• Asia pacific journal of information systems
• /
• v.30 no.4
• /
• pp.719-740
• /
• 2020

The stock market changes continuously as new information emerges, affecting the judgments of investors. Online news articles are valued as a traditional window to inform investors about various information that affects the stock market. This paper proposed new ways to utilize online news articles with technical indicators. The suggested hybrid model consists of three models. First, a self-attention-based convolutional neural network (CNN) model, considered to be better in interpreting the semantics of long texts, uses news content as inputs. Second, a self-attention-based, bi-long short-term memory (bi-LSTM) neural network model for short texts utilizes news titles as inputs. Third, a bi-LSTM model, considered to be better in analyzing context information and time-series models, uses 19 technical indicators as inputs. We used news articles from the previous day and technical indicators from the past seven days to predict the share price of the next day. An experiment was performed with Korean stock market data and news articles from 33 top companies over three years. Through this experiment, our proposed model showed better performance than previous approaches, which have mainly focused on news titles. This paper demonstrated that news titles and content should be treated in different ways for superior stock price prediction.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.5
• /
• pp.2087-2095
• /
• 2012

The editorial elements which are used in this paper are made up of news categories, photograph articles, titles of article written bold strokes and contents of article. Of these elements, only the three elements of photographic articles, the titles of article written bold strokes and contents of article had some effects on the choice of articles. For the portal news, only the news categories, the titles of article written bold strokes and the name of newspaper had an effect on the choice of articles. Of the news genres such as politics, business, social affairs, sports, culture/entertainment, world news and IT/science, only the three genres of social affairs, culture/entertainment and world news had some effects on news users' choice. For the difference between man group and woman group in analyzing the choice of articles, there was the difference in four elements of business, sports, culture/entertainment and IT/science.

• The Journal of the Korea Contents Association
• /
• v.20 no.6
• /
• pp.556-578
• /
• 2020

There are many different researches about how to analyze issues based on real-time news streams. But, there are few researches which analyze issues hierarchically from news articles and even a previous research of hierarchical issue analysis make clustering speed slower as the increment of news articles. In this paper, we propose a hierarchical and incremental clustering for semi real-time issue analysis on news articles. We trained siamese neural network based weighted cosine similarity model, applied this model to k-means algorithm which is used to make word clusters and converted news articles to document vectors by using these word clusters. Finally, we initialized an issue cluster tree from document vectors, updated this tree whenever news articles happen, and analyzed issues in semi real-time. Through the experiment and evaluation, we showed that up to about 0.26 performance has been improved in terms of NMI. Also, in terms of speed of incremental clustering, we also showed about 10 times faster than before.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.12
• /
• pp.8195-8209
• /
• 2015

The purpose of this paper is to propose a method for recommendation and personalization of important news articles based on evaluating news value. Evaluation of news is the approach by which editors select news articles for cover-story in traditional offline news papers area. For this, my study proposes a suite of methods to select and personalize a set of news based on evaluating news articles, not just on the personal preference for them. The aforementioned the value of news articles including social impact, novelty, relevance to each audience, and human interest, all of which have been factorized in many previous studies, is a main concept for a procedural and structural application methodology deduced in this study. After a comparative case study with other online news services, it was shown that my research provides more effective way to select important news articles in terms of user satisfaction than others.

• Journal of Digital Contents Society
• /
• v.19 no.4
• /
• pp.631-637
• /
• 2018

In order to read newspaper articles, the use of paper newspapers is decreasing and smartphone are increasingly used. As a result, the number of news apps continues to increase. Many of the news apps in the Android Play Store fall into two categories. The first is an app that is developed by a specific newspaper company and distributes only the articles of the newspaper company. The rest is an app that shows a list of newspapers and shows the homepage when a newspaper is selected. In this paper, we have designed and implemented a Real-Time News app for collecting articles from many newspapers and providing them in real time. Newspapers provide up-to-date articles with RSS feeds. The server program stores them in the DB, and transmits the articles requested in the Real-Time News app in real time. In order to see the latest news, it is possible to collect the articles of each newspaper without visiting the websites of the various newspapers, and it is possible to reduce the mobile data usage used to access each website.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.8
• /
• pp.1015-1021
• /
• 2020

Mobile news users want news articles without advertising, meanwhile the news providers require advertisement displays in several types to attain advertising revenue. In this paper, we classified the types of advertisements on mobile news articles into fixed article type which is fixed on some areas of articles, fixed screen type which is fixed on mobile screens, and a combination type of them. In addition, we proposed a smart solution based on overlapping method which effectively organize advertisements to not distract the readers. The proposed method is similar to fixed article type and overlapping technique of advertisements on news article's photo or virtual area. The performance evaluation result shows that the proposed method provides more spaces for news articles effectively than the existing methods. Although only some areas of advertisements may be blocked according to the number or size of advertisements, the effect is not critical.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.8
• /
• pp.999-1005
• /
• 2020

While search portals' 'Portal News' account for the largest portion of aggregated news outlet, its neutrality as an outlet is questionable. This is because news aggregation may lead to prejudiced information consumption by recommending biased news articles. In this paper we introduce a new method of measuring political bias of news articles by using deep learning. It can provide its readers with insights on critical thinking. For this method, we build the dataset for deep learning by analyzing articles' bias from keywords, sourced from the National Assembly proceedings, and assigning bias to said keywords. Based on these data, news article bias is calculated by applying deep learning with a combination of Convolution Neural Network and Recurrent Neural Network. Using this method, 95.6% of sentences are correctly distinguished as either conservative or progressive-biased; on the entire article, the accuracy is 46.0%. This enables analyzing any articles' bias between conservative and progressive unlike previous methods that were limited on article subjects.

• Journal of the Korean Society for Library and Information Science
• /
• v.53 no.2
• /
• pp.159-178
• /
• 2019

The purpose of the study is to examine the trend in the change of relation between countries (Czech and Korea) through analysis of portal new articles. In order to achieve the purpose, we analyzed news articles about Czech from 1990 to March 31st, 2019. We divided it into 6 periods by every 5 years, reviewed 200 news articles for each period totaling 1,200 news articles, and categorized them into 4 categories by subject (politics, economy, society and culture, and educations). The result of the study showed the subject of society and culture represented the largest proportion of all news articles. We also found that the range of changes in the sub-categories of society and culture occurred most extensively. We concluded the paper with several suggestions that could promote cooperation between Korea and Czech.

• Jounal of The Korea Society of Information Technology Policy & Management
• /
• v.11 no.6
• /
• pp.1503-1507
• /
• 2019

With the softness of online entertainment news articles and the increasing number of quick-reporting articles in the entertainment sector, many people have access to entertainment front-page articles and are now able to make reviews of celebrities. It is not easy to systematically analyze which news articles are about which celebrities in a real-time environment, although their reputation is a key factor in the entertainment agency's business strategy, which should make the most of its affiliated celebrity resources. Based on the amount of celebrity references mentioned in entertainment news data, this paper proposes an entertainment news keyword analysis system, which extracts celebrities that are the subject of the article and associates them with the celebrity entertainment agency in question. Through the system proposed in this paper, advertisers or entertainment agencies can judge the value of the celebrity as reference material for the business. In addition, it can lay the groundwork for an investment strategy by predicting the outlook for the entertainment company for brokerages and investors.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.11 no.6
• /
• pp.63-71
• /
• 2011

The current Web portal sites provide significant keywords with high popularity or importance; specifically, user-friendly services such as tag clouds and associated word search are provided. However, in general, since news articles are classified only with their date and categories, it is not easy for users to find other articles related to some articles while reading news articles classified with categories. And the conventional associated keyword service has not satisfied users sufficiently because it depends only upon user queries. This paper proposes a way of searching news articles by utilizing the keywords tightly associated with users' queries. Basically, the proposed method discovers a set of keyword association patterns by using the association rule mining technique that extracts association patterns for keywords by focusing upon sentences containing some keywords. The method enables users to navigate the space of associated keywords hidden in large news articles.