• Journal of Computing Science and Engineering
• /
• v.10 no.4
• /
• pp.103-110
• /
• 2016

Emotion recognition is an important component of affective computing, and is significant in the implementation of natural and friendly human-computer interaction. An effective approach to recognizing emotion from text is based on a machine learning technique, which deals with emotion recognition as a classification problem. However, in emotion recognition, the texts involved are usually very short, leaving a very large, sparse feature space, which decreases the performance of emotion classification. This paper proposes to resolve the problem of feature sparseness, and largely improve the emotion recognition performance from short texts by doing the following: representing short texts with word cluster features, offering a novel word clustering algorithm, and using a new feature weighting scheme. Emotion classification experiments were performed with different features and weighting schemes on a publicly available dataset. The experimental results suggest that the word cluster features and the proposed weighting scheme can partly resolve problems with feature sparseness and emotion recognition performance.

• Journal of Korean Library and Information Science Society
• /
• v.34 no.2
• /
• pp.273-296
• /
• 2003

Because all the subjects may become the targets of classification, it is necessary to change the enumerated into the faceted scheme. This study is to confirm that possibility of change and to proposed the new faceted structure scheme(faceted classification) based on the Literature. The proposed schemes are : 1) facet structure is simple, 2) meaning of facet is clearness, 3) because using mixed notation, it is complicated.

• Proceeding of Spring/Autumn Annual Conference of KHA
• /
• 2008.11a
• /
• pp.469-474
• /
• 2008

In the last decade many advanced home network services have been provided by construction industry, and many researches on the area have been carried out. One of the problems we are confronted with is that we do not have a standard classification scheme for home network services yet. In this paper a more stable classification scheme is suggested after comparative analysis of many different schemes in previous researches. More commonly highly attended service categories in the scheme is selected, and those are examined in terms of how commonly they are supplied by construction industry. In the later part of this paper the current home network building certification code is reviewed if the code complies with the highly attended categories in the classification scheme suggested in this paper. Some ideas to improve the current certification code is suggested by adding or optionally adding the highly attended categories.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.17 no.9
• /
• pp.1006-1012
• /
• 2004

In this paper, neural networks is studied to apply as a PD source classification in XLPE power cable specimen. Two learning schemes are used to classification; BP(Back propagation algorithm), SOM(self organized map - kohonen network). As a PD source, using treeing discharge sources in the specimen, three defected models are made. And these data making use of a computer-aided discharge analyser, statistical and other discharge parameters is calculated to discrimination between different models of discharge sources. And a]so these distribution characteristics are applied to classify PD sources by two scheme of the neural networks. In conclusion, recognition efficiency of BP is superior to SOM.

• Journal of Korean Institute of Industrial Engineers
• /
• v.24 no.3
• /
• pp.447-455
• /
• 1998

It is important to identify and evaluate poor working postures properly to prevent work-related musculoskeletal disorders. The purpose of this study is to develope a new upper body micropostural classification scheme for analyzing postural stress in industry. Most of the existing postural classification schemes were based either on the literature, or on simple biomechanical principles, or on a subjective ranking system. The scheme suggested in this study was based on perceived joint discomfort measured through experiment, in which nineteen subjects participated and the magnitude estimation method was employed to obtain subjects' joint discomfort. Also, the criteria for evaluating postural stress of working postures were presented for practitioners of health and safety to be able to redesign working methods and workplaces, which was based on maximum holding time by Miedema and other people. It is expected that the scheme developed in this study could be used as a valuable tool when evaluating working postures.

• The Journal of the Korea Contents Association
• /
• v.18 no.10
• /
• pp.443-452
• /
• 2018

With the explosively increasing intellectual property rights, securing technological competitiveness of companies is more and more important. In particular, since patents include core technologies and element technologies, patent analysis researches are actively conducted to measure the technological value of companies. Various patent analysis studies have been conducted by the International Patent Classification(IPC), which does not include the latest technical classification, and the technical classification accuracy is low. In order to overcome this problem, the Cooperative Patent Classification(CPC), which includes the latest technology classification and detailed technical classification, has been developed. In this paper, we propose a model to analyze the classification of the technologies included in the patent by using the detailed classification system of CPC. It is possible to analyze the inventor's patents in consideration of the relation, importance, and efficiency between the detailed classification schemes of the CPCs to extract the core technology fields and to analyze the details more accurately than the existing IPC-based methods. Also, we perform the comparative evaluation with the existing IPC based patent analysis method and confirm that the proposed model shows better performance in analyzing the inventor's core technology classification.

• Journal of Korean Library and Information Science Society
• /
• v.15
• /
• pp.179-205
• /
• 1988

The purpose of this research is to analyze the differences between electronical engineering class in KDC3 and electronical engineering class in DDC 19. The results of the study can be summarized as follows ; 1. The characteristics of the classification items in classification schemes of the electronical engineering fields is that KDC is classified by the vacuum tubes, electron tubes, special tubes, computers etc and DDC is classified by the microwave electronics, short-and long-wave electronics, X-ray and gamma-ray electronics, computers etc. 2. There is a blank in KDC and there are three blanks in DDC of the electronical engineering fields, but the total classification items number is much more in DDC than in KDC. This means that DDC is more classified than in the special classification items. 3. Classification items over 70% in the total classification items of computer fields in KDC, are classified a classification items of the short-and long-wave electronics in DDC. 4. There is a blank of computer fields in KDC and there are three blanks of computer fields in DDC, but the total classification items are many eighteen items. This means that DDC is more classified than KDC. 5. The sections of computers in DDC were established the 621.38195 classification item and the subsection of analogue computers in DDC were established the 621.381957 classification item and the subsection of digital computer in DDC were established the 621.381958 classification item. As we have seen, because of the development in technology and the subdivision of knowledge, the number of items is increased, and the terminology also become simplified and specified in the field of electronic engineering in KDC and DDC.

• The Journal of Korea Robotics Society
• /
• v.15 no.1
• /
• pp.62-69
• /
• 2020

This paper proposes a pattern recognition and classification algorithm based on a circular structure that can reflect the characteristics of the sEMG (surface electromyogram) signal measured in the arm without putting the placement limitation of electrodes. In order to recognize the same pattern at all times despite the electrode locations, the data acquisition of the circular structure is proposed so that all sEMG channels can be connected to one another. For the performance verification of the sEMG pattern recognition and classification using the developed algorithm, several experiments are conducted. First, although there are no differences in the sEMG signals themselves, the similar patterns are much better identified in the case of the circular structure algorithm than that of conventional linear ones. Second, a comparative analysis is shown with the supervised learning schemes such as MLP, CNN, and LSTM. In the results, the classification recognition accuracy of the circular structure is above 98% in all postures. It is much higher than the results obtained when the linear structure is used. The recognition difference between the circular and linear structures was the biggest with about 4% when the MLP network was used.

• Journal of Internet Computing and Services
• /
• v.1 no.1
• /
• pp.63-72
• /
• 2000

Data mining means a process of nontrivial extraction of hidden knowledge or potentially useful information from data in large databases. Data mining algorithm is a multi-disciplinary field of research; machine learning, statistics, and computer science all make a contribution. Different classification schemes can be used to categorize data mining methods based on the kinds of tasks to be implemented and the kinds of application classes to be utilized, and classification has been identified as an important task in the emerging field of data mining. Since classification is the basic element of human's way of thinking, it is a well-studied problem in a wide varietyof application. In this paper, we propose a classifier system based on genetic algorithm with robust property, and the proposed system is evaluated by applying it to nDmC problem related to classification task in data mining.

• Knowledge Management Research
• /
• v.20 no.3
• /
• pp.139-154
• /
• 2019

With the increasing use of smartphone apps, many apps are coming out in various fields. In order to analyze the current status and trends of apps in a specific field, it is necessary to establish a classification scheme. Various schemes considering users' behavior and characteristics of apps have been proposed, but there is a problem in that many apps are released and a fixed classification scheme must be updated according to the passage of time. Although it is necessary to consider many aspects in establishing classification scheme, it is possible to grasp the trend of the app through the proposal of a classification scheme according to the characteristic of the app. This research proposes a method of establishing an app classification scheme through the description of the app written by the app developers. For this purpose, we collected explanations about apps in the tourism field and identified major categories through topic modeling. Using only the apps corresponding to the topic, we construct a network of words contained in the explanatory text and identify subcategories based on the networks of words. Six topics were selected, and Clauset Newman Moore algorithm was applied to each topic to identify subcategories. Four or five subcategories were identified for each topic.