As WWW grows at an increasing speed, a classifier targeted hypertext has become in high demand. While document categorization is quite mature, the issue of utilizing structure and hyperlinks been relatively unexplored. In this paper, we propose practical method for enhancing both speed quality using hyperlinks. comparison against recently proposed technique that appears to be only one kind, obtained up 18.5% improvement effectiveness while reducing processing time dramatically. We attempt...

Since postoperative complications after gastrectomy for gastric cancer are associated with poor clinical outcomes, it is important to predict and prepare the occurrence of preoperatively. Conventional models predicting have limitations, prompting interest in machine learning algorithms. Machine a superior ability identify complex interactions among variables nonlinear relationships, potentially revealing new risk factors. This study aimed explore previously overlooked factors compare linear...

Though controversial, water fluoridation has been hailed as one of the top-ten public-health achievements 20th century in United States America. In this article, we aim to investigate public sense reflected on Twitter, using data from 2009 2017. To end, tweets related were collected queries such “fluoridated or fluoride water,” “water and hashtags fluoridation. The ( n = 218,748) examined through informetric, linguistic (word sentiment, word frequency, network analyses), issue tweet...

A question answering (QA) system can be built using multiple QA modules that individually serve as a in and of themselves. This paper proposes learnable, strategy-driven model aims at enhancing both efficiency effectiveness. strategy is learned learning-based classification algorithm determines the sequence to invoked decides when stop invoking additional modules. The invokes most suitable module for given attempts verify answer by consulting other until level confidence reaches threshold....

In the recent explosion of Web information, it is important to find not only appropriate, but also more trustworthy answers user questions. This paper proposes an improved ranking model for question answering (QA) which focused on various answer trustworthiness factors. Contrary past research that simply document quality, we have identified three different factors in multiple layers processes: authority and reputation sources, appropriateness method a given question. Each these used...

Purpose This study explored spatial usage of a public library based on activity logs produced by real users. The purpose this paper is to provide preliminary data for decision-making when establishing the operation policy. Design/methodology/approach To achieve goal, author collected variety including 274,242 seat reservations logs, 3,361,284 collection and 96,098 user information four years in which National Library Korea, Sejong actually operated. crawled were analyzed statistically terms...

Recently there is a need for QA system to answer not only factoid questions but also descriptive questions. Descriptive are which answers that contain definitional information about the search term or describe some special events. We have proposed new model and presented result of we built defined 10 Answer Type(DAT)s as types discussed how our was applied question with experiments.

Purpose The purpose of this paper is to describe the development an algorithm for realizing web crawlers that automatically collect dynamically generated webpages from deep web. Design/methodology/approach This study proposes and develops information as if crawler gathers static by managing script commands links. proposed actually experiments with collecting webpages. Findings Among findings actual crawling process provides search results pages, outcome only collects first page. However, can...

This paper presents a case study of edutainment robot, which is an intelligent robot for educational use with voice-QA model applied. The emphatic functions our are: analyzing spoken question from student, finding appropriate answer in Korean encyclopedia, and then serving the speech synthesis. We develop ESTk, Automatic Speech Recognition (ASR) system based on Finite State Network (FSN) processing questions. For extraction, we utilize machine learning techniques pattern extraction method....

최근 사회망 서비스를 활용하여 사용자 관점에서의 선호지역을 분석하여 맞춤 정보를 제공하기 위한 지능형 위치기반 서비스에 관한 연구가 다수 발표되고 있다. 본 논문에서는 소셜 네트워크 서비스(SNS)에서 생성되는 빅데이터의 내용(content)에 기반하여 추출, 특성에 따른 파악하고 이를 시각화하는 방법을 제안한다. 특히 가장 대표적인 소셜미디어인 트위터를 대상으로 1년치 데이터를 수집, 기초조사를 수행하였다. 기초실험 결과를 통해 성별 및 연령 다양한 사례를 선정, 지역 시간에 따라 구분하고 그에 선호지역과 이동패턴을 분석 제시함으로써 제안된 방법의 효과와 시각화 결과의 활용성을 보인다.

In this paper, we propose an intelligent TV interface using a voice-enable dialogue system. This paper rests on the both directions: new type of management model and its use for practical systems to commercialize. We devise based statistical learning methods. To analyze discourse context, utilize techniques anaphora resolution history management. Contrary rule-based system, develop incremental method construct strategies from training corpus. Several service models equipped with proposed are...

The main motivation of this paper is to devise a way select the best answers collected from multiple web sources. Depending on questions, we need combine QA modules. To end, analyze real-life questions for their characteristics and classify them into different domains genres. In proposed distributed framework, local optimal are selected by several specialized sub-QAs. For fining global answers, merged candidates re-ranked adjusting confidence weights based question analysis. We adopt idea...

See the article "Non-Exposure Endoscopic-Laparoscopic Cooperative Surgery for Stomach Tumors: First Experience from Czech Republic" in Volume 51 on page 167.

Water fluoridation has been cited as one of the top 10 public health achievements 20th century. Herein, we analyzed water articles related to dental caries published in PubMed between 1950 and 2016 using informetrics linguistic methods investigate trends studies. To this aim, queries such “dental (water or fluoridated water),” (fluoride fluoridation),” caries” were submitted retrieve information about on within area fluoride – their titles, abstracts, publication dates, author affiliations,...

Although it has been a long subject of study why researchers prefer some cited documents to others, the existing relative researches have had variety perspectives on nature and complexity citation behavior not provided complete answer this question. In particular, Korea mainly used statistical analysis bibliographic information, which limitations in revealing dynamic complex cognitive aspects process. study, I investigate citer perception citing motives factors through survey compared...

1995년 드라마 <모래시계> 촬영지가 유명 관광지로 각광받고 수많은 관광객들을 불러 모은 이후, 각 지자체는 영화 촬영을 유치하고자 다양한 노력을 기울이고 있다. 지자체 중에서도 특히 전주시는 국제영화제 개최, 전주영상위원회 및 전주영화종합촬영소 설립 등 적극적으로 노력하고 그러나 이러한 풍부한 기반 환경을 갖췄음에도 불구하고, 촬영 이후에 이를 장기적으로 활용하고자 하는 노력은 타 도시에 비해 미흡한 실정이다. 본 연구에서는 이와 같은 한계를 보완하기 위하여 대표적인 SNS(Social Network Service)인 트위터(twitter)를 대상으로 특정 구문을 포함한 트윗을 수집, 그 추이를 분석하였다. 트윗 멘션의 내용 분석을 통해, 연구는 전주 지역의 촬영지와 관광지에 실제로 방문하는 관광객의 주요 방문 요인을 알아보고, 나아가 촬영지의 관광자원화를 위한 방안을 제안하였다. In 1995, the filming location of drama <Hourglass>...

Abstract Since postoperative complications after gastrectomy for gastric cancer are associated with poor clinical outcomes, it is crucial to predict and prepare the occurrence of preoperatively. We evaluated machine learning predicting surgery, emphasizing its advantage in uncovering unnoticed risk factors improving preoperative strategies over linear regression models. retrospectively reviewed cohort data from 865 patients who underwent 2018–2022. A total 83 variables including...