Hasil untuk "Information resources (General)"

Aleksandra Edwards, Antonio F Pardiñas, George Kirov et al.

Abstract BackgroundFree-text clinical data are unstructured and narrative in nature, providing a rich source of patient information, but extracting research-quality clinical phenotypes from these data remains a challenge. Manually reviewing and extracting clinical phenotypes from free-text patient notes is a time-consuming process and not suitable for large-scale datasets. On the other hand, automatically extracting clinical phenotypes can be challenging because medical researchers lack gold-standard annotated references and other purpose-built resources, including software. Recent large language models (LLMs) can understand natural language instructions, which help them adapt to different domains and tasks without the need for specific training data. This makes them suitable for clinical applications, though their use in this field is limited. ObjectiveWe aimed to develop an LLM pipeline based on the few-shot learning framework that could extract clinical information from free-text clinical summaries. We assessed the performance of this pipeline for classifying individuals with confirmed or suspected comorbid intellectual disability (ID) from clinical summaries of patients with severe mental illness and performed genetic validation of the results by testing whether individuals with LLM-defined ID carried more genetic variants known to confer risk of ID when compared with individuals without LLM-defined ID. MethodsWe developed novel approaches for performing classification, based on an intermediate information extraction (IE) step and human-in-the-loop techniques. We evaluated two models: Fine-Tuned Language Text-To-Text Transfer Transformer (Flan-T5) and Large Language Model Architecture (LLaMA). The dataset comprised 1144 free-text clinical summaries, of which 314 were manually annotated and used as a gold standard for evaluating automated methods. We also used published genetic data from 547 individuals to perform a genetic validation of the classification results; Firth’s penalized logistic regression framework was used to test whether individuals with LLM-defined ID carry significantly more de novo variants in known developmental disorder risk genes than individuals without LLM-defined ID. ResultsThe results demonstrate that a 2-stage approach, combining IE with manual validation, can effectively identify individuals with suspected IDs from free-text patient records, requiring only a single training example per classification label. The best-performing method based on the Flan-T5 model and incorporating the IE step achieved an F1P−5 ConclusionsLLMs and in-context learning techniques combined with human-in-the-loop approaches can be highly beneficial for extraction and categorization of information from free-text clinical data. In this proof-of-concept study, we show that LLMs can be used to identify individuals with a severe mental illness who also have suspected ID, which is a biologically and clinically meaningful subgroup of patients.

E. Kitch

Pierfrancesco Bellini, Daniele Bologna, Paolo Nesi et al.

Smart city/IoT frameworks are becoming more complex for the needs regarding multi-tenancy, data streams, real-time event-driven processing, data, and visual analytics. The infrastructures also need to support multiple organizations and optimizations in terms of data, processes/services, and tools cross-exploited by multiple applications and developers. In this paper, we addressed these needs to provide platform operators and developers effective models and tools to: (i) identify the causes of problems and dysfunctions at their inception; (ii) identify references to data, processes, and APIs to add/develop new scenarios in the infrastructure, minimizing effort; (iii) monitor resources and the work performed by developers to exploit the complex multi-application platform. To this end, we developed a semantic unified knowledge model, UKM, and a number of tools for its implementation and exploitation. The UKM, with its inferences, allows to browse and extract information from complex relationships among entities. The proposed solution has been designed, implemented, and validated in the context of the open source Snap4City.org platform and applied in many geographical areas with 18 organizations, 40 cities, thousands of operators and developers, and free trials to keep platform complexity under control, as in the interconnected scenarios of the Herit-Data Interreg Project, which is a lighthouse project of the European Commission.

Eungi Kim

Journals that have consistently maintained uninterrupted indexing over an extended period can be assumed to possess stability and sustainability in journal indexing. Building on this assumption, the objective of this study is to scrutinize the years omitted in the indexing of Scopus-indexed journals. To conduct this study, three coverage duration indicators—nyears-covered (total years covered), nyears-skipped (years skipped), and skipped/covered ratio (proportion of years skipped to total years covered)—were formulated. Data from SCImago Journal Rank (SJR) for 2022, consisting of 16,762 records (62% of downloaded data) with a coverage duration of 25 years or less, were used for this study. The results revealed that around 10% of Scopus-indexed journals experienced exclusions or coverage gaps. Longer coverage correlates positively with documents published, h-index, and citations, while skipped years decrease with these indicators. Open access (OA) journals exhibited a lower skipped/covered ratio than non-OA journals, suggesting a better sustainability of indexing than non-OA journals. Disciplinary differences in Scopus journal coverage duration revealed notable variation, suggesting that coverage duration indicators can be effectively used to evaluate journal stability within Scopus. Overall, the coverage gaps reflect Scopus’s efforts to regulate the journals it indexes. The coverage duration indicators proposed in this study can be applied to assess the stability of periodicals in any database, providing insights into the broader dynamics and quality standards maintained by a database, where the database periodically adds and removes its indexed contents.

Mehrvash Moradi, Zahra Karimian, Nahid Zarifsanaiey

Background: Medical students receive health information from various media sources and messages. Receiving health information can have a significant impact on the students’ dietary habits and healthy lifestyles. This study aimed to investigate the status of medical students’ use of health media and messaging, as well as the contextual variables that affect it.Methods: A survey study was conducted on 500 students of Shiraz University of Medical Sciences during September to December 2022 that were selected by simple random sampling. The research tool was a researcher-made questionnaire consisting of 15 items and six domains on a 4-point Likert scale, with a cutting point of 2.5. The validity of the tool was assessed using the CVI and CVR index, and its reliability was confirmed by internal consistency of 0.818. Data were collected through an e-questionnaire and analyzed using one-sample t-tests, independent t-tests, and ANOVA, using SPSS v.24 software.Results: About 40% of the participants were male. 36.4% were BS students, 27.8% MD, 20.5% MSc degree, and the remaining were in the PhD program. 73.3% of the participants were single, 43.8% lived with their parents, and 28.8% lived in the dormitory. Among the types of health media, Internet searches (M=2.950), verbal conversations (M=2.394), scientific resources (M=2.236), and official health information websites (M=2.128) had the highest means, while mass media like radio and television were the lowest. Contextual factors such as age, field of study, place of residence, marital status, and educational level had a significant effect on the students’ preference (P<0.05), but gender had no significant effect (P>0.05).Conclusion: Students have access to the Internet and receive most of their information through this medium. Monitoring the sources and content of health information websites, as well as strengthening health media literacy among students, are of great importance

Siosaia Vaihola, Stuart Kininmonth

Despite the crucial role played by international and regional tuna fisheries in facilitating the successful implementation of the ecosystem approach to fisheries management, there exist disparities in viewpoints among these stakeholders, resulting in gaps between regional fisheries management and local communities. Nevertheless, the Tongan government, under the Ministry of Fisheries, is dedicated to the efficient management of its tuna resources, aiming to establish it as the preferred and optimal approach for ensuring the long-term sustainability of its tuna fisheries and the ecosystem services they provide to the community. Recognizing that an appropriate legal, policy and institutional framework is in place for sustainable management of tuna, the first part of this paper presents a review of current Tonga fisheries laws and policies for its tuna fisheries. This review reflects the implementation of an information-based management framework, namely the Tonga National Tuna Fishery Management and Development Plan. The tuna fisheries in Tonga mainly catch albacore (<i>Thunnus alalunga</i>), bigeye (<i>Thunnus obesus</i>), skipjack (<i>Katsuwonus pelamis</i>), and yellowfin (<i>Thunnus albacares</i>) tuna. These tuna species are caught within Tonga’s exclusive economic zones and play a crucial role in the country’s economy; hence, it is crucial to examine the spatio-temporal distributions of their catch in relation to their environmental conditions. In pursuit of this goal, the tasks of mapping (i) the spatio-temporal distribution of catch landed at ports and (ii) the spatio-temporal of environmental conditions were performed. The study utilizes longline catch per unit effort data spanning from 2002 to 2018 for albacore, bigeye, skipjack, and yellowfin tuna. It also incorporates data on environmental conditions, including sea surface temperature, sea surface chlorophyll, sea surface current, and sea surface salinity. Additionally, the El Nino Southern Oscillation Index is mapped in relation to catch data to examine the potential effects of climate change on the tuna catch. Results show that bigeye, skipjack, and yellowfin CPUE show a central–northernmost distribution and are primarily caught between latitudes 14° S–22° S, while albacore, shows a central–southern distribution. The highest CPUE for all species are in latitudes 15.5° S–22.5° S and longitudes 172.5° W–176.5° W. The data indicate that sea surface current velocities range from −0.03 to 0.04 ms⁻¹, sea surface salinity ranges from 34.8 to 35.6 PSU, sea surface chlorophyll concentration varies from 0.03 to 0.1 mg m⁻³, and sea surface temperature fluctuates seasonally, ranging from 18 °C to 30 °C. Mapping also reveals that times of reduced catches in Tonga coincide with periods of moderate to strong El Nino events from 2002 to 2018.

Alana R. Westwood, Jenna Hutchen, Tyreen Kapoor et al.

Abstract Knowledge produced by scientists is essential to the policy and practice of managing natural resources, including forests. However, there has never been systematic mapping of which techniques in knowledge exchange (KE) have been applied in the forest sciences, by whom, and to what effect. We examined KE techniques documented in the forest sciences globally. We used standardized search strings in English and French across two academic search engines (BASE and Scopus) and a specialist website (ResearchGate) to locate relevant items. We screened items, extracted data, conducted qualitative and quantitative analysis, and built a network visualization diagram to demonstrate knowledge flow. Our final map included 122 items published from 1998 to 2020, with most published after 2010. Items mentioned organizations from 66 countries as knowledge producers or users. The interactive network visualization diagram displays linkages between organizations, sectors and countries. We found that most of the KE activity involved the Global North (89%). Governments were the most common knowledge users, and industry was frequently reported as a user but rarely a producer. Academia was both producer and user. Indigenous, local, traditional or community knowledge was included in 24% of items, but these communities were not associated with any coauthor affiliations. Reported funders were universities, governments, non‐profits or foundations. We found 90 unique terms in the items related to KE with less than 25% of terms used in more than one item. Fifteen per cent of item keywords related to KE. The most commonly identified enabling conditions for KE were trust, funding and established relationships, while major barriers were challenges for translation of science and lack of time. To improve searchability of information related to KE and encourage a culture of considering KE in scientific research and forest management work, we recommend a common lexicon of ‘knowledge exchange’/‘échange de connaisances’. We recommend that more effort be given to forest science‐related KE connections between the Global North and South as well as a deliberate collection of evidence for the effectiveness of KE techniques. Researchers and practitioners can use our KE typology to identify their goals and design appropriate evaluation measures.

Allison B. Ledford, Gregory Harris, Gregory Purdy

In the age of digitalization and the fourth industrial revolution, the concept of the digital thread has captured both the attention and resources of countless organizations. However, despite its growing popularity and a plethora of technologies being offered that promise to create the digital thread, this research revealed that no current techniques allow for the systematic uncovering of data and information flows in a way that reveals individual threads of data elements. Therefore, data element mapping and analysis (DEMA) is proposed as a tool capable of identifying all the data elements at the granularity necessary to connect digital threads. DEMA is a technology agnostic approach that allows an enterprise to move from a functional, document-centric view of data and information flows to a data element level view, with the data elements serving as the connectors of the digital thread.

Shu-Fan Hsieh, Chia-Ying Chan, Ming-Chun Wang

Abstract In this paper, we argue that when individual investors can obtain information from public resources such as Google search, the degree of investor attention to a particular underlying company is positively linked with herding behavior for retail investors. Empirical results confirm that Google Search Volume Index can be a proxy for the information demand of uninformed individual investors. Empirical evidence also shows that reaching the price limit generates an attention-grabbing effect. Further, in general, small cap firms generate more intensive individual investor herding. In addition, we explore the asymmetric impact of abnormal search volume index on individual investor herding behavior for bull and bear markets, and confirm that the individual investor buy herding phenomenon is stronger in bull markets, especially for small capitalization firms. In bear markets, with greater price deterioration for large cap firms, we detect herding behavior on the sell side.

Xiaojuan Yan, Hailin Xiao, Kang An et al.

Allowing multiple users access simultaneously, a power-domain nonorthogonal multiple access (NOMA) scheme can provide enhanced resource utilization efficiency and has been viewed as a promising strategy in the satellite networks, to provide high quality-of-service for large number of users within limited spectrum and time resources. Taking into account the random location information of served users, imperfect channel state information (CSI), and antenna-pointing error, we study the ergodic capacity of a NOMA-based uplink satellite network in this article. Specially, we first develop a general framework for the NOMA-based system by applying standard recommendations, configurations, and channel characteristics. Then, the analytical expression for the achievable ergodic capacity of the considered network is derived. Finally, numerical simulations are provided to attest the validity of theoretical results and show the impact of various key parameters, such as location information, link quality, transmission power, imperfect CSI, and antenna-pointing error, on the considered network.

Andrew R. A. Conway, R. Engle

Federico Brunero, P. Elia

This work identifies the fundamental limits of multi-access coded caching (MACC) where each user is connected to multiple caches in a manner that follows a generalized combinatorial topology. This topology stands out as it allows for unprecedented coding gains, even with very modest cache resources. First, we extend the setting and the scheme presented by Muralidhar et al. to a much more general topology that supports both a much denser range of users and the coexistence of users connected to different numbers of caches, all while maintaining the astounding coding gains — here proven to be exactly optimal — associated with the combinatorial topology. This is achieved, for this generalized topology, with a novel information-theoretic converse that we present here, which establishes, together with the scheme, the exact optimal performance under the assumption of uncoded placement. We subsequently consider different connectivity ensembles, including the very general scenario of the entire ensemble of all possible network connectivities/topologies, where any subset of caches can serve any arbitrary number of users. For these settings, we develop novel converse bounds on the optimal performance averaged over the ensemble’s different connectivities. This novel analysis of topological ensembles leaves open the possibility that currently-unknown topologies may yield even higher gains, a hypothesis that is part of the bigger question of which network topology yields the most caching gains.

Gabriel Khan, Jun Zhang

Information geometry and optimal transport are two distinct geometric frameworks for modeling families of probability measures. During the recent years, there has been a surge of research endeavors that cut across these two areas and explore their links and interactions. This paper is intended to provide an (incomplete) survey of these works, including entropy-regularized transport, divergence functions arising from $c$-duality, density manifolds and transport information geometry, the para-Kähler and Kähler geometries underlying optimal transport and the regularity theory for its solutions. Some outstanding questions that would be of interest to audience of both these two disciplines are posed. Our piece also serves as an introduction to the Special Issue on Optimal Transport of the journal Information Geometry.

Alvaro Sanchez, Usue Elizondo-Alzola, Jose I. Pijoan et al.

IntroductionA substantial proportion of individuals with low cardiovascular risk receive inappropriate statin prescription for primary prevention of cardiovascular disease (CVD) instead of the evidence-based recommendations to promote healthy lifestyle behaviors. This study reports on the structured process performed to design targeted de-implementation strategies to reduce inappropriate prescription of statins and to increase healthy lifestyle promotion in low cardiovascular risk patients in Primary Care (PC).MethodsA formative study was conducted based on the Theoretical Domains Framework and the Behavior Change Wheel (BCW). It comprised semi-structured interviews with PC professionals to define the problem in behavioral terms; focus groups with Family Physicians and patients to identify the determinants (barriers and facilitators) of inappropriate statin prescription and of healthy lifestyle promotion practice; mapping of behavioral change interventions operationalized as de-implementation strategies for addressing identified determinants; and consensus techniques for prioritization of strategies based on perceived effectiveness, feasibility and acceptability.ResultsIdentified key determinants of statin prescription and healthy lifestyle promotion were: the lack of time and clinical inertia, external resources, patients' preferences and characteristics, limitation of available clinical tools and guidelines, social pressures, fears about negative consequences of not treating, and lack of skills and training of professionals. Fourteen potential de-implementation strategies were mapped to the identified determinants and the following were prioritized: 1) non-reflective decision assistance strategies based on reminders and decision support tools for helping clinical decision-making; 2) decision information strategies based on the principles of knowledge dissemination (e.g., corporative diffusion of evidence-based Clinical Practice Guidelines and Pathways for CVD primary prevention); 3) reflective decision-making restructuring strategies (i.e., audit and feedback provided along with intention formation interventions).ConclusionsThis study supports the usefulness of the BCW to guide the design and development of de-implementation strategies targeting the determinants of clinicians' decision-making processes to favor the abandonment of low-value practices and the uptake of those recommended for CVD primary prevention in low-risk patients. Further research to evaluate the feasibility and effectiveness of selected strategies is warranted.Clinical trial registrationSanchez A. De-implementation of Low-value Pharmacological Prescriptions (De-imFAR). ClinicalTrials.gov, Identifier: NCT04022850. Registered July 17, 2019. In: ClinicalTrials.gov. Bethesda (MD): U.S. National Library of Medicine (NLM). Available from: https://www.clinicaltrials.gov/ct2/show/NCT04022850.

Aled Jones, Jill Maben, Mary Adams et al.

Background: The introduction of ‘Freedom to Speak Up Guardians’ into every NHS trust in England was intended to support workers and trusts to better raise, respond to and learn from speaking-up concerns. However, only broad guidance was provided on how to implement the role. As a result, there is the potential for important local differences to emerge as the role is implemented across England. Objectives: The overall aim of this study was to better understand the implementation of Guardians in acute trusts and mental health trusts. Design: The Freedom to Speak Up Guardian role was conceptualised as a complex intervention consisting of several interacting and interlocking components spanning the macro level (national organisations), the meso level (individual trusts) and the micro level (employees, teams and wards/units). A mixed-methods study was designed, which consisted of three work packages: (1) a systematic narrative review of the international literature regarding interventions promoting ‘speaking up’ by health-care employees; (2) semistructured telephone interviews with Guardians working in acute hospital trusts and mental health trusts; and (3) qualitative case studies of Freedom to Speak Up Guardian implementation, consisting of observations and interviews undertaken in four acute trusts and two mental health trusts. Interviews were also undertaken with national stakeholders. Setting: Acute trusts and mental health NHS trusts in England. Participants: Work package 2: Freedom to Speak Up Guardians (n = 87) were interviewed. Work package 3: 116 interviews with key stakeholders involved in pre-implementation and early implementation decision-making, workers who had spoken up to the Guardian, and national stakeholders. Results: Wide variability was identified in how the Guardian role had been implemented, resourced and deployed by NHS trusts. ‘Freedom to Speak Up Guardian’ is best considered an umbrella term, and multiple versions of the role exist simultaneously across England. Any comparisons of Guardians’ effectiveness are likely to be possible or meaningful only when this variability is properly accounted for. Many Freedom to Speak Up Guardians identified how a lack of available resources, especially time scarcity, negatively and significantly affected their ability to effectively respond to concerns; their opportunities to collect, analyse and learn from speaking-up data; and, more generally, the extent to which they developed their role and speak-up culture. Limitations: It is possible that those whom we interviewed were more receptive of Freedom to Speak Up Guardians or may have been biased by ‘socially desirability’, and their answers may not always have represented respondents’ true perceptions. Conclusions: Optimal implementation of the Guardian role has five components: (1) establishing an early, collaborative and coherent strategy congruent with the values of Freedom to Speak Up fosters the implementation of (2) policies and robust, yet supportive, practices (3) informed by frequent and reflexive monitoring of Freedom to Speak Up implementation that is (4) underpinned by sufficient time and resource allocation that leads to (5) a positive implementation climate that is congruent with Freedom to Speak Up values and is well placed to engender positive and sustainable Freedom to Speak Up culture and the well-being of a Guardian. Future work: The following recommendations for future research are considered to be of equal priority. Studies of the speaking-up experiences of minority communities and ‘seldom-heard’ workforce groups are a priority requirement. There is also value in undertaking a similar study in non-hospital settings and where peripatetic working is commonplace, such as in ambulance services and in primary care settings. The role of human resources and ‘middle managers’ in the management of concerns is an area requiring further research, especially regarding concerns relating to unprofessional and transgressive behaviours. Devolved administrations in Scotland and Wales have adopted different approaches to speaking up; research undertaken in these contexts would offer valuable comparative insights. Researching the Guardian role ≥ 5 years post implementation is recommended to understand the medium-term impact and the longer-term sustainability of the role and well-being of Guardians. Study registration: This study is registered as ISRCTN38163690 and has the study registration CRD42018106311. Funding: This project was funded by the National Institute for Health and Care Research (NIHR) Health and Social Care Delivery Research programme and will be published in full in Health and Social Care Delivery Research; Vol. 10, No. 23. See the NIHR Journals Library website for further project information.

Wasim Ahmed, P. Bath, L. Sbaffi et al.

BACKGROUND Infectious disease outbreaks have the potential to cause a high number of fatalities and are a very serious public health risk. OBJECTIVES Our aim was to utilise an indepth method to study a period of time where the H1N1 Pandemic of 2009 was at its peak. METHODS A data set of n = 214 784 tweets was retrieved and filtered, and the method of thematic analysis was used to analyse the data. RESULTS Eight key themes emerged from the analysis of data: emotion and feeling, health related information, general commentary and resources, media and health organisations, politics, country of origin, food, and humour and/or sarcasm. DISCUSSION A major novel finding was that due to the name 'swine flu', Twitter users had the belief that pigs and pork could host and/or transmit the virus. Our paper also considered the methodological implications for the wider field of library and information science as well as specific implications for health information and library workers. CONCLUSIONS Novel insights were derived on how users communicate about disease outbreaks on social media platforms. Our study also provides an innovative methodological contribution because it was found that by utilising an indepth method it was possible to extract greater insight into user communication.

Michael Timothy Bennett, Yoshihiro Maruyama

We attempt to define what is necessary to construct an Artificial Scientist, explore and evaluate several approaches to artificial general intelligence (AGI) which may facilitate this, conclude that a unified or hybrid approach is necessary and explore two theories that satisfy this requirement to some degree.

Domonic A. Bearfield, A. Bowman

Jonathan Scarlett, Albert Guillén i Fàbregas, Anelia Somekh-Baruch et al.

Shannon's channel coding theorem characterizes the maximal rate of information that can be reliably transmitted over a communication channel when optimal encoding and decoding strategies are used. In many scenarios, however, practical considerations such as channel uncertainty and implementation constraints rule out the use of an optimal decoder. The mismatched decoding problem addresses such scenarios by considering the case that the decoder cannot be optimized, but is instead fixed as part of the problem statement. This problem is not only of direct interest in its own right, but also has close connections with other long-standing theoretical problems in information theory. In this monograph, we survey both classical literature and recent developments on the mismatched decoding problem, with an emphasis on achievable random-coding rates for memoryless channels. We present two widely-considered achievable rates known as the generalized mutual information (GMI) and the LM rate, and overview their derivations and properties. In addition, we survey several improved rates via multi-user coding techniques, as well as recent developments and challenges in establishing upper bounds on the mismatch capacity, and an analogous mismatched encoding problem in rate-distortion theory. Throughout the monograph, we highlight a variety of applications and connections with other prominent information theory problems.

Sunday C. Eze, Vera C. A. Chinedu-Eze, Clinton K. Okike et al.

Abstract This paper examines critical success factors shaping the adoption of digital marketing devices (DMD) by micro-businesses. The study adopted a qualitative approach. Both unstructured and semi-structured interviews were conducted with a total of 26 micro-businesses drawn purposefully from the online database and underpinned by technology, organisation and environment (TOE) framework. This helped to unravel 14 critical success factors influencing the adoption of DMD by micro-businesses in Nigeria. These factors include functional capability, adaptive capacity and expandability which are related to technology context. Collective understanding, degree of partnership and diversity of information are linked to the organisation context while the level of training, quality of service delivery, customer fulfilment and intense competition are associated with environmental context. The study expanded the TOE framework by unravelling the impact expectancy context which is associated with factors, such as budget, business expansion, diversity and return on investment. This study will be of importance to academics and practitioners because it provides further awareness into DMD adoption framework, factors critical to the DMD adoption and may assist in reducing the number of resources spent in search of information aimed at helping DMD adoption by micro-businesses.