• Title/Summary/Keyword: Feature selection

Search Result 1,076, Processing Time 0.033 seconds

Creation and labeling of multiple phonotopic maps using a hierarchical self-organizing classifier (계층적 자기조직화 분류기를 이용한 다수 음성자판의 생성과 레이블링)

  • Chung, Dam;Lee, Kee-Cheol;Byun, Young-Tai
    • The Journal of Korean Institute of Communications and Information Sciences
    • /
    • v.21 no.3
    • /
    • pp.600-611
    • /
    • 1996
  • Recently, neural network-based speech recognition has been studied to utilize the adaptivity and learnability of neural network models. However, conventional neural network models have difficulty in the co-articulation processing and the boundary detection of similar phonmes of the Korean speech. Also, in case of using one phonotopic map, learning speed may dramatically increase and inaccuracies may be caused because homogeneous learning and recognition method should be applied for heterogenous data. Hence, in this paper, a neural net typewriter has been designed using a hierarchical self-organizing classifier(HSOC), and related algorithms are presented. This HSOC, during its learing stage, distributed phoneme data on hierarchically structured multiple phonotopic maps, using Kohonen's self-organizing feature maps(SOFM). Presented and experimented in this paper were the algorithms for deciding the number of maps, map sizes, the selection of phonemes and their placement per map, an approapriate learning and preprocessing method per map. If maps are divided according to a priorlinguistic knowledge, we would have difficulty in acquiring linguistic knowledge and how to alpply it(e.g., processing extended phonemes). Contrarily, our HSOC has an advantage that multiple phonotopic maps suitable for given input data are self-organizable. The resulting three korean phonotopic maps are optimally labelled and have their own optimal preprocessing schemes, and also confirm to the conventional linguistic knowledge.

  • PDF

Understanding the Selective Attention and Animation Induction Device According to the Visual Capture of Audience (관객의 시각포획현상에 따른 선택적 주의집중과 애니메이션 유도장치의 이해)

  • Lee, Jong-Han
    • Cartoon and Animation Studies
    • /
    • s.41
    • /
    • pp.133-152
    • /
    • 2015
  • Some artists and scientists in physics and animation originating from research on its form of expression thanks to the rapid development of the example in the late 20th century image production technology integrating existing media feature, perform a re-creation and pop culture content has been recognized as a key factor. animation of the modern emphasis is also commercial and artistic activities as show whether the artist can not be excluded that also target audience. The audience does not want only to receive offers simply 'seeing' and 'hearing' in the animation requires a more indirect mental met. the other side, the director should lead the audience to immerse myself in work as intended mystification induce the world. where a conflict occurs between the audience and the director and The director needs to have its troubleshooting point to 'Technology of the communication'. Which is reduced to 'How will tell,' is technology communication technologies that are abbreviated representations of animation director is accessible to the audience and it is a close relationship between the psychological aspect of audience. Because, the audience is reproduced in a limited space, but he called on the board of directors and the same time the screen, the audience located at reception and the director located at provide. It is given. led force is given to the director. for this reason, The director needs to pay attention to the psychological aspect of audience this can be explained based on psychoanalytic theory. In this paper, "How can you lie to the audience and the director is the same line?" put down logic that is the animation audience under the logic that takes place visually capture phenomenon "selective attention" and sub-concept of "goal-directed selection' and 'stimulus-driven capturel' for theory of psychology. also, Induction device to elicit selective attention of the audience accordingly, let's consider whether and how they apply in animation.

Influence of Social Support and Social Network on Quality of Life among the Elderly in a Local Community (지역사회 거주 일반노인의 사회적지지, 사회적관계망이 삶의 질에 미치는 영향)

  • Kim, Hyeong-Min;Sim, Kyoung-Bo;Kim, Hwan;Kim, Souk-Boum
    • The Journal of Korean society of community based occupational therapy
    • /
    • v.3 no.1
    • /
    • pp.11-20
    • /
    • 2013
  • Objective : The purpose of this study is to identify the impact of the social support and social network on the quality of life of the elderly residing in a local community. Method : The subjects of this study were 75 healthy old men and women of 13 sites of welfare centers for the disabled and public health centers and senior welfare centers in Busan and Gyeongju. A survey was conducted with a questionnaire that include general characteristics, cognitive ability, social support, social network and quality of life. The analysis was made on 63 replies except 12 subjects who had been excluded by the subject selection criteria. Result : As a result of analyzing correlation of variables affecting life quality, there was positive correlation in contact frequency(p<.05), intimacy(p<.001), and social support(p<.001). Finally, it was analyzed that the variable of intimacy (p<.001) affected life quality of general aged people living in regional community. Conclusion : It was found that intimacy of general aged people living in regional community was a major variable to affect life quality. It could be identified that intimacy which is qualitative feature of social, relational network for the aged who live passive life was important.

  • PDF

A Study on the Economic Efficiency of Capital Market (자본시장(資本市場)의 경제적(經濟的) 효율성(效率性)에 관한 연구(硏究))

  • Nam, Soo-Hyun
    • The Korean Journal of Financial Management
    • /
    • v.2 no.1
    • /
    • pp.55-75
    • /
    • 1986
  • This article is to analyse the economic efficiency of capital market, which plays a role of resource allocation in terms of financial claims such as stock and bond. It provides various contributions to the welfare theoretical aspects of modern capital market theory. The key feature that distinguishes the theory described here from traditional welfare theory is the presence of uncertainty. Securities has time dimensions and the state and outcome of the future are really uncertain. This problem resulting from this uncertainty can be solved by complete market, but it has a weak power to explain real stock market. Capital Market is faced with the uncertainity because it is a kind of incomplete market. Individuals and firms in capital market made their consumption-investment decision by their own criteria, i. e. the maximization of expected utility form intertemporal consumption and the maximization of the market value of firm. We noted that allocative decisions that had to be made in the economy could be naturally subdivided into two groups. One set of decisions concerned the allocation of first-period resources among consumption $C_i$, investment in risky firms $I_j$, and riskless investment M. The other decisions concern the distribution among individuals of income available in the second period $Y_i(\theta)$. Corresponing to this grouping, the theoretical analysis of efficiency has also been dichotomized. The optimality of the distribution of output in the second period is distributive efficiency" and the optimality of the allocation of first-period resources is 'the efficiency of investment'. We have found in the distributive efficiency that the conditions for attainability is the same as the conditions for market optimality. The necessary and sufficient conditions for attainability or market optimality is that (1) all utility functions are such that -$\frac{{U_i}^'(Y_i)}{{U_i}^"(Y_i)}={\mu}_i+{\lambda}Y_i$-linear risk tolerance function where the coefficients ${\mu}_i$ and $\lambda$ are independent of $Y_i$, and (2) there are homogeneous expectations, i. e. ${\Large f}_i(\theta)={\Large f}(\theta)$ for every i. On the other hand, the efficiency of investment has disagreement about optimal investment level. The investment level for market rule will not generally lead to Pareto-optimal allocation of investment. This suboptimality is caused by (1)the difference of Diamond's decomposable production function and mean-variance valuation model and (2) the selection of exelusive investment or competitive investment. In conclusion, this article has made an analysis of conditions and processes of Pareto-optimal allocation of resources in capital marker and tried to connect with significant issues in modern finance.

  • PDF

Reinforcement Method for Automated Text Classification using Post-processing and Training with Definition Criteria (학습방법개선과 후처리 분석을 이용한 자동문서분류의 성능향상 방법)

  • Choi, Yun-Jeong;Park, Seung-Soo
    • The KIPS Transactions:PartB
    • /
    • v.12B no.7 s.103
    • /
    • pp.811-822
    • /
    • 2005
  • Automated text categorization is to classify free text documents into predefined categories automatically and whose main goals is to reduce considerable manual process required to the task. The researches to improving the text categorization performance(efficiency) in recent years, focused on enhancing existing classification models and algorithms itself, but, whose range had been limited by feature based statistical methodology. In this paper, we propose RTPost system of different style from i.ny traditional method, which takes fault tolerant system approach and data mining strategy. The 2 important parts of RTPost system are reinforcement training and post-processing part. First, the main point of training method deals with the problem of defining category to be classified before selecting training sample documents. And post-processing method deals with the problem of assigning category, not performance of classification algorithms. In experiments, we applied our system to documents getting low classification accuracy which were laid on a decision boundary nearby. Through the experiments, we shows that our system has high accuracy and stability in actual conditions. It wholly did not depend on some variables which are important influence to classification power such as number of training documents, selection problem and performance of classification algorithms. In addition, we can expect self learning effect which decrease the training cost and increase the training power with employing active learning advantage.

A Research of LEACH Protocol improved Mobility and Connectivity on WSN using Feature of AOMDV and Vibration Sensor (AOMDV의 특성과 진동 센서를 적용한 이동성과 연결성이 개선된 WSN용 LEACH 프로토콜 연구)

  • Lee, Yang-Min;Won, Joon-We;Cha, Mi-Yang;Lee, Jae-Kee
    • The KIPS Transactions:PartC
    • /
    • v.18C no.3
    • /
    • pp.167-178
    • /
    • 2011
  • As the growth of ubiquitous services, various types of ad hoc networks have emerged. In particular, wireless sensor networks (WSN) and mobile ad hoc networks (MANET) are widely known ad hoc networks, but there are also other kinds of wireless ad hoc networks in which the characteristics of the aforementioned two network types are mixed together. This paper proposes a variant of the Low Energy Adaptive Cluster Hierarchy (LEACH) routing protocol modified to be suitable in such a combined network environment. That is, the proposed routing protocol provides node detection and route discovery/maintenance in a network with a large number of mobile sensor nodes, while preserving node mobility, network connectivity, and energy efficiency. The proposed routing protocol is implemented with a multi-hop multi-path algorithm, a topology reconfiguration technique using node movement estimation and vibration sensors, and an efficient path selection and data transmission technique for a great many moving nodes. In the experiments, the performance of the proposed protocol is demonstrated by comparing it to the conventional LEACH protocol.

Machine Learning Based Structural Health Monitoring System using Classification and NCA (분류 알고리즘과 NCA를 활용한 기계학습 기반 구조건전성 모니터링 시스템)

  • Shin, Changkyo;Kwon, Hyunseok;Park, Yurim;Kim, Chun-Gon
    • Journal of Advanced Navigation Technology
    • /
    • v.23 no.1
    • /
    • pp.84-89
    • /
    • 2019
  • This is a pilot study of machine learning based structural health monitoring system using flight data of composite aircraft. In this study, the most suitable machine learning algorithm for structural health monitoring was selected and dimensionality reduction method for application on the actual flight data was conducted. For these tasks, impact test on the cantilever beam with added mass, which is the simulation of damage in the aircraft wing structure was conducted and classification model for damage states (damage location and level) was trained. Through vibration test of cantilever beam with fiber bragg grating (FBG) sensor, data of normal and 12 damaged states were acquired, and the most suitable algorithm was selected through comparison between algorithms like tree, discriminant, support vector machine (SVM), kNN, ensemble. Besides, through neighborhood component analysis (NCA) feature selection, dimensionality reduction which is necessary to deal with high dimensional flight data was conducted. As a result, quadratic SVMs performed best with 98.7% for without NCA and 95.9% for with NCA. It is also shown that the application of NCA improved prediction speed, training time, and model memory.

A Study on the Document viewer optimized for VR environment (VR 환경에 최적화 된 문서 뷰어에 관한 연구)

  • Joo, Yong-Ho;Kim, Sang-Mok;Cho, Ok-Hue
    • Journal of the Korea Convergence Society
    • /
    • v.12 no.5
    • /
    • pp.139-145
    • /
    • 2021
  • Through this study, we intend to study user satisfaction in order to verify whether there is a need for full-scale research, development and commercialization of document viewers in a VR environment. VR content consists of realistic 3D graphics and 360-degree video, and provides a synesthesia experience and immersion. We developed and tested a VR document viewer prototype that can utilize this concept as a document viewing system. It can act as a viewer that provides an interactive viewing environment according to the user's body interaction and the direction of the field of view, and it can be said that the feature of VR document viewer is that it can draw the user's high level of immersion and concentration when using the viewer. The developed prototype was tested in a test group consisting of 100 VR experiences and device owners for about 1 hour and 3 days a day, and then a questionnaire survey in the form of a fixed selection question was conducted. This study is a prototype study of a document viewer suitable for a virtual reality environment, and can lead to a sense of immersion when reading a document, and suggest a new document viewer direction that is effective for visual fatigue and visual perception of the document.

Multimodal Sentiment Analysis Using Review Data and Product Information (리뷰 데이터와 제품 정보를 이용한 멀티모달 감성분석)

  • Hwang, Hohyun;Lee, Kyeongchan;Yu, Jinyi;Lee, Younghoon
    • The Journal of Society for e-Business Studies
    • /
    • v.27 no.1
    • /
    • pp.15-28
    • /
    • 2022
  • Due to recent expansion of online market such as clothing, utilizing customer review has become a major marketing measure. User review has been used as a tool of analyzing sentiment of customers. Sentiment analysis can be largely classified with machine learning-based and lexicon-based method. Machine learning-based method is a learning classification model referring review and labels. As research of sentiment analysis has been developed, multi-modal models learned by images and video data in reviews has been studied. Characteristics of words in reviews are differentiated depending on products' and customers' categories. In this paper, sentiment is analyzed via considering review data and metadata of products and users. Gated Recurrent Unit (GRU), Long Short-Term Memory (LSTM), Self Attention-based Multi-head Attention models and Bidirectional Encoder Representation from Transformer (BERT) are used in this study. Same Multi-Layer Perceptron (MLP) model is used upon every products information. This paper suggests a multi-modal sentiment analysis model that simultaneously considers user reviews and product meta-information.

State of the Art Technology Trends and Case Analysis of Leading Research in Harmony Search Algorithm (하모니 탐색 알고리즘의 선도 연구에 관한 최첨단 기술 동향과 사례 분석)

  • Kim, Eun-Sung;Shin, Seung-Soo;Kim, Yong-Hyuk;Yoon, Yourim
    • Journal of the Korea Convergence Society
    • /
    • v.12 no.11
    • /
    • pp.81-90
    • /
    • 2021
  • There are various optimization problems in real world and research continues to solve them. An optimization problem is the problem of finding a combination of parameters that maximizes or minimizes the objective function. Harmony search is a population-based metaheuristic algorithm for solving optimization problems and it is designed to mimic the improvisation of jazz music. Harmony search has been actively applied to optimization problems in various fields such as civil engineering, computer science, energy, medical science, and water quality engineering. Harmony search has a simple working principle and it has the advantage of finding good solutions quickly in constrained optimization problems. Especially there are various application cases showing high accuracy with a low number of iterations by improving the solution through the empirical derivative. In this paper, we explain working principle of Harmony search and classify the leading research in recent 3 years, review them according to category, and suggest future research directions. The research is divided into review by field, algorithmic analysis and theory, and application to real world problems. Application to real world problems is classified according to the purpose of optimization and whether or not they are hybridized with other metaheuristic algorithms.