Search | Korea Science

Seol, Jinah
- Journal of Internet Computing and Services
- /
- v.19 no.4
- /
- pp.123-133
- /
- 2018
This paper examined how the millennial generation consumes mobile news through social networking sites with regards to user patterns, preference topics and news values, and whether news topics and news values may influence their overall mobile SNS news consumption and interactivity. The findings show that more than 2/3 of respondents consumed mobile SNS news at least once everyday for 30minutes to one-hour. Male millennials tended to use Facebook and Kakao-talk more than female. While the portal site was the most accessed channel for consuming mobile news, SNS was the second, more than the combined use of national daily papers, TV, and internet newspapers. The respondents' demographic characteristics and news topics also affect the form and degree of news interactivity. With regards to their preferences and prioritization of news values, millennials tend to perceive 'impact' and 'usefulness' as being most important, despite the differences of their demographic characteristics. They also preferred those news values most. There were significant differences in terms of preferred news topics according to the demographics' characteristics.
https://doi.org/10.7472/jksii.2018.19.4.123 인용 PDF KSCI

Wang, Guanwen;Yu, Zhengtao;Xian, Yantuan;Zhang, Yu
- Journal of Information Processing Systems
- /
- v.17 no.6
- /
- pp.1057-1070
- /
- 2021
Case-related news filtering is crucial in legal text mining and divides news into case-related and case-unrelated categories. Because case-related news originates from various fields and has different writing styles, it is difficult to establish complete filtering rules or keywords for data collection. In addition, the labeled corpus for case-related news is sparse; therefore, to train a high-performance classification model, it is necessary to annotate the corpus. To address this challenge, we propose topic-enhanced positive-unlabeled learning, which selects positive and negative samples guided by topics. Specifically, a topic model based on a variational autoencoder (VAE) is trained to extract topics from unlabeled samples. By using these topics in the iterative process of positive-unlabeled (PU) learning, the accuracy of identifying case-related news can be improved. From the experimental results, it can be observed that the F1 value of our method on the test set is 1.8% higher than that of the PU learning baseline model. In addition, our method is more robust with low initial samples and high iterations, and compared with advanced PU learning baselines such as nnPU and I-PU, we obtain a 1.1% higher F1 value, which indicates that our method can effectively identify case-related news.
https://doi.org/10.3745/JIPS.01.0081 인용 PDF KSCI