• Title/Summary/Keyword: 시간적 연관성

Search Result 918, Processing Time 0.024 seconds

An Enhancement of Learning Speed of the Error - Backpropagation Algorithm (오류 역전도 알고리즘의 학습속도 향상기법)

  • Shim, Bum-Sik;Jung, Eui-Yong;Yoon, Chung-Hwa;Kang, Kyung-Sik
    • The Transactions of the Korea Information Processing Society
    • /
    • v.4 no.7
    • /
    • pp.1759-1769
    • /
    • 1997
  • The Error BackPropagation (EBP) algorithm for multi-layered neural networks is widely used in various areas such as associative memory, speech recognition, pattern recognition and robotics, etc. Nevertheless, many researchers have continuously published papers about improvements over the original EBP algorithm. The main reason for this research activity is that EBP is exceeding slow when the number of neurons and the size of training set is large. In this study, we developed new learning speed acceleration methods using variable learning rate, variable momentum rate and variable slope for the sigmoid function. During the learning process, these parameters should be adjusted continuously according to the total error of network, and it has been shown that these methods significantly reduced learning time over the original EBP. In order to show the efficiency of the proposed methods, first we have used binary data which are made by random number generator and showed the vast improvements in terms of epoch. Also, we have applied our methods to the binary-valued Monk's data, 4, 5, 6, 7-bit parity checker and real-valued Iris data which are famous benchmark training sets for machine learning.

  • PDF

Considering Customer Buying Sequences to Enhance the Quality of Collaborative Filtering (구매순서를 고려한 개선된 협업필터링 방법론)

  • Cho, Yeong-Bin;Cho, Yoon-Ho
    • Journal of Intelligence and Information Systems
    • /
    • v.13 no.2
    • /
    • pp.69-80
    • /
    • 2007
  • The preferences of customers change over time. However, existing collaborative filtering (CF) systems are static, since they only incorporate information regarding whether a customer buys a product during a certain period and do not make use of the purchase sequences of customers. Therefore, the quality of the recommendations of the typical CF could be improved through the use of information on such sequences. In this study, we propose a new methodology for enhancing the quality of CF recommendation that uses customer purchase sequences. The proposed methodology is applied to a large department store in Korea and compared to existing CF techniques. Various experiments using real-world data demonstrate that the proposed methodology provides higher quality recommendations than do typical CF techniques with better performance.

  • PDF

The Literature Study about the Origin of Sipgan-Sipiji (십간십이지의 기원에 관한 문헌적 고찰)

  • Na, Hyeok-Jin;Kim, Ki-Seung
    • Industry Promotion Research
    • /
    • v.5 no.1
    • /
    • pp.117-124
    • /
    • 2020
  • The purpose of this paper is to find out the origin and significance of Sipgan-Sipiji that Yinyang and five elements are assigned. By looking at the hypotheses that Sipgan-Sipiji may have originated from the ancient Chinese calendar, the contents of the materials recorded in the classical literature, and the records left in the Oracle bone script, we wanted to explore the starting point of Sipgan-Sipiji in Eastern humanities and social sciences that could never be overlooked. Through the literature review of Sipgan-Sipiji, the researcher believes that this code system was completed by Hwangje and his teacher Daeyo, who led paternal clan society in the latter part of the Chinese Neolithic Age, because of supporting the use of the October calendar associated with Sipgan, and because Sipgan in the Oracle bone script seems to keep the sacredness and at the same time to be used to document the time. I believe that this consideration of Sipgan-Sipiji will be helpful for future related research in the process of underlying the research effort of Myeongri Science, such as the security of the logic of the theoretical framework for the theory system and the proof through statistical social science techniques.

The Ontology Based, the Movie Contents Recommendation Scheme, Using Relations of Movie Metadata (온톨로지 기반 영화 메타데이터간 연관성을 활용한 영화 추천 기법)

  • Kim, Jaeyoung;Lee, Seok-Won
    • Journal of Intelligence and Information Systems
    • /
    • v.19 no.3
    • /
    • pp.25-44
    • /
    • 2013
  • Accessing movie contents has become easier and increased with the advent of smart TV, IPTV and web services that are able to be used to search and watch movies. In this situation, there are increasing search for preference movie contents of users. However, since the amount of provided movie contents is too large, the user needs more effort and time for searching the movie contents. Hence, there are a lot of researches for recommendations of personalized item through analysis and clustering of the user preferences and user profiles. In this study, we propose recommendation system which uses ontology based knowledge base. Our ontology can represent not only relations between metadata of movies but also relations between metadata and profile of user. The relation of each metadata can show similarity between movies. In order to build, the knowledge base our ontology model is considered two aspects which are the movie metadata model and the user model. On the part of build the movie metadata model based on ontology, we decide main metadata that are genre, actor/actress, keywords and synopsis. Those affect that users choose the interested movie. And there are demographic information of user and relation between user and movie metadata in user model. In our model, movie ontology model consists of seven concepts (Movie, Genre, Keywords, Synopsis Keywords, Character, and Person), eight attributes (title, rating, limit, description, character name, character description, person job, person name) and ten relations between concepts. For our knowledge base, we input individual data of 14,374 movies for each concept in contents ontology model. This movie metadata knowledge base is used to search the movie that is related to interesting metadata of user. And it can search the similar movie through relations between concepts. We also propose the architecture for movie recommendation. The proposed architecture consists of four components. The first component search candidate movies based the demographic information of the user. In this component, we decide the group of users according to demographic information to recommend the movie for each group and define the rule to decide the group of users. We generate the query that be used to search the candidate movie for recommendation in this component. The second component search candidate movies based user preference. When users choose the movie, users consider metadata such as genre, actor/actress, synopsis, keywords. Users input their preference and then in this component, system search the movie based on users preferences. The proposed system can search the similar movie through relation between concepts, unlike existing movie recommendation systems. Each metadata of recommended candidate movies have weight that will be used for deciding recommendation order. The third component the merges results of first component and second component. In this step, we calculate the weight of movies using the weight value of metadata for each movie. Then we sort movies order by the weight value. The fourth component analyzes result of third component, and then it decides level of the contribution of metadata. And we apply contribution weight to metadata. Finally, we use the result of this step as recommendation for users. We test the usability of the proposed scheme by using web application. We implement that web application for experimental process by using JSP, Java Script and prot$\acute{e}$g$\acute{e}$ API. In our experiment, we collect results of 20 men and woman, ranging in age from 20 to 29. And we use 7,418 movies with rating that is not fewer than 7.0. In order to experiment, we provide Top-5, Top-10 and Top-20 recommended movies to user, and then users choose interested movies. The result of experiment is that average number of to choose interested movie are 2.1 in Top-5, 3.35 in Top-10, 6.35 in Top-20. It is better than results that are yielded by for each metadata.

Toxicity of Chloramine-T on the Flounder, Paralichthys olivaceus, and Bactertcidal Activity Against Fish Pathogenic Bacteria and Blue Green Algae, Tetraselmis suecica (Chloramine-T의 넙치에 대한 독성과 어병세균 및 Blue Green Algae에 대한 살균 효과)

  • Hwang, Eun-A;Jeong, Hyun-Do
    • Journal of fish pathology
    • /
    • v.13 no.1
    • /
    • pp.37-44
    • /
    • 2000
  • CT(Sodium N-chloro-para toluenesulfonamide, Chloramine-T) known to be a strong oxidative agent was investigated to use as a candidate of disinfectant in the marine aquaculture industry by the analysis of the bactericidal activity against different aquatic microorganisms. One hundred percent mortality appeared at and above 16 ppm CT in sea water in flounder, the predominant species in the marine aquaculture of Korea, when exposed for 48hr. However, bactericidal activity was appeared to be very effective, and all different species of the fish pathogenic bacteria exposed to CT of less than 2 ppm in sea water were dead within 15 min. It allowed us to confirm that CT could be a very effective disinfectant to protect the spread of fish pathogenic bacteria derived from diseased fish or sea water in marine aquaculture. High concentration(10 ppm) with longer exposing time(24 hrs) was required for the algicidal activity of CT, at which concentration might induce the acute toxicity against fish, however, restricted the expanded use of CT for the elimination of zooplanktons at phytoplanktons in marine farms. Moreover the bactericidal activity of CT inhibited almost completely in the present of more than 10 ppm of organic materials pointed out that quality of the culturing sea water should be considered carefully for the application CT to the aquatic farms as a therapeutic agent.

  • PDF

A Study of Science Teachers' Perception on Knowledge Information Processing Competency (지식정보처리역량에 대한 과학교사의 인식 조사)

  • Son, Mihyun;Jeong, Daehong
    • Journal of The Korean Association For Science Education
    • /
    • v.38 no.5
    • /
    • pp.693-703
    • /
    • 2018
  • One of the important competencies in knowledge and information society is the competency of Knowledge information processing. This is closely related to science education and described as one of the core competencies in the 2015 national curriculum revision. Even with a general agreement of the educational necessity of this competency, its practice in class depends mostly on teachers' will. Therefore, we surveyed science teachers' perceptions and classroom condition about the competency of Knowledge information processing, and we analyzed the questionnaires of 64 middle and high school teachers in Seoul and interviewed three teachers. As a result, all teachers shared the importance of Knowledge information processing competency and explained it in terms of social paradigm and goal of the subject. However, there were not many cases of actual practice in class. The teachers answered that time for Knowledge information processing competency is not enough in class and they also lack of PCK on this competency. About half of the teachers had experiences in teacher training about Knowledge information processing competency but most of their experiences are related to information utilization literacy. The importance of Knowledge information processing competency was very high in three factors (collection, analyzation, and utilization), among which information analysis was the highest. Middle school teachers showed higher scores in most questions even though the differences were not significant, and high school teachers showed higher scores in the importance of information sources at a significant level. In order to cultivate competencies, it is necessary to expand the meaning of science inquiry so as to cover this competency and to consider ways of linking with other subjects and develop methods of teaching and learning.

Characteristics of Low-frequency Ambient Seismic Noise in South Korea (국내 저주파수 무작위 지진잡음의 특성 연구)

  • Park, Iseul;Kim, Ki Young;Byu, Joongmoo
    • Geophysics and Geophysical Exploration
    • /
    • v.19 no.2
    • /
    • pp.67-75
    • /
    • 2016
  • To investigate spatial and temporal variations of low-frequency (${\leq}5Hz$) ambient seismic noise, we analyzed the noise data recorded for one whole year of 2014 at surface accelerometer stations in South Korea. After decomposed into low-frequency (LF; < 1 Hz) and high-frequency (HF; ${\geq}1Hz$) components, the root-mean-squared (RMS) amplitudes and power spectral densities (PSD) of the noise data were computed. The RMS amplitudes were larger on islands and near-shore stations, but also large RMS amplitudes were observed at inland stations in large cities only for HF components. The RMS amplitudes of HF components were larger in the daytime than at nighttime and during weekdays than on Sunday and holidays. This indicates the HF components are closely related to human activities. On the contrary, daily and weekly variations were not clear in the LF components while they showed seasonal variations with its maximum during the winter and a good correlation with significant wave height. Therefore, we interpret the mechanism of LF components is closely related to natural phenomena such as sea. The amplitude of LF components decreased as an exponential function of the distance to the center of typhoons. The exponential index of -0.76 suggested that ambient seismic noise included both surface and body waves. Peak frequencies of the PSD curves were near 0.34 Hz indicating the double frequency. No temporal variation in the peak frequency was clearly noticed.

Channel Searching Method of IEEE 802.15.4 Nodes for Avoiding WiFi Traffic Interference (WiFi 트래픽 간섭을 피하기 위한 IEEE 802.15.4 노드의 채널탐색방법)

  • Song, Myong Lyol
    • Journal of Internet Computing and Services
    • /
    • v.15 no.2
    • /
    • pp.19-31
    • /
    • 2014
  • In this paper, a parallel backoff delay procedure on multiple IEEE 802.15.4 channels and a channel searching method considering the frequency spectrum of WiFi traffic are studied for IEEE 802.15.4 nodes to avoid the interference from WiFi traffic. In order to search the channels being occupied by WiFi traffic, we analyzed the methods measuring the powers of adjacent channels simultaneously, checking the duration of measured power levels greater than a threshold, and finding the same periodicity of sampled RSSI data as the beacon frame by signal processing. In an wireless channel overlapped with IEEE 802.11 network, the operation of CSMA-CA algorithm for IEEE 802.15.4 nodes is explained. A method to execute a parallel backoff procedure on multiples IEEE 802.15.4 channels by an IEEE 802.15.4 device is proposed with the description of its algorithm. When we analyze the data measured by the experimental system implemented with the proposed method, it is observed that medium access delay times increase at the same time in the associated IEEE 802.15.4 channels that are adjacent each other during the generation of WiFi traffic. A channel evaluation function to decide the interference from other traffic on an IEEE 802.15.4 channel is defined. A channel searching method considering the channel evaluations on the adjacent channels together is proposed in order to search the IEEE 802.15.4 channels interfered by WiFi, and the experimental results show that it correctly finds the channels interfered by WiFi traffic.

Petrochmical study on the Volcanic Rocks Related to Depth to the Benioff Zone and Crustal Thickness in the Kyongsang Basin, Korea: A Review (경상분지 화산암류의 지화학적 연구. 섭입대(베니오프대)의 깊이와 지각의 두께)

  • Jong Gyu Sung
    • Economic and Environmental Geology
    • /
    • v.32 no.4
    • /
    • pp.323-337
    • /
    • 1999
  • Late Cretaceous to early Tertiary volcanic rocks in the Kyongsang basin exhibit high-K calc-alkaline characteristics, and originated from the magmatism related genetically to subduction of Kula-Pacific plate. They represent HFSE depletion and LlLE enrichment characteristics as shown by magmas related to subduction. Early studies on the depth of magma generation has been estimated as 180-230 km based on K-h relation should be reevaluated, because the depth of peridotite partial melting with 0.4 wt. % water is 80-120 km at subduction zone, and subducting slab in premature arc can melted even lower than 70 km. Moreover the increase of potassium contents depends on either contamination of crustal material and fluids of subducting slab or low degree of partial melting. If the inclination of subduction zone is 30 degrees and the depth to the Benioff zone is 180-230 km, the calculated distance between the volcanic zone and trench axis would be 310-400 km. It is unlikely because the distance between the Kyongsang basin and trench during late Cretaceous to early Tertiary is closer than this value and not comparable with generally-accepted models in subduction zone magmatism. $K_{55}$ of the volcanics in the Kyongsang basin is 0.3-2.3 wt.% and the average indicate that the depth ranges between 80-170 km on the diagram of Marsh, Carmichael (1974). Fractionation from garnet lherzolite, assumed the depth of 180-230km, is not consistent with the REE patterns of the volcanoes in the Kyongsang basin. Futhermore, the range of depth suggested by many workers, who studied magmatism related to subduction, imply shallower than this depth. Crustal thickness calculated by the content of CaO and $Na_2O$ is about 30 km and about 35 km, respectively. Paleo-crustal thickness during late Cretaceous to early Tertiary times in the Kyongsang basin inferred about 30 km calculated by La/Sm versus LaJYb data, which is also supported by many previous studies.

  • PDF

Historical Studies on the Characteristics of Jondeokjeong in the Rear Garden of Changdeok Palace (창덕궁 후원 존덕정(尊德亭)의 조영사적 특성)

  • Song, Suk-Ho;Sim, Woo-Kyung
    • Journal of the Korean Institute of Traditional Landscape Architecture
    • /
    • v.32 no.1
    • /
    • pp.31-43
    • /
    • 2014
  • The purposes for this research were to find the correct building background of Jondeokjeong(尊德亭) in the rear garden of Changdeok Palace(昌德宮) based on the historical facts for the area of Jondeokjeong and investigate the internal theory of construction through the metaphysical consideration. The results were as follows. 1. Building background of Jondeokjeong was related to Sohyunseja(昭顯世子) and Bongrimdaegun(鳳林大君) as forms of rear garden where was not built in the area of Jondeokjeong, bamboo pavilion, hexagonal pavilion, octagonal pavilion, etc. They were built two or three times after returning of Sohyunseja and Bongrimdaegun from China, and the area of Jondeokjeong was continuously developed by building Chunhyanggak(天香閣), Mangchunjeong(望春亭) and Cheoknoidang(滌惱堂) after Hyojong(孝宗) succeeded the royal authority of Hyunjong(顯宗) who was born in Shenyang(瀋陽) and hung a signboard of Jondeokjeong after then, etc. 2. Dazhengjeon(大政殿) of Shenyang Palace(瀋陽故宮, 1625) played the roles of major hall, Jeongjeon(正殿) which held national big events as a one-storied building with two piles of roof similar with Jondeokjeong (1644). Also, it was the building encountered when Sohyunseja and Bongrimdaegun participated in breakfasts held in the palace or banquets supervised by a king, so building background of Jondeokjeong was judged to be related with Dazhengjeon. 3. In consideration of characteristics which are shown commonly in Jondeokjeong and Dazhengjeon, relations of two buildings were verified and characteristics of Jondeokjeong were examined. First, dragon which is representatively symbolizing royal authority was formed. Therefore, Jondeokjeong was judged as a garden building which purposes were to obtain and train natural reasons, govern the nation and let people comfortable. Second, the purposes of sun dial, Ilyoungdae(日影臺) were judged to examine the accurate time and express appropriateness of the king. Third, Taechungmun(太淸門) around there is related to Samcheong(三淸) of Taoism and judged to be caused by floral wall and secular happiness accomplishment for king's longevity without disease. Fourth, for building style, one-stories pavilion of duplicated roof and building with only cylinder were to king's embodiment of supporting the sky and governing the nation.4) 4. By examining the differences between Jondeokjeong and Dazhengjeon, Jondeokjeong's own characteristics were considered. First, constructive characteristics of Jondeokjeong were changed by existing of pond and it was changed for characteristics, location and function of pavilion built in the garden while accepting oversea culture and embodied by absorbing to Joseon culture. Second, the appearance of the pond in Jondeokjeong was shown as the situation that half moon typed pond with the form of young moon is full by stream to the East. It was to express movement of moon which was always changed through the form of pond and stream of water iconographycally and it was considered as an imaginary environment method of the period.