precedent and succeeding activities in a workflow process. In terms of analysis, the temporal work transference network is an analytical property that has significant value to be analyzed to discover organizational knowledge for human resource management and related decision-making steps for process-centric organizations. Also, the beginning point of implementinga human-centered workflow intelligence framework dealing with work transference networks is to develop an algorithm for discovering temporal work transference cases on workflow execution logs. To this end, we first formalize a concept of temporal work transference network, and next, we present a discovery algorithm which is for the construction of temporal work transference network from workflow execution logs. Then, as a verification of the proposed algorithm, we apply the algorithm to an XES-formatted log dataset that was released by the process mining research group and finally summarize the discovery result.

• 전자통신동향분석
• /
• 제36권3호
• /
• pp.106-118
• /
• 2021

The field of brain science (or neuroscience in a broader sense) has inspired researchers in artificial intelligence (AI) for a long time. The outcomes of neuroscience such as Hebb's rule had profound effects on the early AI models, and the models have developed to become the current state-of-the-art artificial neural networks. However, the recent progress in AI led by deep learning architectures is mainly due to elaborate mathematical methods and the rapid growth of computing power rather than neuroscientific inspiration. Meanwhile, major limitations such as opacity, lack of common sense, narrowness, and brittleness have not been thoroughly resolved. To address those problems, many AI researchers turn their attention to neuroscience to get insights and inspirations again. Biologically plausible neural networks, spiking neural networks, and connectome-based networks exemplify such neuroscience-inspired approaches. In addition, the more recent field of brain network analysis is unveiling complex brain mechanisms by handling the brain as dynamic graph models. We argue that the progress toward the human-level AI, which is the goal of AI, can be accelerated by leveraging the novel findings of the human brain network.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.1293-1298
• /
• 2005

In the environment where human coexists with robot, the problem of safety is very important. But it is difficult to separate the robot from the human in time-domain or space-domain unlike the case of factory automation, so a new concept is needed. One approach is to notice sensory and emotional feeling of human, and in this study "pain" is focused, which is a typical unpleasant feeling when the robot contacts us. In this paper, to design the controller based on the pain, an artificial superficial pain model caused by impact is proposed. This ASPM model consists of mechanical pain model, skin model and gate control by artificial neural networks (ANNs). The proposed ASPM is evaluated by experiments.

• 융합신호처리학회논문지
• /
• 제24권2호
• /
• pp.90-96
• /
• 2023

합성곱 신경망을 비롯하여 딥러닝 신경망의 학습에서 많은 양의 훈련데이터의 확보는 과적합 현상을 피하고 우수한 성능을 가지기 위해서 매우 중요하다. 하지만, 딥러닝 신경망에서의 레이블화된 훈련데이터의 확보는 실제로는 매우 제한적이다. 이를 극복하기 위해, 이미 획득한 훈련데이터를 변형, 조작 등으로 추가로 훈련데이터를 생성하는 여러 증강 방법이 제안되었다. 하지만, 이미지, 문자 등의 훈련데이터와 달리, 인간 동작 인식을 행하는 합성곱 신경망의 생체신호 훈련데이터를 추가로 생성하는 증강 방법은 연구 문헌에서 찾아보기 어렵다. 본 연구에서는 합성곱 신경망에 기반한 인간 동작 인식을 위한 생체신호 훈련데이터를 생성하는 간편하지만, 효과적인 증강 방법을 제안한다. 본 연구의 제안된 증강 방법의 유용성은 추가로 생성된 생체신호 훈련데이터로 학습하여 합성곱 신경망이 인간 동작을 높은 정확도로 인식하는 것을 보임으로써 검증하였다.