• Genomics & Informatics
• /
• 제19권3호
• /
• pp.24.1-24.7
• /
• 2021

Tracking the most recent advances in Coronavirus disease 2019 (COVID-19)-related research is essential, given the disease's novelty and its impact on society. However, with the publication pace speeding up, researchers and clinicians require automatic approaches to keep up with the incoming information regarding this disease. A solution to this problem requires the development of text mining pipelines; the efficiency of which strongly depends on the availability of curated corpora. However, there is a lack of COVID-19-related corpora, even more, if considering other languages besides English. This project's main contribution was the annotation of a multilingual parallel corpus and the generation of a recommendation dataset (EN-PT and EN-ES) regarding relevant entities, their relations, and recommendation, providing this resource to the community to improve the text mining research on COVID-19-related literature. This work was developed during the 7th Biomedical Linked Annotation Hackathon (BLAH7).

• 경영정보학연구
• /
• 제17권3호
• /
• pp.95-111
• /
• 2015

추천시스템은 과거 구매행동을 통해 사용자가 향후 구매할 것이라 예상되는 제품을 자동으로 검색하여 제공하는 시스템이다. 이러한 추천시스템은 여러 전자상거래 업체에서 도입하고 있으며, 사용자의 편의성 및 수익에 긍정적인 영향을 미치고 있다. 하지만 사용자가 어떠한 기준을 가지고 제품을 평가하는지, 어떠한 요소가 구매 의사 결정에 영향을 미치는지는 반영할 수 없다는 한계가 있다. 이에 본 연구에서는 사용자가 직접 작성한 구매후기를 통해, 사용자 별 제품 평가요소를 활용할 수 있는 추천 모형 알고리즘을 개발하였다. 토픽 모델링을 활용하여 사용자들의 구매후기를 분석하였으며, 이러한 후기의 특성이 반영된 커널과 평가 점수가 반영된 커널 등을 함께 활용하여 다중 커널 학습 기반의 추천 모형을 개발하였다. 또한, 이러한 모형을 BestBuy 사례에 적용하여 검증하였다. 검증 결과, 기존 협업적 필터링 알고리즘보다 다중 커널 학습에 의한 추천 모형의 정확도가 우수하였고, 구매후기의 유사성을 반영하였기에, 사용자가 어떠한 요소를 평가하는지를 확인할 수 있었다. 또한, 기존 협업적 필터링 알고리즘보다 다양한 제품에 대한 추천이 가능함을 확인할 수 있었다. 본 연구는 토픽 모델링과 커널 학습 기반을 사용한 융합적인 추천모형으로서, 온라인 추천시스템의 새로운 방법을 제안한다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2006년도 하계종합학술대회
• /
• pp.49-50
• /
• 2006

Digital satellite systems are usually integrated with terrestrial systems to provide various services, and in these cases they should satisfy the performance objectives defined by the terrestrial systems. Recommendation ITU-R S.1062 specifies the performance of digital satellite systems. The performance objectives were given in terms of bit error probability divided by the average number of errors per burst versus percentage of time. This paper presents theoretical method to estimate performance measure of digital satellite systems defined in Recommendation ITU-R S.1062. We show performance estimation results of duo-binary Turbo codes, and verify them by comparing to the simulation results.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권5호
• /
• pp.1272-1290
• /
• 2013

Privacy-preserving collaborative filtering schemes are becoming increasingly popular because they handle the information overload problem without jeopardizing privacy. However, they may be susceptible to shilling or profile injection attacks, similar to traditional recommender systems without privacy measures. Although researchers have proposed various privacy-preserving recommendation frameworks, it has not been shown that such schemes are resistant to profile injection attacks. In this study, we investigate two memory-based privacy-preserving collaborative filtering algorithms and analyze their robustness against several shilling attack strategies. We first design and apply formerly proposed shilling attack techniques to privately collected databases. We analyze their effectiveness in manipulating predicted recommendations by experimenting on real data-based benchmark data sets. We show that it is still possible to manipulate the predictions significantly on databases consisting of masked preferences even though a few of the attack strategies are not effective in a privacy-preserving environment.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 2004년도 춘계공동학술대회 논문집
• /
• pp.510-514
• /
• 2004

Recommender systems are a personalized information filtering technology to help customers find the products they would like to purchase. Collaborative filtering is known to be the most successful recommendation technology, but its widespread use has exposed some problems such as sparsity and scalability in the e-business environment. In this paper, we propose a recommendation methodology based on Web usage mining and product taxonomy to enhance the recommendation quality and the system performance of original CF-based recommender systems. Web usage mining populates the rating database by tracking customers' shopping behaviors on the Web, so leading to better quality recommendations. The product taxonomy is used to improve the performance of searching for nearest neighbors through dimensionality reduction of the rating database. Several experiments on real e-commerce data show that the proposed methodology provides higher quality recommendations and better performance than original collaborative filtering methodology.

• Asia pacific journal of information systems
• /
• 제12권4호
• /
• pp.121-137
• /
• 2002

Researches on P2P, the information sharing model from person to person, up to date have focused on the technical side and there have been lacking of the business side researches such as customer loyalty. Considering the problem, this study tries to examine empirically in what way the factors of service, market, and customer affect the service loyalty of P2P sites. Results of the study show that the three factors have statistically significant effects on the service loyalty in general. In particular, the results uncover that those factors have different impacts on the reuse intention and the recommendation intention, which are the two measures of service loyalty. For example, the cost of service use affects the reuse intention significantly whereas the same element does not affect the recommendation intention. Interestingly, some of the results are not coincide with the results of previous studies and do not meet general expectation. For example, users' tendency of pursuing variety has positive impacts on the service loyalty, especially, the intention of recommendation.

• Management Science and Financial Engineering
• /
• 제12권2호
• /
• pp.71-85
• /
• 2006

Collaborative filtering, among other recommender systems, has been known as the most successful recommendation technique. However, it requires the user-item rating data, which may not be easily available. As an alternative, some collaborative filtering algorithms have been developed recently by utilizing the market basket data in the form of the binary user-item matrix. Viewing the recommendation scheme as a two-class classification problem, we proposed a new collaborative filtering scheme using a regularized discriminant analysis applied to the binary user-item data. The proposed discriminant model was built in terms of the major principal components and was used for predicting the probability of purchasing a particular item by an active user. The proposed scheme was illustrated with two modified real data sets and its performance was compared with the existing user-based approach in terms of the recommendation precision.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1994년도 추계학술대회 논문집
• /
• pp.658-663
• /
• 1994

The classification and coding of part for group technology applications continus to be labour intensive and time-consuming process, and therefore much effort is dedicated to the structure and creation of automatic coding systems. IN this paper, Neural networks is used to generate processes-related digit as well as part geometry-related digit of the TS code where part name is provided as input.since part name, which is appropriately designated, provides much information about part geometry and manufacturing processes. THe developed GT recommendation system is integrated with interactive TS coding system and database in order to handle the changes of production environment, such as the change of production part of plant. It is found to recommend codes accurately and promises to be a useful tool for consistent, reliable and convenient coding processes.

• Journal of Information Processing Systems
• /
• 제17권2호
• /
• pp.352-368
• /
• 2021

Sign-in point of interest (POI) are extremely sparse in location-based social networks, hindering recommendation systems from capturing users' deep-level preferences. To solve this problem, we propose a content-aware POI recommendation algorithm based on a convolutional neural network. First, using convolutional neural networks to process comment text information, we model location POI and user latent factors. Subsequently, the objective function is constructed by fusing users' geographical information and obtaining the emotional category information. In addition, the objective function comprises matrix decomposition and maximisation of the probability objective function. Finally, we solve the objective function efficiently. The prediction rate and F1 value on the Instagram-NewYork dataset are 78.32% and 76.37%, respectively, and those on the Instagram-Chicago dataset are 85.16% and 83.29%, respectively. Comparative experiments show that the proposed method can obtain a higher precision rate than several other newer recommended methods.

• 데이타베이스연구회지:데이타베이스연구
• /
• 제34권3호
• /
• pp.99-123
• /
• 2018

웹상에서의 기하급수적으로 증가하는 정보의 양으로 인해, 중요하고 가치 있는 데이터를 변별 해 내는 작업은 그 어느 때보다도 중요하다고 하겠다. 추천 시스템은 이러한 정보의 과 공급 문제를 해결하기 위한 가장 효과적인 방법 중 하나임에도 불구하고, 그 성능은 기존 방식들에서 크게 진전을 이루지 못하고 있는 것이 사실이다. 따라서 본 논문에서는 이 문제를 진전시키기 위해, 협업태그를 활용한 새로운 사용자 프로파일링 기법을 제안하고 사용자의 평가 및 태깅패턴을 분석, 그 활용 또한 모색한다. 본 논문에서 제안하는 기법의 검증을 위해, 해당 프로파일링 기법을 활용 한 혼합 영화 추천 시스템을 구현하고 실제 데이터를 사용하여 기존의 추천 방식 대비 그 경쟁력을 검증하였다. 그와 더불어, 민감도 분석을 통해 사용자의 태깅패턴과 평가패턴에 기반한 차별적인 추천 방식의 잠재적 가능성 또한 제안, 검증한다.