Diese Datenbank enthält über 40.000 Dokumente zu Themen aus den Bereichen Formalerschließung – Inhaltserschließung – Information Retrieval.
© 2015 W. Gödert, TH Köln, Institut für Informationswissenschaft / Powered by litecat, BIS Oldenburg (Stand: 04. Juni 2021)
1Lee, Y.-Y. ; Ke, H. ; Yen, T.-Y. ; Huang, H.-H. ; Chen, H.-H.: Combining and learning word embedding with WordNet for semantic relatedness and similarity measurement.
In: Journal of the Association for Information Science and Technology. 71(2020) no.6, S.657-670.
Abstract: In this research, we propose 3 different approaches to measure the semantic relatedness between 2 words: (i) boost the performance of GloVe word embedding model via removing or transforming abnormal dimensions; (ii) linearly combine the information extracted from WordNet and word embeddings; and (iii) utilize word embedding and 12 linguistic information extracted from WordNet as features for Support Vector Regression. We conducted our experiments on 8 benchmark data sets, and computed Spearman correlations between the outputs of our methods and the ground truth. We report our results together with 3 state-of-the-art approaches. The experimental results show that our method can outperform state-of-the-art approaches in all the selected English benchmark data sets.
Themenfeld: Semantisches Umfeld in Indexierung u. Retrieval
2Huang, H.-H. ; Wang, J.-J. ; Chen, H.-H.: Implicit opinion analysis : extraction and polarity labelling.
In: Journal of the Association for Information Science and Technology. 68(2017) no.9, S.2076-2087.
Abstract: Opinion words are crucial information for sentiment analysis. In some text, however, opinion words are absent or highly ambiguous. The resulting implicit opinions are more difficult to extract and label than explicit ones. In this paper, cutting-edge machine-learning approaches - deep neural network and word-embedding - are adopted for implicit opinion mining at the snippet and clause levels. Hotel reviews written in Chinese are collected and annotated as the experimental data set. Results show the convolutional neural network models not only outperform traditional support vector machine models, but also capture hidden knowledge within the raw text. The strength of word-embedding is also analyzed.
Inhalt: Vgl.: http://onlinelibrary.wiley.com/doi/10.1002/asi.23835/full.
3Huang, H. ; Chu, S. K.-W. ; Chen, D. Y.-T.: Interactions between English-speaking and Chinese-speaking users and librarians on social networking sites.
In: Journal of the Association for Information Science and Technology. 66(2015) no.6, S.1150-1166.
Abstract: Social networking sites (SNSs) can encourage interaction among users. Existing research mainly focuses on the ways in which SNSs are used in libraries and on librarians' or users' attitudes towards these SNSs. This study focused on the flow of information via SNS interactions between librarians and users on library Facebook, Twitter, and Chinese Weibo sites, and developed an SNS user interaction type model based on these information flows. A mixed-method approach was employed combining quantitative data generated from the analysis of 1,753 posts sampled from 40 library SNSs and qualitative data from interviews with 10 librarians. Four types of interactions were identified: information/knowledge sharing, information dissemination, communication, and information gathering. The study found that SNSs were used primarily as channels for disseminating news and announcements about things currently happening in the library. Communication allowed open-ended questions and produced more replies. In Facebook posts, Chinese Facebook users generated less "likes" than English-speaking users. The comparison of data between Facebook-like and Twitter-like SNSs in different library settings suggested that libraries need to coordinate different types of SNSs, and take library settings and sociocultural environments into consideration in order to enhance and encourage user engagement and interaction.
Inhalt: Vgl.: http://onlinelibrary.wiley.com/doi/10.1002/asi.23251/abstract.
4Huang, H. ; Jörgensen, C.: Characterizing user tagging and Co-occurring metadata in general and specialized metadata collections.
In: Journal of the American Society for Information Science and Technology. 64(2013) no.9, S.1878-1889.
Abstract: This study aims to identify the categorical characteristics and usage patterns of the most popular image tags in Flickr. The "metadata usage ratio" is introduced as a means of assessing the usage of a popular tag as metadata. We also compare how popular tags are used as image tags or metadata in the Flickr general collection and the Library of Congress's photostream (LCP), also in Flickr. The Flickr popular tags in the list overall are categorically stable, and the changes that do appear reflect Flickr users' evolving technology-driven cultural experience. The popular tags in Flickr had a high number of generic objects and specific locations-related tags and were rarely at the abstract level. Conversely, the popular tags in the LCP describe more in the specific objects and time categories. Flickr users copied the Library of Congress-supplied metadata that related to specific objects or events and standard bibliographic information (e.g., author, format, time references) as popular tags in the LCP. Those popular tags related to generic objects and events showed a high metadata usage ratio, while those related to specific locations and objects showed a low image metadata usage ratio. Popular tags in Flickr appeared less frequently as image metadata when describing specific objects than specific times and locations for historical images in Flickr LCP collections. Understanding how people contribute image tags or image metadata in Flickr helps determine what users need to describe and query images, and could help improve image browsing and retrieval.
Themenfeld: Social tagging
Behandelte Form: Bilder
5Huang, H. ; Stvilia, B. ; Jörgensen, C. ; Bass, H.W.: Prioritization of data quality dimensions and skills requirements in genome annotation work.
In: Journal of the American Society for Information Science and Technology. 63(2012) no.1, S.195-207.
Abstract: The rapid accumulation of genome annotations, as well as their widespread reuse in clinical and scientific practice, poses new challenges to management of the quality of scientific data. This study contributes towards better understanding of scientists' perceptions of and priorities for data quality and data quality assurance skills needed in genome annotation. This study was guided by a previously developed general framework for assessment of data quality and by a taxonomy of data-quality (DQ) skills, and intended to define context-sensitive models of criteria for data quality and skills for genome annotation. Analysis of the results revealed that genomics scientists recognize specific sets of criteria for quality in the genome-annotation context. Seventeen data quality dimensions were reduced to 5-factor constructs, and 17 relevant skills were grouped into 4-factor constructs. The constructs defined by this study advance the understanding of data quality relationships and are an important contribution to data and information quality research. In addition, the resulting models can serve as valuable resources to genome data curators and administrators for developing data-curation policies and designing DQ-assurance strategies, processes, procedures, and infrastructure. The study's findings may also inform educators in developing data quality assurance curricula and training courses.
6Huang, H. ; Andrews, J. ; Tang, J.: Citation characterization and impact normalization in bioinformatics journals.
In: Journal of the American Society for Information Science and Technology. 63(2012) no.3, S.490-497.
Abstract: Bioinformatics journals publish research findings of intellectual synergies among subfields such as biology, mathematics, and computer science. The objective of this study is to characterize the citation patterns in bioinformatics journals and their correspondent knowledge subfields. Our study analyzed bibliometric data (impact factor, cited-half-life, and references-per-article) of bioinformatics journals and their related subfields collected from the Journal Citation Reports (JCR). The findings showed that bioinformatics journals' citations are field-dependent, with scattered patterns in article life span and citing propensity. Bioinformatics journals originally derived from biology-related subfields have shorter article life spans, more citing on average, and higher impact factors. Those journals, derived from mathematics and statistics, demonstrate converse citation patterns. Journal impact factors were normalized, taking into account the impacts of article life spans and citing propensity. A comparison of these normalized factors to JCR journal impact factors showed rearrangements in the ranking orders of a number of individual journals, but a high overall correlation with JCR impact factors.
7Sawyer, S. ; Huang, H.: Conceptualizing information, technology, and people : comparing information science and information.
In: Journal of the American Society for Information Science and Technology. 58(2007) no.10, S.1436-1447.
Abstract: Through this article, we highlight that there are discernibly different patterns among conceptualizations of information, technology, and people across information systems and information science literatures. We do this to clarify the differences in these two areas of scholarship and to further encourage the substantial overlap possible, but not yet engaged, in the research pursued in these areas. We engage this by analyzing published literature in these areas to frame our discussion of the challenges and opportunities for scholars in information science and information systems disciplines to engage in collaborative work.