applsci-logo

Journal Browser

Journal Browser

New Advances for Open-Domain Information Mining in Theories and Applications

A special issue of Applied Sciences (ISSN 2076-3417). This special issue belongs to the section "Computing and Artificial Intelligence".

Deadline for manuscript submissions: closed (20 May 2024) | Viewed by 2552

Special Issue Editors


E-Mail Website
Guest Editor
College of Computer and Information Science, Southwest University, Chongqing 400715, China
Interests: textual data mining; knowledge graphs; graph representation learning; code understanding and representation
Special Issues, Collections and Topics in MDPI journals

E-Mail Website
Guest Editor
Hundred Talents Program of Tongji University, Tongji University, Shanghai 200092, China
Interests: artificial intelligence; knowledge graphs; natural language processing; conversational user interface; intelligent content creation

E-Mail Website
Guest Editor
School of Information Management, Wuhan University, Wuhan 430072, China
Interests: knowledge graph; graph data mining; fintech

E-Mail Website
Guest Editor
School of Computer Science and Engineering, Southeast University, Nanjing 211189, China
Interests: knowledge graph; knowledge representation and reasoning; data mining
Special Issues, Collections and Topics in MDPI journals
Chongqing Research Institute of Big Data, Peking University, Chongqing 401122, China
Interests: knowledge graphs; distributed graph database; graph data management

Special Issue Information

Dear Colleagues,

Open-domain information mining is a crucial task of natural language processing that involves extracting structured and semantic information that can be interpreted easily by a machine or a program, from open-domain-based environments, especially for plain unstructured text. The corresponding results can be formulated as key topics, named entities, summarization, semantic linkage to knowledge graphs, types of knowledge, and so forth. Achieving high-value information mining is a challenging problem. The main difficulties are two-fold: on one hand, the most prevalent models are based on deep learning in open domains, and the creation process highly relies on an unprecedentedly large amount of annotated data for efficient performance. Unfortunately, the models fail to adapt to new emerging key information extraction techniques in many real-life applications, with limited or even a lack of annotation settings. On the other hand, the human language described by the plain text is ambiguous in nature, context matters, and individuals frequently use the same word and acronyms to represent a multitude of different meanings, and thus the models must have a capacity to better comprehend and differentiate the correct semantic signal and noise from different human language contexts (e.g., finding the equivalent or complementary knowledge expression to the same fact across sentences).

In this Special Issue, we welcome researchers from both the academic community and industry to share and discuss their state-of-the-art research on the original algorithmic, methodological, theoretical, or systems-based contributions to open-domain information mining research and relevant applications broadly related to knowledge graphs, social networks, stock prediction, online shopping, recommendation systems, self-driving cars, smart grids, bioinformatics, and medical informatics. Research papers and comprehensive reviews may focus on (but are not restricted to) the following research topics:

  • Document parsing, named entity recognition, topic/knowledge entity/relation/attribute/span/mention/document/argument/event/mathematical information/scientific information/metadata/dataset extraction under a closed-world or open-world assumption;
  • Structured-friendly information extraction from articles, tables, images, and bibliographies;
  • Novel Open IE tools on domain-specific articles and interaction with users;
  • Entity/concept/knowledge graph/document/report/multi-table summarization in open-domain settings;
  • Open-domain taxonomy and knowledge graph construction and learning;
  • Open-domain information representation based on text, images, and graphs, as well as multi-modal data;
  • Transfer learning, multi-agent adaptation, and self-paced learning via open-domain information mining theories and techniques;
  • Visual searching and browsing of structured information from open-domain information mining;
  • Novel applications of open-domain information mining in e-commerce, text mining, stock prediction, recommendation systems, self-driving cars, smart grids, bioinformatics and medical informatics, and so on;
  • Open-domain information mining for explainable AI.

Dr. Yongpan Sheng
Dr. Haofen Wang
Prof. Dr. Liang Hong
Dr. Tianxing Wu
Dr. Wenjie Li
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • open-domain information extraction
  • structured-friendly and semantically information mining
  • open-domain knowledge structure construction and learning
  • applications in open-domain environments
  • explainable AI

Published Papers (1 paper)

Order results
Result details
Select all
Export citation of selected articles as:

Research

23 pages, 4196 KiB  
Article
Trust-Aware Evidence Reasoning and Spatiotemporal Feature Aggregation for Explainable Fake News Detection
by Jing Chen, Gang Zhou, Jicang Lu, Shiyu Wang and Shunhang Li
Appl. Sci. 2023, 13(9), 5703; https://doi.org/10.3390/app13095703 - 5 May 2023
Viewed by 1518
Abstract
Fake news detection has become a significant topic based on the fast-spreading and detrimental effects of such news. Many methods based on deep neural networks learn clues from claim content and message propagation structure or temporal information, which have been widely recognized. However, [...] Read more.
Fake news detection has become a significant topic based on the fast-spreading and detrimental effects of such news. Many methods based on deep neural networks learn clues from claim content and message propagation structure or temporal information, which have been widely recognized. However, firstly, such models ignore the fact that information quality is uneven in propagation, which makes semantic representations unreliable. Additionally, most models do not fully leverage spatial and temporal structures in combination. Finally, internal decision-making processes and results are non-transparent and unexplained. In this study, we developed a trust-aware evidence reasoning and spatiotemporal feature aggregation model for more interpretable and accurate fake news detection. Specifically, we first designed a trust-aware evidence reasoning module to calculate the credibility of posts based on a random walk model to discover high-quality evidence. Next, from the perspective of spatiotemporal structure, we designed an evidence-representation module to capture the semantic interactions granularly and enhance the reliable representation of evidence. Finally, a two-layer capsule network was designed to aggregate the implicit bias in evidence while capturing the false portions of source information in a transparent and interpretable manner. Extensive experiments on two benchmark datasets indicate that the proposed model can provide explanations for fake news detection results, and can also achieve better performance, boosting the F1-score 3.5% on average. Full article
Show Figures

Figure 1

Back to TopTop