• Journal of Information Processing Systems
• /
• 제14권6호
• /
• pp.1438-1444
• /
• 2018

Recently, with the development of Internet technologies and propagation of smart devices, use of microblogs such as Facebook, Twitter, and Instagram has been rapidly increasing. Many users check for new information on microblogs because the content on their timelines is continually updating. Therefore, clustering algorithms are necessary to arrange the content of microblogs by grouping them for a user who wants to get the newest information. However, microblogs have word limits, and it has there is not enough information to analyze for content clustering. In this paper, we propose a semantic-based K-means clustering algorithm that not only measures the similarity between the data represented as a vector space model, but also measures the semantic similarity between the data by exploiting the TagCluster for clustering. Through the experimental results on the RepLab2013 Twitter dataset, we show the effectiveness of the semantic-based K-means clustering algorithm.

• Asia pacific journal of information systems
• /
• 제19권1호
• /
• pp.1-31
• /
• 2009

This paper empirically examines factors that potentially influence the success of a Web-based semantic search engine. A research model has been proposed that shows the impact of quality-related factors upon the effectiveness of a semantic search engine, based on DeLone and McLean's(2003) information systems success model. An empirical study has been conducted to test hypotheses formulated around the research model, and statistical methods were applied to analyze gathered data and draw conclusions. Implications for academics and practitioners are offered based on the findings of the study. The proposed model includes three quality dimensions of a Web-based semantic search engine-namely, information quality, system quality and service quality. These three dimensions each have measures designed to collectively assess the respective dimension. The model is intended to examine the relationship between measures of these quality dimensions and measures of two dependent constructs, including individuals' net benefit and user satisfaction. Individuals' net benefit was measured by the extent to which the user's information needs were adequately met, whereas user satisfaction was measured by a combination of the perceived satisfaction with search results and the perceived satisfaction with the overall system. A total of 23 hypotheses have been formulated around the model, and a questionnaire survey has been conducted using a functional semantic search website created by KT and Hakia, so as to collect data to validate the model. Copies of a questionnaire form were handed out in person to 160 research associates and employees working in the area of designing and developing semantic search engines. Those who received the form, 148 respondents returned valid responses. The survey form asked respondents to use the given website to answer questions concerning the system. The results of the empirical study have indicated that, of the three quality dimensions, information quality was found to have the strongest association with the effectiveness of a Web-based semantic search engine. This finding is consistent with the observation in the literature that the aspects of the information quality should serve as a basis for evaluating the search outcomes from a semantic search engine. Measures under the information quality dimension that have a positive effect on informational gratification and user satisfaction were found to be recall and currency. Under the system quality dimension, response time and interactivity, were positively related to informational gratification. On the other hand, only one measure under the service quality dimension, reliability was found to have a positive relationship with user satisfaction. The results were based on the seven hypotheses that have been accepted. One may wonder why 15 out of the 23 hypotheses have been rejected and question the theoretical soundness of the model. However, the correlations between independent variables and dependent variables came out to be fairly high. This suggests that the structural equation model yielded results inconsistent with those of coefficient analysis, because the structural equation model intends to examine the relationship among independent variables as well as the relationship between independent variables and dependent variables. The findings offer some useful implications for owners of a semantic search engine, as far as the design and maintenance of the website is concerned. First, the system should be designed to respond to the user's query as fast as possible. Also it should be designed to support the search process by recommending, revising, and choosing a search query, so as to maximize users' interactions with the system. Second, the system should present search results with maximum recall and currency to effectively meet the users' expectations. Third, it should be capable of providing online services in a reliable and trustworthy manner. Finally, effective increase in user satisfaction requires the improvement of quality factors associated with a semantic search engine, which would in turn help increase the informational gratification for users. The proposed model can serve as a useful framework for measuring the success of a Web-based semantic search engine. Applying the search engine success framework to the measurement of search engine effectiveness has the potential to provide an outline of what areas of a semantic search engine needs improvement, in order to better meet information needs of users. Further research will be needed to make this idea a reality.

• Journal of Information Processing Systems
• /
• 제16권2호
• /
• pp.261-276
• /
• 2020

Digital contents services are one of the topics that have been intensively studied in the media industry, where various semantic and ontology techniques are applied. However, query execution for ontology data is still inefficient, lack of sufficient extensible definitions for node relationships, and there is no specific semantic method fit for media data representation. In order to make the machine understand digital contents (DCs) data well, we analyze DCs data, including static data and dynamic data, and use ontology to specify and classify objects and the events of the particular objects. Then the formal representation method is proposed which not only redefines DCs data based on the technology of OWL/RDF, but is also combined with media segmentation methods. At the same time, to speed up the access mechanism of DCs data stored under the persistent database, an ontology-based DCs query solution is proposed, which uses the specified distance vector associated to a surveillance of semantic label (annotation) to detect and track a moving or static object.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제35권5호
• /
• pp.306-312
• /
• 2008

목적 지향성 대화에서 화자의 의도는 화행과 개념열 쌍으로 구성되는 의미 구조로 근사화될 수 있다. 그러므로 지능형 대화 시스템을 구현하기 위해서는 의미 구조를 올바르게 파악하는 것이 매우 중요하다. 본 논문에서는 능동학습(active learning) 방법을 이용하여 효율적으로 의미 구조를 분석하는 모델을 제안한다. 제안 모델은 언어 분석에 따른 부담을 덜기위하여 형태소 자질들과 이전 의미 구조만을 입력 자질로 사용한다. 그리고 정확률 향상을 위하여 자연어 처리 분야에서 높은 성능을 보이고 있는 CRFs(Conditional Random Fields)를 기본 통계 모델로 사용한다. 일정 관리 영역에서 제안 모델을 실험한 결과는 기존 모델들과 비교하여 1/3 정도의 훈련데이타를 사용하고도 비슷한 정확률(화행 92.4%, 개념열 89.8%)을 나타내고 있음을 알 수 있었다.

• 한국컴퓨터정보학회논문지
• /
• 제15권1호
• /
• pp.31-42
• /
• 2010

GPM(Generic Product Model)은 원자력 플랜트의 라이프 사이클 데이터를 통합, 공유하기 위하여 일본의 Hitachi에서 개발한 공통 데이터 모델이다. GPM은 추상 모델 성격의 GPM 핵심 모델과 핵심 모델의 기술을 위한 구현 언어 그리고 구현 언어로 작성된 참조 라이브러리로 구성되어 있다. GPM 핵심 모델은 객체들 사이의 의미가 부여된 관계 정의를 통하여 객체 의미 관계 네트워크 모델을 구성할 수 있는 특성이 있다. 초기의 GPM은 GPM 핵심 모델의 특성을 반영한 신택스의 GPML이라는 구현 언어를 개발하여 제공하였으나 원자력 플랜트 라이프 사이클동안 다양한 목적으로 GPM 데이터 모델에 접근하는 이기종 애플리케이션들과의 상호운용성을 위하여 XML을 기반으로 하는 GPM-XML로 교체되었다. 그러나 현재 GPM-XML을 사용하여 구축한 GPM 데이터 모델이 객체 의미 관계 네트워크 모델로서 활용되기 위한 GPM-XML 기반의 연구가 미비한 상태이다. 따라서 본 논문에서는 객체 의미 관계 네트워크와 유사한 성격의 온톨로지를 기술할 수 있으며 이를 지원하는 기술 표준 및 도구들이 이미 활용 가능한 수준에 있는 OWL을 GPM 핵심 모델을 위한 구현 언어로서 제안한다. OWL은 XML 기반의 RDF/XML 형식으로 기술될 수 있으므로 상호운용성 또한 보장받을 수 있다. 본 논문은 OWL의 세 가지 하위 언어 사양 중 추론기능을 완벽히 제공받을 수 있는 가운데 문법적 제약이 가장 덜 엄격한 OWL DL을 사용한다. 본 논문은 OWL DL을 GPM 핵심 모델의 구현 언어로서 사용하기 위하여 GPM과 OWL 두 모델 사이의 차이점을 도출한 후 이를 해소할 수 있는 방법을 제안하며 이 방법을 적용하여 GPML로 작성된 참조 라이브러리를 OWL DL 기반의 온톨로지로 변환하여 구축하는 방법을 기술한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권12호
• /
• pp.4706-4724
• /
• 2020

With various structured data, such as the company size, loan balance, and savings accounts, the voice of customer (VOC), which is text data containing contact history and counseling details was analyzed in this study. To analyze unstructured data, the term frequency-inverse document frequency (TF-IDF) analysis, semantic network analysis, sentiment analysis, and a convolutional neural network (CNN) were implemented. A performance comparison of the models revealed that the predictive model using the CNN provided the best performance with regard to predictive power, followed by the model using the TF-IDF, and then the model using semantic network analysis. In particular, a character-level CNN and a word-level CNN were developed separately, and the character-level CNN exhibited better performance, according to an analysis for the Korean language. Moreover, a systematic selection model for optimal text mining techniques was proposed, suggesting which analytical technique is appropriate for analyzing text data depending on the context. This study also provides evidence that the results of previous studies, indicating that individual customers leave when their loyalty and switching cost are low, are also applicable to corporate customers and suggests that VOC data indicating customers' needs are very effective for predicting their behavior.

• 정보과학회 논문지
• /
• 제42권1호
• /
• pp.54-67
• /
• 2015

최근 사물인터넷 환경에서는 발생하는 센서데이터의 가치와 데이터의 상호운용성을 증진시키기 위해 시맨틱웹 기술과의 접목에 대한 연구가 활발히 진행되고 있다. 이를 위해서는 센서데이터와 서비스 도메인 지식의 융합을 위한 센서데이터의 시맨틱화는 필수적이다. 하지만 기존의 시맨틱 변환기술은 정적인 메타데이터를 시맨틱 데이터(RDF)로 변환하는 기술이며, 이는 사물인터넷 환경의 실시간성, 대용량성의 특징을 제대로 처리할 수 없는 실정이다. 따라서 본 논문에서는 사물인터넷 환경에서 발생하는 대용량 스트리밍 센서데이터의 실시간 병렬처리를 통해 시맨틱 데이터로 변환하는 기법을 제시한다. 본 기법에서는 시맨틱 변환을 위한 변환규칙을 정의하고, 정의된 변환규칙과 온톨로지 기반 센서 모델을 통해 실시간 병렬로 센서데이터를 시맨틱 변환하여 시맨틱 레파지토리에 저장한다. 성능향상을 위해 빅데이터 실시간 분석 프레임워크인 아파치 스톰을 이용하여, 각 변환작업을 병렬로 처리한다. 이를 위한 시스템을 구현하고, 대용량 스트리밍 센서데이터인 기상청 AWS 관측데이터를 이용하여 제시된 기법에 대한 성능평가를 진행하여, 본 논문에서 제시된 기법을 입증한다.

• 말소리와 음성과학
• /
• 제7권4호
• /
• pp.41-47
• /
• 2015

This paper examines whether the Global Vector model is applicable to Korean data as a universal learning algorithm. The main purpose of this study is to compare the global vector model (GloVe) with the word2vec models such as a continuous bag-of-words (CBOW) model and a skip-gram (SG) model. For this purpose, we conducted an experiment by employing an evaluation corpus consisting of 70 target words and 819 pairs of Korean words for word similarities and analogies, respectively. Results of the word similarity task indicated that the Pearson correlation coefficients of 0.3133 as compared with the human judgement in GloVe, 0.2637 in CBOW and 0.2177 in SG. The word analogy task showed that the overall accuracy rate of 67% in semantic and syntactic relations was obtained in GloVe, 66% in CBOW and 57% in SG.

• Journal of information and communication convergence engineering
• /
• 제11권3호
• /
• pp.179-184
• /
• 2013

Traditional clustering methods are usually based on the bag-of-words (BOW) model. A disadvantage of the BOW model is that it ignores the semantic relationship among terms in the data set. To resolve this problem, ontology or matrix factorization approaches are usually used. However, a major problem of the ontology approach is that it is usually difficult to find a comprehensive ontology that can cover all the concepts mentioned in a collection. This paper proposes a new document clustering method using semantic features and fuzzy relations for solving the problems of ontology and matrix factorization approaches. The proposed method can improve the quality of document clustering because the clustered documents use fuzzy relation values between semantic features and terms to distinguish clearly among dissimilar documents in clusters. The selected cluster label terms can represent the inherent structure of a document set better by using semantic features based on non-negative matrix factorization, which is used in document clustering. The experimental results demonstrate that the proposed method achieves better performance than other document clustering methods.

• 정보과학회 논문지
• /
• 제43권1호
• /
• pp.106-118
• /
• 2016

최근 사용자들 사이의 웹 데이터의 생성 및 공유가 활발해 지면서 시맨틱 웹과 소셜 웹이 결합한 소셜 시맨틱 웹에 대한 중요성이 증가되고 있다. 본 논문은 소셜 시맨틱 웹 환경에서 PROV 모델을 확장하여 프로버넌스 기반의 웹 데이터 신뢰성 평가 기법을 제안한다. 제안하는 기법은 소셜 시맨틱 웹 환경에서 웹 데이터의 프로버넌스를 관리하고 신뢰성 평가를 위해 W3C의 PROV 모델에 필요한 요소를 추가하였다. 이와 같이 확장된 PROV 모델은 웹 데이터의 관리 및 프로버넌스 추적을 지원한다. 제안하는 신뢰성 평가 기법은 사용자의 신뢰도, 원본 데이터의 신뢰도 그리고 사람들의 평가 등과 같은 다양한 파라미터를 고려한다. 평가된 신뢰도는 프로버넌스 정보로 관리되고 사용자의 질의를 처리할 때 이 신뢰도를 고려하여 결과를 생성한다. 따라서 제안하는 기법은 소셜 시맨틱 웹 데이터의 프로버넌스를 관리할 수 있고 다양한 파라미터를 이용하여 웹 데이터 신뢰도를 정확하게 계산할 수 있다. 평가된 신뢰도는 사용자가 질의 결과의 신뢰 여부를 판단할 수 있는 기준이 된다. 제안하는 기법의 타당성을 보이기 위해 SPARQL 질의를 이용하여 신뢰성 평가의 성능을 검증한다.