• 전기학회논문지
• /
• 제58권3호
• /
• pp.632-638
• /
• 2009

Probabilistic Latent Semantic Analysis has many applications in information retrieval and filtering, natural language processing, machine learning from text, and in related areas. In this paper, we propose an algorithm using weighted Probabilistic Latent Semantic Analysis Model to find the contextual phrases and opinions from documents. The traditional keyword search is unable to find the semantic relations of phrases, Overcoming these obstacles requires the development of techniques for automatically classifying semantic relations of phrases. Through experiments, we show that the proposed algorithm works well to discover semantic relations of phrases and presents the semantic relations of phrases to the vector-space model. The proposed algorithm is able to perform a variety of analyses, including such as document classification, online reputation, and collaborative recommendation.

• 디지털산업정보학회논문지
• /
• 제14권4호
• /
• pp.35-45
• /
• 2018

Users are sharing many of contents such as text, image, video, and so on in SNS. There are various information as like as personal interesting, opinion, and relationship in social media contents. Therefore, many of recommendation systems or search systems are being developed through analysis of social media contents. In order to extract subject-related topics of social context being collected from social media channels in developing those system, it is necessary to develop ontologies for semantic analysis. However, it is difficult to develop formal ontology because social media contents have the characteristics of non-formal data. Therefore, we develop a social topic system based on semantic and subject correlation. First of all, an extracting system of social topic based on semantic relationship analyzes semantic correlation and then extracts topics expressing semantic information of corresponding social context. Because the possibility of developing formal ontology expressing fully semantic information of various areas is limited, we develop a self-extensible architecture of ontology for semantic correlation. And then, a classifier of social contents and feed back classifies equivalent subject's social contents and feedbacks for extracting social topics according semantic correlation. The result of analyzing social contents and feedbacks extracts subject keyword, and index by measuring the degree of association based on social topic's semantic correlation. Deep Learning is applied into the process of indexing for improving accuracy and performance of mapping analysis of subject's extracting and semantic correlation. We expect that proposed system provides customized contents for users as well as optimized searching results because of analyzing semantic and subject correlation.

• 한국멀티미디어학회논문지
• /
• 제24권1호
• /
• pp.106-116
• /
• 2021

To increase the quality of the system in the subjective-type question grading and document classification, we need the paragraph detection. But it is not easy because it is accompanied by semantic analysis. Many researches on the paragraph detection solve the detection problem using the word based clustering method. However, the word based method can not use the order and dependency relation between words. This paper suggests the paragraph detection system using syntactic-semantic relation between words with the Korean syntactic-semantic analysis. This system is the hybrid system of word based, concept based, and syntactic-semantic tree based detection. The experiment result of the system shows it has the better result than the word based system. This system will be utilized in Korean subjective question grading and document classification.

• 한국콘텐츠학회논문지
• /
• 제6권3호
• /
• pp.12-19
• /
• 2006

본 논문에서는 질의 응답 시스템의 성능 향상을 위한 질의문 심층 분석을 제안한다. 일반적인 질의응답 시스템들은 사용자의 자연언어 질의의 의미를 분석하지 않기 때문에 정확한 정답을 제공하는 것이 어렵다. 질의문 심층 분석은 의미자질 추출 문법과 자연언어 질의 특성을 이용하여 사용자의 질의를 의미적으로 분석하고, 의미자질들을 추출한다. 의미자질 추출 문법과 자연언어 질의 특성은 사용자 질의의 의미와 구문 구조를 반영하기 위해 의미자질과 형식형태소로 표현된다. 웹에서 추출한 세부 정답 유형이 '인물'인 100개의 질의에 대한 실험을 통해, 비교적 짧지만 사용자의 질의 의도를 충분히 표현하고 있는 자연언어 질의에 대해 질의문 심층 분석을 수행함으로써 사용자의 질의 의도를 분석하고, 의미자질들을 추출할 수 있음을 보였다.

• 한국문헌정보학회지
• /
• 제47권3호
• /
• pp.343-361
• /
• 2013

본 연구는 스피치 요약을 위해서 태그를 확장하고 또한 태그 간의 의미적 관계 정보를 이용할 수 있는 태그의미분석 방법을 제안하고 평가하였다. 이를 위해서, 먼저 비디오 태그를 확장하고 태그 간의 의미적 관계를 분석하는데 있어서 플리커의 태그 클러스터와 워드넷의 동의어 정보가 얼마나 효과적으로 이용될 수 있는가 조사해 보았다. 그런 다음 태그의미분석 방법의 특성과 효율성을 조사해 보기 위해서 제안한 방법을 잠재의미분석(Latent Semantic Analysis) 방법과 비교해 보았다. 분석 결과, 플리커의 태그 클러스터는 효과적으로 이용되었지만 워드넷은 효과적으로 이용되지 못한 것으로 나타났다. F측정을 사용하여 두 방법의 효율성을 비교한 결과, 제안한 방법의 F값(0.27)이 잠재의미분석 방법의 F값(0.22)보다 높게 나타났다.

• 한국정보처리학회논문지
• /
• 제3권7호
• /
• pp.1792-1802
• /
• 1996

영한 기계번역에서 전치사구의 해석 부착의 문제(Attachment Problem)와 의미 해석의 문제, 그리고 해석에 필요한 정보 획득의 문제가 있다. 이 세 가지 문제를 해결하기 위하여 본 논문은 전치사구 해석 시스템을 제시한다. 이 시스템은 규칙 제어기와 신경망의 하이브리드 구문해석 시스템, 격의미 해석 시스템, 그리고 신경망 의 입력 정보를 자동으로 생성하는 의미속성 생성기로 구성한다. 의미속성 생성기는 시스템의 입력이 되는 의미속성을 자동으로 생성하는 방법으로 인위적인 방법의 단점 을보완하여 객관성 있는 전치사구 해석을 하게 한다. 격의미 해석 시스템은 영한 기계 번역에 맞는 격의미를 찾아내어 자연스런 한국어 생성을 하게 하고 구문해석 시스템은 규칙 방법의 장점과 신경망 방법의 장점을 취한 하이브리드 방식의 시스템으로 전치사 구 부착의 문제를 해결한다.

• 한국프랜차이즈경영연구
• /
• 제14권3호
• /
• pp.17-35
• /
• 2023

Purpose: As more and more food franchise companies want to expand overseas, related research is becoming more and more necessary. This study aims to examine the critical factors for successful overseas expansion according to the stages of overseas expansion, derive vital associations, and examine the success factors of overseas expansion through semantic network analysis. Research Design, Data, and Methodology: This study conducted in-depth interviews with three food franchise companies that have experienced overseas expansion and conducted semantic network analysis among crucial associations. The semantic network analysis was conducted using the Textom program. Results: Based on the results of the in-depth interview analysis, the factors considered when expanding overseas were categorized as 1) standardization and localization strategies of overseas franchisees, 2) physical environment of overseas franchisees, 3) entry types of overseas franchisees, 4) constraints of overseas franchisees, and 5) success criteria of overseas franchisees. The semantic network analysis based on the corresponding keywords showed that the importance of local partners is very high in common. Conclusion: This study examined and re-categorized the important factors to consider when a restaurant franchise company expands overseas in a step-by-step manner. In addition, an attempt was made to examine the keywords derived from the semantic network analysis objectively. The results provided theoretical and practical implications for the successful overseas expansion of franchise companies.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2018년도 추계학술대회
• /
• pp.197-200
• /
• 2018

본 연구에서는 <네이버 영화> 페이지의 리뷰 데이터를 수집하여, 출현 빈도가 높은 단어를 중심으로 영화 관람객의 반응을 시각화하는 작업을 수행하였다. 이를 위해 총 6편의 영화를 선정하여 데이터 수집 및 정제과정을 거쳤으며, 의미연결망 분석(Semantic network analysis)을 활용하여 단어 간 관계성을 파악하고자 하였다. 데이터 시각화 작업에는 UCINET과 함께 패키지화된 NetDraw가 사용되었다. 본 연구의 시사점은 문장으로 작성된 영화 관람객의 리뷰를 키워드 중심으로 시각화하여, 소비자들의 반응을 한 눈에 확인하는 리뷰 인터페이스 구현이 가능한지 탐색하였다는 점이다.

• 대한인간공학회지
• /
• 제34권6호
• /
• pp.563-581
• /
• 2015

Objective: This study was conducted to investigate the potential of sematic network analysis for design research. Background: As HCD (Human-Centered Design) was emphasized, lots of design research methodologies were developed and used in order to find user needs. However, it is still difficult to discover users' latent needs. This study suggests the semantic network analysis as a complementary means for design research, and proved its potential through the practical application, which compares multi-screen purchase and usage behaviors between America and China. Method: We conducted an in-depth interview with 32 consumers from USA and China, and analyzed interview texts through semantic network analysis. Cross cultural differences in purchase and usage behaviors were investigated, based on measuring centrality and community modularity of devices, functions, key buying factors and brands. Results: Americans use more services and functions in the multi-screen environment, compared to Chinese. As a device substitutes other devices, traditional boundaries of the devices are disappearing in the USA. Americans consider function to recall Apple, but Chinese consider function, design and brand to recall Apple, Sony and Samsung as an important brand at the time of their purchase. Conclusion: This study shows the potential of semantic network analysis for design research through the practical application. Semantic network analysis presents how the concepts regarding a theme are structured in the cognitive map of users with visual images and quantitative data. Therefore, it can complement the qualitative analysis of the existing design research. Application: As the design environment becomes more and more complicated like multi-screen environment, semantic network analysis, which is able to provide design insights in the intuitive and holistic perspective, will be acknowledged as an effective tool for further design research.

• 대한건축학회논문집:계획계
• /
• 제35권7호
• /
• pp.101-108
• /
• 2019

In this research, I conducted a semantic network analysis centering on media articles on purchasing, revitalizing, and utilizing the former Busan branch of Hansung Bank, a modern architectural heritage. We sought the most efficient analysis elements for the analysis of the social arguments about preservation and utilization embedded in media articles. For this reason, Degree Centrality measures how many connections the word described in the media article has, and Betweenness Centrality measures the influence that controls the flow of information through correlation I examined. In addition, keyword that express the theme well examined the aggregation structure in each sub-network. In this research, in theoretical terms, it makes sense in that the social discussion embedded in the article of the mass media is grasped empirically through semantic network analysis of words. Methodological aspect is best when it includes nouns and adjectives and the distance between words is more than four words in the analysis of the cohesive structure of the semantic network to determine whether the influence of social discussions is best assessed through the connection between words to media articles.