Article
Clinical Neurology
Kristen Barbour, Niu Tian, Elissa G. G. Yozawitz, Steven Wolf, Patricia E. E. McGoldrick, Tristan T. T. Sands, Aaron Nelson, Natasha Basma, Zachary M. M. Grinspan
Summary: This study evaluated the use of keyword search as an alternative method for identifying individuals with rare epilepsies in electronic health records. The results showed that keyword search was effective in identifying rare epilepsy cases and promoting their access to specialized care, clinical research, and support groups.
Article
Computer Science, Artificial Intelligence
Sebastian Schultheiss, Dirk Lewandowski, Sonja von Mach, Nurce Yagci
Summary: Search engine queries are often used in studies to make statements about social phenomena, but the queries used in these studies are usually not systematic and do not reflect actual user behavior. We developed a method called query sampler, which samples queries from commercial search engines using keyword research tools, to address this problem. Our approach allows for the generation of large numbers of queries related to a given topic and provides information on query volume. Empirical testing showed that our approach can significantly expand the number of queries and total search volume. It has wide applications for studies using search engine queries to draw conclusions about social phenomena.
PEERJ COMPUTER SCIENCE
(2023)
Article
Computer Science, Software Engineering
Fatmana Senturk, Gurhan Gunduz
Summary: Big data has attracted the attention of governments and many companies today, with search engines playing a crucial role in accessing information on the Internet. This work serves as a guide to help understand the workings of search engines and access desired information more effectively.
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
(2022)
Article
Computer Science, Artificial Intelligence
Xin Hu, Jiangli Duan, Depeng Dang
Summary: This study introduces a framework that incorporates keyword search into natural language question answering to address the shortcomings of existing methods. Experimental results demonstrate that this framework can answer more questions effectively.
KNOWLEDGE AND INFORMATION SYSTEMS
(2021)
Article
Multidisciplinary Sciences
Robert Epstein, Vivian Lee, Roger Mohr, Vanessa R. Zankich
Summary: ABE, also known as Answer Bot Effect, is a relatively new form of influence that can shift opinions and voting preferences without people's awareness. Through three experiments, researchers found that answer boxes and interactions with intelligent personal assistants (IPAs) can significantly impact people's voting preferences, posing a serious threat to democracy and human autonomy.
Article
Computer Science, Information Systems
Arif Usta, Akifhan Karakayali, Ozgur Ulusoy
Summary: Translating natural language queries (NLQs) into structured query language (SQL) in relational databases is a challenging task. This study proposes a deep learning based supervised approach called DBTagger that utilizes POS tags of NLQs to address the keyword mapping problem. DBTagger achieves state-of-the-art accuracy results and is significantly faster than traditional methods, making it practical for various relational databases.
PROCEEDINGS OF THE VLDB ENDOWMENT
(2021)
Article
Mathematical & Computational Biology
Michael Gusenbauer, Neal R. Haddaway
Summary: The COVID-19 pandemic has highlighted the limitations of academic searching capabilities, calling for improvements in search methods and systems. Researchers require dedicated education and training to enhance search efficiency. Matching goals, heuristics, and systems is key to improving academic searching.
RESEARCH SYNTHESIS METHODS
(2021)
Review
Computer Science, Information Systems
Yanwu Yang, Huiran Li
Summary: In this paper, an overarching framework for keyword decisions in sponsored search advertising is presented, which includes four levels of keyword decisions: domain-specific keyword pool generation, keyword targeting, keyword assignment and grouping, and keyword adjustment. Using this framework, the state-of-the-art research literature on keyword decisions is reviewed, considering techniques, input features, and evaluation metrics. Evolving issues, potential gaps, and novel research perspectives for future exploration are also discussed.
INFORMATION PROCESSING & MANAGEMENT
(2023)
Article
Computer Science, Information Systems
Yang Zhou, Chenjiao Zhi, Feng Xu, Weiwei Cui, Huaqiong Wang, Aihong Qin, Xiaodiao Chen, Yaqi Wang, Xingru Huang
Summary: This paper proposes a method based on Keyword-Aware Transformers Network (KAT) to fuse contextual keywords, enabling keyword semantic enhancement. Experimental results on two Chinese open-domain dialogue datasets show that our model outperforms existing methods in both semantic and non-semantic evaluation metrics, improving Coherence, Fluency, and Informativeness in manual evaluation.
Article
Substance Abuse
Mario Haim, Sebastian Scherr, Florian Arendt
Summary: The study found that Google displays counselling services at a high rate when users search suicide-related terms, but significantly lower when drug-related terms are involved, indicating a need for adjustment in search engine algorithms to increase display rates of prevention resources for drug-related suicides. Consequently, search engines should direct users to resources that can help in preventing drug-related suicides, especially during critical search moments.
DRUG AND ALCOHOL DEPENDENCE
(2021)
Article
Business
Huiran Li, Yanwu Yang
Summary: This paper addresses the keyword targeting problem in sponsored search advertising by proposing a data distribution estimation model and a stochastic keyword targeting model. Experimental results show the effectiveness of the proposed strategy in terms of profit and budget constraints. The proposed data distribution estimation approach effectively addresses the problem of incomplete information and uncertainty, making significant contributions to keyword targeting decisions.
ELECTRONIC COMMERCE RESEARCH AND APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Zeyu Zhao, Wei-Qiang Zhang
Summary: Keyword search (KWS) involves searching for keywords from continuous speech, and recent advancements in deep learning have allowed for end-to-end (E2E) training of KWS systems. The proposed E2E model outperforms baseline models and demonstrates effectiveness under low resource conditions.
Editorial Material
Multidisciplinary Sciences
Amanda Heidt
Summary: Developers aim to assist scientists in drawing connections from a vast amount of literature, enabling them to concentrate on discovery and innovation.
Article
Environmental Sciences
Cristina Santin, Aristides Moustakas, Stefan H. Doerr
Summary: Interactions between humans and wildfires have increased due to climate and land-use changes. It is important to understand public interest in wildfires as they play a crucial role in policy decisions. Using Google Trends, temporal patterns of public interest in wildfires were assessed globally and in five case study countries. Public interest shows cyclic patterns with spikes during fire seasons and catastrophic events, with wildfires in Western countries, particularly the USA, receiving the most interest. However, overall global interest in wildfires is low compared to earthquakes or hurricanes which are more economically costly. The seasonal interest in wildfires may hinder the implementation of year-round wildfire mitigation policies, but the internet can be utilized to educate the public about wildfires during interest spikes.
ENVIRONMENTAL SCIENCE & POLICY
(2023)
Article
Chemistry, Multidisciplinary
Zaira Hassan Amur, Yew Kwang Hooi, Gul Muhammad Soomro, Hina Bhanbhro, Said Karyem, Najamudin Sohu
Summary: Keyword extraction is a critical task for various applications, but the lack of a suitable dataset for semantic analysis remains a problem. To address this, a study was conducted to identify a dataset based on structure, complexity, and quality factors.
APPLIED SCIENCES-BASEL
(2023)