• 디지털융복합연구
• /
• 제17권1호
• /
• pp.123-130
• /
• 2019

웹 추천기법에서 가장 많이 사용하는 방식 중의 하나는 협업필터링 기법이다. 협업필터링 관련 많은 연구에서 정확도를 개선하기 위한 방안이 제시되어 왔다. 본 연구는 Word2Vec과 앙상블 합성곱 신경망을 활용한 영화추천 방안에 대해 제안한다. 먼저 사용자, 영화, 평점 정보에서 사용자 문장과 영화 문장을 구성한다. 사용자 문장과 영화 문장을 Word2Vec에 입력으로 넣어 사용자 벡터와 영화 벡터를 구한다. 사용자 벡터는 사용자 합성곱 모델에 입력하고, 영화 벡터는 영화 합성곱 모델에 입력한다. 사용자 합성곱 모델과 영화 합성곱 모델은 완전연결 신경망 모델로 연결된다. 최종적으로 완전연결 신경망의 출력 계층은 사용자 영화 평점의 예측값을 출력한다. 실험결과 전통적인 협업필터링 기법과 유사 연구에서 제안한 Word2Vec과 심층 신경망을 사용한 기법에 비해 본 연구의 제안기법이 정확도를 개선함을 알 수 있었다.

• 한국정보처리학회논문지
• /
• 제6권11S호
• /
• pp.3399-3409
• /
• 1999

Information filtering for internet search, in which new information retrieval environment is given, is different from traditional methods such as bibliography information filtering, news-group and E-mail filtering. Therefore, we cannot expect high performance from the traditional information filtering models when they are applied to the new environment. To solve this problem, we inspect the characteristics of the new filtering environment, and propose a semantic-based filtering model which includes a new filtering method using WordNet. For extracting keywords from documents, this model uses the SDCC(Semantic Distance for Common Category) algorithm instead of the TF/IDF method usually used by traditional methods. The world sense ambiguation problem, which is one of causes dropping efficiency of internet search, is solved by this method. The semantic-based filtering model can filter web pages selectively with considering a user level and we show in this paper that it is more convenient for users to search information in internet by the proposed method than by traditional filtering methods.

• 한국정보통신학회논문지
• /
• 제7권2호
• /
• pp.334-343
• /
• 2003

본 논문은 청소년들의 적극적인 정보제공 활동 시점에 발생하고 있는 정보화 역기능 현상을 해결하는 방안으로 인터넷 게시판에서 유해단어 필터링에 의한 정보윤리 교육 기법을 제안한다. 인터넷 게시판에서 유해단어 필터링 기법은 초ㆍ중등학교 홈페이지 게시판에서 나타나는 부적절한 행동과 상대방에 대한 성적 모욕, 욕설의 사용, 상대방 비하 등에 관련된 유해단어를 추출하여 유해단어 사전을 구축하고 필터링하는 방법이다 필터링 된 결과에 따라 글 쓰는 시점에서 정보윤리 컨텐츠를 제공한다. 이 기법을 학교 홈페이지 게시판에 적용한 결과 유해단어 사용과 부적절한 쓰기에서 90% 이상의 감소효과를 나타내었다.

• International Journal of Advanced Culture Technology
• /
• 제5권1호
• /
• pp.83-89
• /
• 2017

In this paper, we propose an extended method of one-hot Long Short-Term Memory (LSTM) and evaluate the performance on spam filtering task. Most of traditional methods proposed for spam filtering task use word occurrences to represent spam or non-spam messages and all syntactic and semantic information are ignored. Major issue appears when both spam and non-spam messages share many common words and noise words. Therefore, it becomes challenging to the system to filter correct labels between spam and non-spam. Unlike previous studies on information filtering task, instead of using only word occurrence and word context as in probabilistic models, we apply a neural network-based approach to train the system filter for a better performance. In addition to one-hot representation, using term weight with attention mechanism allows classifier to focus on potential words which most likely appear in spam and non-spam collection. As a result, we obtained some improvement over the performances of the previous methods. We find out using region embedding and pooling features on the top of LSTM along with attention mechanism allows system to explore a better document representation for filtering task in general.

• 정보처리학회논문지B
• /
• 제16B권1호
• /
• pp.85-92
• /
• 2009

오늘날 웹이 일반화되면서 사람들은 원하는 정보를 웹을 통해 얻고, 또한 제공하고 있다. 웹이 다양한 정보의 제공과 습득의 장이라는 편의성을 제공하고 있지만, 반면에 너무 많은 정보, 무분별한 유해 정보의 범람 등 여러 가지 문제를 내포하고 있다. 현재 유해 웹 문서를 분류하기 위한 다양한 방법이 연구되고 사용되고 있다. 그러나 각각의 방법들이 갖는 단점들로 인해 획기적인 성과를 내지 못하고 있다. 본 논문에서는 유해 정보로부터 사회적으로 보호를 받아야 할 사용자들을 보호하기 위한 수단으로 유해 웹 문서 차단 방법에 대해 제안하고자 한다. 본 논문에서는 키워드 필터링과 SVM 알고리즘을 이용한 2단계 분류 과정을 통해 분류의 정확률을 높이고자 하였다.

• 통합자연과학논문집
• /
• 제14권4호
• /
• pp.183-188
• /
• 2021

Collaborative filtering is one of the commonly used methods in the web recommendation system. Numerous researches on the collaborative filtering proposed the numbers of measures for enhancing the accuracy. This study suggests the movie recommendation system applied with Word2Vec and ensemble convolutional neural networks. First, user sentences and movie sentences are made from the user, movie, and rating information. Then, the user sentences and movie sentences are input into Word2Vec to figure out the user vector and movie vector. The user vector is input on the user convolutional model while the movie vector is input on the movie convolutional model. These user and movie convolutional models are connected to the fully-connected neural network model. Ultimately, the output layer of the fully-connected neural network model outputs the forecasts for user, movie, and rating. The test result showed that the system proposed in this study showed higher accuracy than the conventional cooperative filtering system and Word2Vec and deep neural network-based system suggested in the similar researches. The Word2Vec and deep neural network-based recommendation system is expected to help in enhancing the satisfaction while considering about the characteristics of users.

• 한국멀티미디어학회논문지
• /
• 제19권11호
• /
• pp.1878-1890
• /
• 2016

Web 2.0 has features produced the content through the user of the participation and share. The content production activities have became active since social network service appear. The social bookmark, one of social network service, is service that lets users to store useful content and share bookmarked contents between personal users. Unlike Internet search engines such as Google and Naver, the content stored on social bookmark is searched based on tag keyword information and unnecessary information can be excluded. Social bookmark can make users access to selected content. However, quick access to content that users want is difficult job because of the user of the participation and share. Our paper suggests a method recommending search word to be able to access quickly to content. A method is suggested by using Collaborative Filtering and Jaccard similarity coefficient. The performance of suggested system is verified with experiments that compare by 'Delicious' and "Feeltering' with our system.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제3권7호
• /
• pp.271-276
• /
• 2014

휴대폰에서 문자 메시지 전송 기능은 현대인들에게 매우 편리한 새로운 형태의 의사소통 방식이다. 반면에 문자 메시지 기능을 악용한 광고성 문자들이 너무 많이 쏟아져서 휴대폰 사용자들은 스팸 문자 공해에 시달리는 심각한 부작용을 낳게 되었다. 광고성 문자를 발송하는 사람들은 문자 메시지가 자동으로 차단되는 것을 회피하기 위해 한글 문장을 다양한 형태로 변형하거나 왜곡시키고 있으며, 이러한 문자 메시지를 자동으로 차단하기 위해서는 변형되거나 왜곡된 문장들을 정상적인 한글 문장으로 정규화하는 기술이 필수적이다. 본 논문에서는 변형되거나 왜곡된 광고성 문자 메시지를 정상적인 문장으로 정규화하고 정규화된 문장으로부터 자동 띄어쓰기 및 복합명사 분해 과정을 거쳐 키워드를 추출하기 위한 방법을 제안하였다.

• 한국문헌정보학회지
• /
• 제14권
• /
• pp.93-130
• /
• 1987

Scientific communication is an information exchange activity between scientists. Scientific communication is carried out in a variety of informal and formal ways. Basically, informal communication takes place by word of mouth, whereas formal communication occurs via the written word. Science is a highly interdependent activity in which each scientist builds upon the work of colleagues past and present. Consequently, science depends heavily on scientific communication. In this study, three mathematical models, namly Brillouin measure, logistic equation, and Markov chain are examined. These models provide one with a means of describing and predicting the behavior of scientific communication process. These mathematical models can be applied to construct quality filtering algorithms for subject literature which identify synthesized elements (authors, papers, and journals). Each suggests a different type of application. Quality filtering for authors can be useful to funding agencies in terms of identifying individuals doing the best work in a given area or subarea. Quality filtering with respect to papers can be useful in constructing information retrieval and dissemination systems for the community of scientists interested m the field. The quality filtering of journals can be a basis for the establishment of small quality libraries based on local interests in a variety of situations, ranging from the collection of an individual scientist or physician to research centers to developing countries. The objective of this study is to establish the theoretical framework for informetrics which is defined as the quantitative analysis of scientific communication, by investigating mathematical models of scientific communication.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권1호
• /
• pp.305-318
• /
• 2022

Abusive behaviors have become a common issue in many online social media platforms. Profanity is common form of abusive behavior in online. Social media platforms operate the filtering system using popular profanity words lists, but this method has drawbacks that it can be bypassed using an altered form and it can detect normal sentences as profanity. Especially in Korean language, the syllable is composed of graphemes and words are composed of multiple syllables, it can be decomposed into graphemes without impairing the transmission of meaning, and the form of a profane word can be seen as a different meaning in a sentence. This work focuses on the problem of filtering system mis-detecting normal phrases with profane phrases. For that, we proposed the deep learning-based framework including grapheme and syllable separation-based word embedding and appropriate CNN structure. The proposed model was evaluated on the chatting contents from the one of the famous online games in South Korea and generated 90.4% accuracy.