• 산경연구논집
• /
• 제5권1호
• /
• pp.27-37
• /
• 2014

Purpose - This study analyzed the characteristics of social networking sites (SNSs) using related literatures, and researched the models discussed in precedent studies, to investigate the effects of SNS characteristics upon consumers'awareness, purchase intention, and recommendation. The purpose of the study was to investigate the use of SNSs as a marketing tool. Design, methodology, and approach - For an empirical analysis, the author distributed questionnaires online and offline, to verify the models and hypotheses. Respondents were persons aged 17 or older, who were frequent users of SNSs. The questionnaire survey was conducted for 11 days from September 30, 2013 to October 10, 2013. The author distributed 450 copies and received 430 responses. Finally, 412 copies were used for the analysis after excluding 18 copies having poor answers. Results - The findings about SNS users' behavior could be used as material in the future use of SNS as a marketing tool. Further, the study provided not only theories about SNS characteristics, but also variables and items that were verified during the empirical study. Conclusions - Further studies are needed to overcome the limitations and to establish various kinds of SNS marketing strategies in detail.

• Journal of Information Processing Systems
• /
• 제15권3호
• /
• pp.645-654
• /
• 2019

The purpose of this study was to collect and analyze personal bio data and social network services (SNS) data, derive user preference and user life pattern, and propose intuitive and precise user modeling. This study not only tried to conduct eye tracking experiments using various smart devices to be the ground of the recommendation system considering the attribute of smart devices, but also derived classification preference by analyzing eye tracking data of collected bio data and SNS data. In addition, this study intended to combine and analyze preference of the common classification of the two types of data, derive final preference by each smart device, and based on user life pattern extracted from final preference and collected bio data (amount of activity, sleep), draw the similarity between users using Pearson correlation coefficient. Through derivation of preference considering the attribute of smart devices, it could be found that users would be influenced by smart devices. With user modeling using user behavior pattern, eye tracking, and user preference, this study tried to contribute to the research on the recommendation system that should precisely reflect user tendency.

• 디지털콘텐츠학회 논문지
• /
• 제19권4호
• /
• pp.731-737
• /
• 2018

추천 시스템은 컨텐츠, 온라인 커머스, 소셜 네트워크, 광고 시스템 등 많은 분야에서 사용자가 관심 있을 만한 정보를 선별 제안함을 목적으로 활발하게 연구되고 있다. 그러나 과거 선호도 데이터를 기반으로 제안하는 추천시스템이 많고 과거 데이터가 적거나 없는 사용자를 대상으로는 제공하기 어려우므로 낮은 성능을 보인다는 부문에서 문제점이 있다. 따라서 더욱 고차원적인 데이터 분석에 관한 관심이 증가하고 있고 Matrix Factorization이 주목받고 있다. 이 논문은 그 중 추천시스템에서 주목받는 Factorization Machines Learning(FM)모델과 고차원 데이터 분석인 High-order Factorization Machines Learning(HOFM)의 비교와 재연을 연구하고 제안 한다.

• International Journal of Advanced Culture Technology
• /
• 제11권3호
• /
• pp.327-331
• /
• 2023

This paper proposes a tourist destination recommendation application that combines a chatbot and a recommendation system. The data to be entered into the chatbot was through big data on social media. Through TEXTOM, a total of 22,701 data were collected over a one-year period from January 2022 to January 2023. Non-terms that interfere with analysis were removed through the data purification process. Using refined data, network visualization and CONCOR analysis were used to identify the information users want to obtain about travel to Jeju Island, and categories for each cluster were organized. The content was intuitively organized so that even those who approached it for the first time could easily use it, reducing the difficulty of operating the application. In this paper, users can select their own preferences and receive information. In addition, a tool called a chatbot allows users to focus more on the process of acquiring information by gaining a sense of reality while operating the application. This suggests an application that can reach the purpose of the curator by affecting the user's desire to visit tourist attractions.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2022년도 춘계학술발표대회
• /
• pp.385-387
• /
• 2022

With the tremendous rise in popularity of the Internet and technological advancements, many news keeps generating every day from multiple sources. As a result, the information (News) on the network has been highly increasing. The critical problem is that the volume of articles or news content can be overloaded for the readers. Therefore, the people interested in reading news might find it difficult to decide which content they should choose. Recommendation systems have been known as filtering systems that assist people and give a list of suggestions based on their preferences. This paper studies a personalized news recommendation system to help users find the right, relevant content and suggest news that readers might be interested in. The proposed system aims to build a hybrid system that combines collaborative filtering with content-based filtering to make a system more effective and solve a cold-start problem. Twitter social media data will analyze and build a user's profile. Based on users' tweets, we can know users' interests and recommend personalized news articles that users would share on Twitter.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2023년도 춘계학술발표대회
• /
• pp.389-391
• /
• 2023

In the era of Big Data, humanity is facing a huge overflow of information. To overcome such an obstacle, many new cutting-edge technologies are being introduced. The movie recommendation system is also one such technology. To date, many theoretical and practical kinds of research have been conducted. Our research also focuses on the movie recommendation system by implementing methods from Social Network Analysis(SNA) and Parallel Programming. We applied the Girvan-Newman algorithm to detect communities of users, and a future package to perform the parallelization. This approach not only tries to improve the accuracy of the system but also accelerates the execution time. To do our experiment, we used the MovieLense Dataset.

• 정보과학회 논문지
• /
• 제43권5호
• /
• pp.579-589
• /
• 2016

오늘날 여러 기업에서 조직 구성원들의 지식을 활용하여 핵심역량을 강화할 목적으로 효과적으로 기업 내부의 지식을 관리하는 지식관리시스템을 도입하였다. 그러나 기존의 지식관리시스템은 기업 구성원의 적극적인 참여를 독려할만한 요소가 부족하여 고품질의 지식콘텐츠를 공유하지 못하고 있다. 이에 본 연구에서는 소셜 네트워크 서비스(Social Network Service, SNS)의 구조를 차용한 집단지성 기반 지식관리 시스템을 설계하여 기능에 따른 화면구조, 사용자의 편의 향상과 융합적인 지식콘텐츠 생산을 위한 추천알고리즘을 제시한다. 본 연구에서 제안하는 소셜 네트워크 서비스 기반의 지식관리시스템은 기존 지식관리시스템보다 구성원의 참여를 독려하고 수준 높은 지식콘텐츠를 생산 및 공유할 수 있을 것으로 기대된다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권2호
• /
• pp.583-600
• /
• 2014

Social media is considered a valuable platform for gathering and analyzing the collective and subconscious opinions of people in Internet and mobile environments, where they express, explicitly and implicitly, their daily preferences for brands and products. Extracting and tracking the various attitudes and concerns that people express through social media could enable us to categorize brands and decipher individuals' cognitive decision-making structure in their choice of brands. We investigate the cognitive network structure of consumers by building an inter-category map through the mining of big data. In so doing, we create an improved online recommendation model. Building on economic sociology theory, we suggest a framework for revealing collective preference by analyzing the patterns of brand names that users frequently mention in the online public sphere. We expect that our study will be useful for those conducting theoretical research on digital marketing strategies and doing practical work on branding strategies.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2022년도 추계학술발표대회
• /
• pp.399-401
• /
• 2022

POI (Point-of-Interest) 추천은 다양한 위치 기반 서비스에서 중요한 역할을 있다. 기존 연구에서는 사용자의 모바일 선호도를 모델링하기 위해 과거의 체크인의 공간-시간적 관계를 추출한다. 그러나 사용자 궤적에 숨겨진 개인 방문 경향을 반영할 수 있는 structured feature 는 잘 활용되지 않는다. 이 논문에서는 궤적 그래프를 결합한 시공간 인식 attention 네트워크를 제안한다. 개인의 선호도가 시간이 지남에 따라 변할 수 있다는 점을 고려하면 Dynamic GCN (Graph Convolution Network) 모듈은 POI 들의 공간적 상관관계를 동적으로 집계할 수 있다. LBSN (Location-Based Social Networks) 데이터 세트에서 검증된 새 모델은 기존 모델보다 약 9.0% 성능이 뛰어나다.

• 한국전자통신학회논문지
• /
• 제15권4호
• /
• pp.709-716
• /
• 2020

소셜 미디어에서 일반적으로 게시물을 올릴 때 이미지의 태그 정보를 사용하는데, 태그를 이용하여 주로 검색이 이루어지기 때문이다. 사용자는 태그를 게시물에 붙임으로써 게시물을 많은 사람들에게 노출시키길 원한다. 또한, 사용자는 게시물과 함께 태깅될 태그를 붙이는 행위를 번거롭게 여겨 태깅하지 않은 게시물도 올리게 된다. 본 논문에서는 입력 이미지와 유사한 이미지를 찾아 해당 이미지에 부착된 레이블을 추출하여 그 레이블이 태그로 존재하는 인스타그램의 게시물들을 찾아 게시물 속 존재하는 다른 태그들을 추천해주는 방법을 제안한다. 제안하는 방법에서는 CNN(Convolutional Neural Network) 딥러닝 기법의 모델을 통하여 이미지로 부터 레이블을 추출하여 추출된 레이블로 인스타그램을 크롤링하여 레이블 외의 태그를 정렬하여 추천해준다. 추천된 태그를 이용하여 이미지를 게시하기도 편해지고, 검색의 노출을 높일 수 있고, 검색오류가 적어 높은 정확도를 도출할 수 있음을 알 수 있다.