ODA Open Digital Archive (Oslomet)
Not a member yet
19980 research outputs found
Sort by
Remote Work and the Psychosocial Work Environment: The Role of Contact
Full text linkRemote work is fundamentally changing the organization of work, replacing the physical work environment of the office and creating a spatial distance between colleagues. This article inves-tigates the relationship between remote work frequency and three aspects of the psychosocial work environment (quantitative job demands, control, and social support) and to which extent the relationship is mediated by the frequency of contact with supervisors and co-workers. The Norlife Remote Work - Longitudinal Study (NorRemo-LS) of Norwegian employees (n = 2553) across four waves between February 2021 and September 2022 is analyzed using general structural equation modeling (GSEM).The results show that the extent of remote work is associated with increased schedule control. Remote work may diminish employee control and increase quantitative job demands if employees have less frequent contact with their supervisor. Less frequent contact with co-workers not only entails lower job control and less social support but also lower quantitative demands.publishedVersio
Group Recommendation Systems With Pairwise Preference Data
Full text linkBackground and Motivation: Group recommendation systems (GRS) are designed to find what a group of people likes and suggest things they will enjoy together. These systems aim to match the combined tastes of everyone in the group. GRSs are needed for social activities like watching movies, dining out, and planning trips, where decisions must please people with different preferences. The main challenge is to combine these different tastes into one recommendation that makes everyone happy. This research focuses on understanding how groups make decisions and creating algorithms that can accurately predict what a group will enjoy. Successfully solving this challenge can make group activities more enjoyable and harmonious.
Objectives: The main objective of this research is to propose new methods for group recommendation that are fair and precise. To accomplish this goal, six research questions (RQs) have been formulated. [RQ1] How does the utilization of pairwise preference data address the limitations of single-rating data in enhancing the effectiveness of group recommendation systems? [RQ2] How does clustering users based on similar preferences contribute to enhancing the fairness of recommendation systems? [RQ3] How can the prediction of missing data in pairwise preference datasets effectively address the cold start issue in GDM and GRS? [RQ4] How can we develop models to better understand and incorporate the influences among members’ preferences, thus enhancing group recommendation? [RQ5] How can leveraging diverse similarity features of users overcome the limitations of traditional group recommendation systems to enhance recommendation accuracy? [RQ6] How can aggregation and consensus-reaching mechanisms enhance group recommendation systems?
Methods: This research proposed and employed methodologies that can be classified into three primary categories: First, we utilized pairwise preference data, and predicted missing values. Second, we explored user grouping through the introduction of clustering techniques such as GcPp, MFP-based diversity clustering, and GCN-based diversity clustering. In this context, we examined various user similarity score calculations, some of which were proposed for the first time. Third, we developed consensus-reaching or aggregation methods that combine individual user preferences to form a cohesive group preference profile, which is essential for constructing the group recommendation model.
Contributions: The main contributions of this study include:
• Introducing an entropy-based matrix factorization technique for predicting missing values in pairwise preference datasets, which has broad applications in group recommendation systems and group decision-making.
• Proposing several methods for predicting user similarity scores using pairwise preference data, demonstrating higher accuracy compared to single rating data. These similarity scores were calculated using various methods:
1. User similarity scores based on preference graph and graph convolutional networks (GCN).
2. User similarity scores based on user vectors derived from: a) User-item scores obtained from matrix factorization (MF). b) User embedding vectors from a trained matrix factorization model. c) User embedding vectors from the weights of a trained graph neural network.
• Developing clustering methods for grouping the users with similar preferences to facilitate generating fair group recommendations, such as:
1. Dominant set clustering.
2. Diversity-based clustering, which minimizes user diversity scores within
groups.
• Introducing a consensus-reaching method based on user personalities, reflecting real-life scenarios where user contributions to group decisions depend on their personality traits.
• Developing aggregation methods that account for the contributions of individual users in the final group decision. These contributions are calculated using concepts such as the Shapley value and Wonderful Life Utility.
Bakgrunn og Motivasjon: Gruppeanbefalingssystemer (GRS) er utviklet for å finne ut hva en gruppe mennesker liker og til å foreslå ting de vil ha glede av sammen. Disse systemene har som mål å matche de kombinerte preferansene til alle i gruppen. GRS er nyttig for sosiale aktiviteter som å se på film, spise ute og planlegge turer, hvor beslutninger gi et best mulig resultat for flere med ulike preferanser. Den største utfordringen er å kombinere de ulike preferansene til én anbefaling som gjør alle fornøyde. Denne forskningen fokuserer på å forstå hvordan grupper tar beslutninger og å utvikle algoritmer som nøyaktig kan forutsi hva en gruppe vil sette pris på. Å lykkes med å løse denne utfordringen kan gjøre gruppeaktiviteter mer hyggelige og harmoniske.
Målsetninger: Hovedmålet med denne forskningen er å foreslå nye metoder for gruppeanbefalinger som er rettferdige og presise. For å oppnå dette målet er seks forskningsspørsmål (RQs) formulert: [RQ1] Hvordan kan bruk av parvise preferansedata løse begrensningene ved enkeltratingsdata for å øke effektiviteten til GRS? [RQ2] Hvordan bidrar klynging av brukere basert på lignende preferanser til å forbedre rettferdigheten i anbefalingssystemer? [RQ3] Hvordan kan prediksjon av manglende data i parvise preferansedatasett effektivt håndtere kaldstartproblemet i GDM og GRS? [RQ4] Hvordan kan vi utvikle modeller for å bedre forstå og inkorporere påvirkningene blant medlemmers preferanser, og dermed forbedre gruppeanbefalinger? [RQ5] Hvordan kan utnyttelse av mangfoldige likhetsegenskaper hos brukere overkomme begrensningene til tradisjonelle gruppeanbefalingssystemer for å forbedre anbefalingsnøyaktigheten? [RQ6] Hvordan kan aggregasjons- og konsensusmekanismer forbedre gruppeanbefalingssystemer?
Metoder: Denne forskningen foreslo og anvendte metodologier som kan klassifiseres i tre hovedkategorier: Først benyttet vi parvise preferansedata og forutså manglende verdier. For det andre utforsket vi brukergruppering gjennom introduksjon av klyngingsteknikker som GcPp, MFP-basert mangfoldsklynging og GCN-basert mangfoldsklynging. I denne sammenhengen undersøkte vi forskjellige metoder for å beregne brukersimilaritet, hvorav noen ble foreslått for første gang. For det tredje utviklet vi konsensus- eller aggregeringsmetoder som kombinerer individuelle brukerpreferanser for å danne en helhetlig gruppepreferanseprofil, som er essensielt for å
konstruere gruppeanbefalingsmodellen.
Bidrag: Hovedbidragene fra denne studien inkluderer:
• Introduksjon av en entropibasert matrisefaktoriseringsteknikk for å forutsi manglende verdier i parvise preferansedatasett, som har bred anvendelse i gruppeanbefalingssystemer og gruppedynamikk.
• Forslag til flere metoder for å forutsi brukersimilaritet ved bruk av parvise preferansedata, som viser høyere nøyaktighet sammenlignet med enkeltratingsdata. Disse similaritetsscorene ble beregnet ved hjelp av ulike metoder:
1. Brukersimilaritet basert på preferansegraf og grafkonvolusjonsnettverk (GCN).
2. Brukersimilaritet basert på brukervektorer hentet fra: a) Bruker-itemscores oppnådd fra matrisefaktorisering (MF). b) Brukerinnleiringsvektorer fra en trent matrisefaktoreringsmodell. c) Brukerinnleiringsvektorer fra vektene
til et trent grafnevralnettverk.
• Utvikling av klyngemetoder for å gruppere brukere med lignende preferanser for å fasilitere generering av rettferdige gruppeanbefalinger, som:
1. Dominant set klynging.
2. Mangfoldsbasert klynging, som minimerer brukermangfoldsscorer innenfor grupper.
• Introduksjon av en konsensusmetode basert på brukerpersonligheter, som reflekterer virkelige scenarioer hvor brukerens bidrag til gruppebeslutninger avhenger av deres personlighetstrekk.
• Utvikling av aggregeringsmetoder som tar hensyn til bidragene fra individuelle brukere i den endelige gruppedynamikken. Disse bidragene beregnes ved hjelp av konsepter som Shapley-verdi og Wonderful Life Utility.publishedVersio
Social and Emotional Impacts of Airplane Headache: Insights into Patient Expectations and Management Needs
Full text linkA headache attributed to airplane travel is a headache that occurs during landing and/or take-of, also called “airplane headache” (AH). The emotional and social impacts on AH remain uncertain today. The purpose of this study was to assess the social and emotional impacts of airplane headache (AH) from the perspective of affected individuals and to explore their treatment expectations and needs. Specifically, the study examined whether stress, anxiety, and fatigue are significant factors influencing AH patients’ decision to seek medical treatment. An online anonymous survey was conducted in Denmark to investigate the emotional and social effects of AH. According to IHS diagnostic criteria for AH, participants self-identified as AH patients and completed a 36-question survey. Through social media and institutional channels over a 62-day period from March to May 2024, the questionnaire was developed using prior research and pilot-tested for clarity. The study included 33 patients with AH who reported increased pain intensity and significant social and emotional effects. Over-the-counter medications were ineffective, and participants expressed a need for more effective treatments that would have minimal side effects. A strong public awareness and reliable information about AH was needed, as it affected daily activities during fight. The development of treatments and preventative measures for AH should involve patients. In order to implement a holistic approach, both the social and emotional aspects must be considered, along with the need for preventive treatments and better awareness programs.publishedVersio
Exploring Internet-Delivered Treatment for Depression
Full text linkInternet-delivered treatment for patients with depressive disorders is evaluated as an effective and flexible treatment approach. Worldwide, the number of individuals with depressive disorders is increasing, and there is a considerable gap between the number of people in need of treatment and available treatment. Delivering treatments over the internet is proposed as a potential solution to bridge this treatment gap. These treatments have been reported to have both benefits and challenges. This thesis aimed to investigate therapist-guided internet-delivered treatments for depression, focusing on essential aspects to consider when studying, providing, and developing these interventions to complement studies of treatment effects. Three research questions were posed: 1) How have therapist-guided internet-delivered treatments for depression been studied, and are there any gaps in the research? 2) How do therapists experience providing therapist-guided internet-delivered treatment for patients with depression? 3) How do participants with personal experience of depression view a new internet-delivered acceptance and commitment therapy program for depression? Three studies were conducted to answer these questions. Study I is a scoping review, where 111 studies on guided internet-delivered treatment for depression were identified. Several research gaps were uncovered regarding how the studies were designed, the treatment approaches tested, the representation of participants, and how treatment completion was defined and reported. In Study II, therapists with clinical experience of the treatment program “eCoping” were interviewed about providing this treatment. The therapists described that predicting which patients would benefit from the program was complex. The program was experienced as putting high demands on patients, therapists, and clinics. Furthermore, it was emphasized that the program would not be a good fit for every patient, and the lack of possibilities to tailor the treatment was described as a challenge. In Study III, participants with personal experience of depression were interviewed about their views of an internet-delivered acceptance and commitment therapy program. The impact depressive symptoms could have on adherence to the treatment was emphasized. Combining the program with other treatments was viewed as an acceptable and good treatment option, but the program was viewed as insufficient as a standalone program. The participants also expressed the need for personalized treatment and raised concerns about the program’s standardized format. Altogether, the work in this thesis provides a comprehensive insight into vital aspects to consider when studying, providing, and developing internet-delivered interventions. This could be a valuable contribution to a research field aiming to address an increasing treatment gap.
Internett-levert behandling har vist seg å kunne være en effektiv og fleksibel tilnærming til depresjonbehandling. På verdensbasis er det en økning i antall mennesker som rammes av depresjon, og tilgjengelig behandlingstilbud dekker ikke behovet. Å levere behandling over internett er beskrevet som en mulig måte å gjøre behandling tilgjengelig for flere. Formålet med denne avhandlingen var å se nærmere på terapeut-veiledede internettleverte behandlinger for pasienter med depresjon, med særlig søkelys på hva som er viktig når disse behandlingene forskes på, brukes og utvikles. Målet var å avdekke kunnskap som kan komplementere funn fra effektstudier. Det ble laget tre forskningsspørsmål: 1) Hvordan har terapeut-veiledede internett-leverte behandlingsprogram for depresjon blitt forsket på, og er det noen kunnskapshull i forskningen som kan avdekkes? 2) Hvordan opplever terapeuter å levere terapeut-veiledet internett-levert depresjonbehandling? 3) Hvordan vurderer deltakere med personlig erfaring med depresjon et nytt internett-levert depresjonsprogram basert på aksept- og forpliktelsesterapi? Tre studier ble designet for å besvare disse spørsmålene. Studie I er en scoping review som inkluderer 111 studier om veiledet internett-levert behandling for depresjon. Flere kunnskapshull ble avdekket i forbindelse med hvordan studiene var designet, behandlingstilnærmingene som ble undersøkt, hvilke deltakere som var representert, og hvordan fullføring av behandlingsprogrammene var definert og rapportert. I Studie II ble terapeuter intervjuet om deres kliniske erfaring med behandlingsprogrammet «eMestring». Terapeutene beskrev at programmet ikke vil passe for alle, og at det kunne være utfordrende å forutsi hvilke pasienter som ville ha nytte av det. Videre beskrev de at programmet stiller store krav til pasientene, terapeutene og klinikkene som tilbyr behandlingen. Terapeutene beskrev også utfordringer med brukervennlighet og at de opplevde at det var for få muligheter til å tilpasse programmet til pasientenes individuelle behov. I Studie III ble personer med personlig erfaring med depresjon intervjuet om et internett-levert behandlingsprogram basert på aksept og forpliktelsesterapi. Deltakerne var generelt positive til programmet, men beskrev at depresjonssymptomer kunne gjøre det vanskelig å gjennomføre behandlingen. Det ble foreslått at programmet kunne brukes som supplement til andre helsetjenester, men ikke erstatte dem. Videre ble det vektlagt at programmet ikke vil passe for alle og begrensede muligheter for individuell tilpasning ble også her nevnt som en utfordring. Samlet sett bidrar denne avhandlingen med innsikt i viktige aspekter som bør vektlegges når internett-leverte behandlingsprogram forskes på, brukes og utvikles, og vil være et nyttig bidrag i utviklingen av adekvate tilbud til en stor gruppe mennesker i behov av depresjonbehandling.publishedVersio
The Norwegian Parliamentary Debates Dataset
Full text linkRecent advancements in computing power and machine learning techniques have facilitated the digitization of new corpora, as well as new methods for studying high-dimensional data. This has enabled empirical investigations of fundamental questions in the social sciences that were previously restricted by technical limitations or data availability. In this note, we introduce a new dataset covering debates in the Norwegian Parliament in the 1945-2024 period. This dataset, which covers close to one million speeches, includes information about speeches (full text, date of speech, and chamber), speakers’ status (parliamentary president, member of parliament, deputy member of parliament, or cabinet minister), as well as speaker background characteristics (party affiliation, committee membership, district affiliation, rank on electoral lists, gender, and birth year). This dataset will enable extensive research into political representation in a party-centered electoral framework. More broadly, this dataset serves as a vital resource for interdisciplinary research, enabling studies on the evolution of language, rhetoric, and the broader socio-economic factors influencing legislative behavior.publishedVersio
Trygg legemiddelhåndtering i akuttmottak - Et kunnskapsbasert undervisningsprogram for nyansatte sykepleiere om å forebygge legemiddelhåndteringsfeil.
Full text linkSAMMENDRAG
Prosjektets bakgrunn:
Akutt og kritisk syke pasienter i akuttmottak har ofte behov for legemiddelbehandling. Legemiddelhåndtering er en stor del av akuttsykepleierens funksjon og ansvarsområder i akuttmottaket. Legemiddelrelaterte feil er den hyppigste årsaken til pasientskader både nasjonalt og globalt. Akuttmottakets kompleksitet øker risikoen for legemiddelhåndteringsfeil. Økt kompetanse og andre forebyggende tiltak kan redusere forekomsten.
Hensikt:
Hensikten med masteroppgaven er å utvikle et kunnskapsbasert undervisningsprogram for å øke nyansatte sykepleieres kompetanse innen legemiddelhåndtering i akuttmottak. Dette for å forebygge feil og pasientskader, og å bedre pasientsikkerheten.
Problemstilling:
Å utvikle et kunnskapsbasert undervisningsprogram for nyansatte sykepleiere i akuttmottak om trygg legemiddelhåndtering og forebygging av legemiddelhåndteringsfeil.
Metode:
Masteroppgaven er et undervisningsprogram som del av et kvalitetsforbedringsarbeid, utarbeidet ved bruk av modell for kvalitetsforbedring og den didaktiske relasjonsmodellen. Arbeidets kunnskapsgrunnlag består av forsknings-, erfarings og pasientkunnskap innhentet gjennom systematiske forskningssøk med utgangspunkt i kunnskapspyramiden. En pilotundervisning ble gjenomført og evaluert med evalueringsskjema.
Resultater:
Masteroppgaven resulterte i et kunnskapsbasert undervisningsprogram om trygg legemiddelhåndtering i akuttmottak. Undervisningsprogrammet fokuserer på risikofaktorer for legemiddelhåndteringsfeil i akuttmottak og forebyggende tiltak. Evalueringen av pilotundervisningen viser at sykepleierne opplevde undervisningen relevant og faglig tilpasset akuttmottakets kliniske praksis. Deltakerne mente programmet var relevant for nyansatte sykepleiere, men også for mer erfarne sykepleiere i akuttmottak.
Konklusjon:
Undervisningsprogrammet kan bidra til økt kompetanse innen trygg legemiddelhåndtering i akuttmottak. Kompetanse og andre tiltak er viktige for å forebygge legemiddelhåndteringsfeil og kan bidra til å øke pasientsikkerheten.
Nøkkelord: Akuttmottak, legemiddelhåndtering, pasientsikkerhet, undervisningsprogram, legemiddelhåndteringsfeil, risikofaktorer, forebyggende tiltak
Referansebudsjettet 2025
No full textDenne filen gir et sammendrag av referansebudsjettets tall for 2025, og viser et utregnet budsjett for en eksempelfamilie. Du kan også laste ned budsjettet i Excel. Lenker til filene under bildet.This publication is a summary of the calculations for individual-specific and household-specific expences in the 2025 budget and presents an example budget for a family with two adults and two children. Here's also an Excel-version of the budget
Promoting healthy eating in kindergartens - Understanding the factors influencing food provision and implementing meal practices in alignment with national guidelines
Full text linkThere is considerable variability in food and meal practices across Norwegian kindergartens. Given the potential of kindergartens as a setting for health promotion and their potential to encourage healthy eating, this thesis seeks to understand the factors that influence the provision of food and the implementation of meal practices in alignment with national guidelines. Understanding these factors is important in promoting health and nutrition in kindergartens. This thesis consists of three papers. The first paper investigates whether the food budget based on the monthly food budget per child (additional fee paid by parents) is associated with food quality (food provision in alignment with national guidelines). The second paper examines whether there is any association between the knowledge of the content of national guidelines among kindergarten staff, their application of these guidelines, and food quality. The third paper explores the perceived challenges kindergarten staff face in implementing national guidelines. The first two papers are cross-sectional studies based on responses from 324 kindergarten staff collected through a web-based questionnaire measuring kindergartens’ food and meal practices. The statistical analyses in the first paper were performed using logistic regression. The second paper employs multiple linear regression. The third paper uses qualitative data collected through semi-structured interviews with kindergarten leaders and pedagogical staff. Total of 16 participants, spread across eight private and eight public kindergartens, participated in the study. The data analysis followed a content analysis approach based on Graneheim and Lundman's methodology. The results from the first paper revealed no association between an increased food budget and food quality once the per-child, per-month food budget was over 199 NOK. These findings suggest that factors other than the food budget could influence food and meal practices in kindergartens. The second paper revealed a significant association between the knowledge and application of the national guidelines by the kindergarten staff and food quality. We also found an association between knowledge of the content of national guidelines and practical application. In the third paper, priority and structural work with food and meals, pedagogical approaches to food and meals and external expectations and social pressure regarding food and meals were identified as the main themes influencing kindergartens’ food and meal practices. Key insights from our research highlight the substantial role of individual factors among staff in implementing food and meal practices in alignment with national guidelines. These individual factors include staff members’ preferences, interests and personal opinions regarding the importance of food and meals. Notably, the prioritisation of the kindergarten’s food and meals set by leaders also emerges as a significant influence. The present thesis found that the food budget is probably not the most significant factor in food quality. Instead, knowledge about the content of the national guidelines was found to be a substantial factor. Today’s practices seem to be highly influenced by individual factors (prioritisation, personal interests), and one explanation for that could be a lack of prioritising and structure from the municipality. This thesis provides valuable insights into the factors influencing food and meal practices in Norwegian kindergartens. The findings could contribute to developing future strategies to improve current practices, which is highly important from a health promotion perspective, and reduce social inequalities.
Det er betydelig variasjon i mat- og måltidspraksis på tvers av norske barnehager. Gitt barnehagens potensial som en arena for helsefremmende arbeid og deres mulighet til å fremme sunne spisevaner, søker denne avhandlingen å forstå de faktorene som påvirker matservering og implementering av måltidspraksis i tråd med nasjonale retningslinjer. Å forstå disse faktorene er viktig for å fremme helse og ernæring i barnehager. Denne avhandlingen består av tre artikler. Den første artikkelen undersøker om matbudsjettet basert på det månedlige matbudsjettet per barn (tilleggsavgift betalt av foreldre) er assosiert med matkvalitet (matservering i tråd med nasjonale retningslinjer). Den andre artikkelen utforsker om det er noen sammenheng mellom barnehagepersonalets kjennskap til innholdet i nasjonale retningslinjer, deres anvendelse av disse retningslinjene, og matkvalitet. Den tredje artikkelen forsøker å forstå de opplevde utfordringene som barnehagepersonalet møter ved implementering av nasjonale retningslinjer. De to første artiklene er tverrsnittsstudier basert på svar fra 324 barnehageansatte samlet inn gjennom et nettbasert spørreskjema som måler barnehagers mat- og måltidspraksis. De statistiske analysene i den første artikkelen ble utført ved bruk av logistisk regresjon. Den andre artikkelen benytter multippel lineær regresjon. Den tredje artikkelen benytter kvalitative data samlet inn gjennom semi-strukturerte intervjuer med barnehageledere og pedagogisk personale. Totalt deltok 16 deltakere, fordelt på åtte private og åtte offentlige barnehager. Dataanalysen fulgte en innholdsanalytisk tilnærming basert på Graneheim og Lundmans metodikk. Resultatene fra den første artikkelen viste ingen sammenheng mellom et økt matbudsjett og matkvalitet når det månedlige matbudsjettet per barn var over 199 NOK. Disse funnene antyder at andre faktorer enn matbudsjettet kunne påvirke mat- og måltidspraksis i barnehager. Den andre artikkelen avdekket en signifikant sammenheng mellom kjennskap til og anvendelse av nasjonale retningslinjer blant barnehagepersonalet og matkvalitet. Vi fant også en sammenheng mellom kjennskap til innholdet i nasjonale retningslinjer og praktisk anvendelse. I artikkel 3 ble prioritering og strukturelle tilnærminger til mat, pedagogiske tilnærminger, eksterne forventninger og sosialt press, identifisert som de tre hovedtemaene Sentrale innsikter fra vår forskning fremhever imidlertid den betydelige rollen individuelle faktorer blant personalet spiller i implementeringen av mat- og måltidspraksis i tråd med nasjonale retningslinjer. Disse individuelle faktorene inkluderer personalets preferanser, interesser og personlige meninger om viktigheten av mat og måltider. Betydelig er det også at den prioriteringen som settes av barnehagens ledere for mat og måltider framkommer som en vesentlig innflytelse. Denne avhandlingen fant at matbudsjettet sannsynligvis ikke er den mest betydningsfulle faktoren for matkvalitet. I stedet ble kjennskap til innholdet i de nasjonale retningslinjene funnet å være en vesentlig faktor. Dagens praksis ser ut til å være sterkt påvirket av individuelle faktorer (prioritering, personlige interesser), og en forklaring på det kan være mangel på prioritering og struktur fra kommunen. Denne avhandlingen gir verdifull innsikt i faktorer som påvirker mat- og måltidspraksis i norske barnehager. Funnene kan bidra til utvikling av fremtidige strategier for å forbedre dagens praksis, noe som er viktig i et helsefremmende perspektiv og for å utjevne sosiale ulikheter.publishedVersio
Health conspiracy theories: a scoping review of drivers, impacts, and countermeasures
Full text linkBackground Health-related conspiracy theories undermine trust in healthcare, exacerbate health inequities, and contribute to harmful health behaviors such as vaccine hesitancy and reliance on unproven treatments. These theories disproportionately impact marginalized populations, further widening health disparities. Their rapid spread, amplified by social media algorithms and digital misinformation networks, exacerbates public health challenges, highlighting the urgency of understanding their prevalence, key drivers, and mitigation strategies.
Methods This scoping review synthesizes research on health-related conspiracy theories, focusing on their prevalence, impacts on health behaviors and outcomes, contributing factors, and counter-measures. Using Arksey and O’Malley’s framework and the Joanna Briggs Institute guidelines, a systematic search of six databases (PubMed, Embase, Web of Science, CINAHL, PsycINFO, and Scopus) was conducted. Studies were screened using predefined inclusion and exclusion criteria, with thematic synthesis categorizing findings across diverse health contexts.
Results The review revealed pervasive conspiracy beliefs surrounding HIV/AIDS, vaccines, pharmaceutical companies, and COVID-19, linked to reduced vaccine uptake, increased mistrust in health authorities, and negative mental health outcomes such as anxiety and depression. Key drivers included sociopolitical distrust, cognitive biases, low scientific literacy, and the unchecked proliferation of misinformation on digital platforms. Promising countermeasures included inoculation messaging, media literacy interventions, and two-sided refutational techniques. However, their long-term effectiveness remains uncertain, as few studies assess their sustained impact across diverse sociopolitical contexts.
Conclusion Health-related conspiracy theories present a growing public health challenge that undermines global health equity. While several interventions show potential, further research is needed to evaluate their effectiveness across diverse populations and contexts. Targeted efforts to rebuild trust in healthcare systems and strengthen critical health literacy are essential to mitigate the harmful effects of these conspiracy beliefs.publishedVersio
Autonomous Robot Navigation Using Deep Reinforcement Learning in Office Environments
Full text linkAutonomous mobile robots are increasingly being tasked with performing complex logistics operations in environments characterized by unpredictable layouts, dynamic obstacles, and limited infrastructure. In such settings—typified by modern warehouses and distribution centers—robots must navigate reliably and efficiently without relying on premapped routes, fixed waypoints, or extensive human oversight. This thesis explores whether deep reinforcement learning (RL) can serve as a fully standalone navigation paradigm, replacing both global planners and hand-crafted heuristics, by enabling an omnidirectional robot to discover robust navigation strategies solely through sensorimotor interaction.
We formulate the navigation problem as a Markov Decision Process in which the robot’s observations consist of 2D LiDAR sector readings, odometry-derived pose and velocity estimates, and the relative direction to its goal. Actions are continuous linear and angular velocity commands, permitting fine-grained control essential for maneuvering through narrow corridors and around obstacles. A dense, composite reward function is designed to simultaneously encourage progress toward intermediate subgoals and the final destination, penalize collisions and unsafe proximities, discourage inefficient motions (e.g., reversing or excessive turning), and grant substantial bonuses upon subgoal or final goal attainment. No external maps, cost-maps, or path planners inform the learning process—every aspect of navigation strategy emerges from this reward-driven exploration.
Two state-of-the-art continuous-control RL algorithms are implemented and rigorously compared. Proximal Policy Optimization (PPO) offers on-policy stability through a clipped surrogate objective and entropy regularization, while Deep Deterministic Policy Gradient (DDPG) employs an off-policy replay buffer augmented by an elite-sampling mechanism and Ornstein–Uhlenbeck exploration noise for sample-efficient learning. Both agents are trained in a custom Gazebo–ROS2 simulation that mimics warehouse-like environments with randomized obstacle placements for each episode. The goal position remained fixed within each training run; for any new target location, the agent was re-initialized and trained anew.
Experimental results demonstrate that DDPG rapidly converges to high-performance policies, achieving over 95% success on three distinct goal scenarios—including long, multi-turn detours and tight hallway passages—within 60 to 80 episodes. Its deterministic actor and replay-based learning enable the robot to recall and refine successful navigation trajectories, yielding smooth, direct paths and minimal collisions. PPO, by contrast, exhibits greater training stability but slower convergence: it ultimately succeeds in simpler corridor tasks yet fails to generalize to the most complex, multi-turn goal within 100 episodes, often becoming trapped or oscillating at tight corners due to its stochastic policy updates and shorter effective horizon.
These findings confirm that—and delineate how—deep reinforcement learning can alone produce robust, map-free navigation behaviors in dynamic, unstructured environments using minimal sensor inputs. We identify key factors underpinning this success, including reward shaping, algorithmic exploration strategies, and sensor abstraction via LiDAR sectorization. Finally, we discuss the remaining challenges for real-world deployment—such as sensor noise, dynamic obstacles, and sim-to-real transfer—and propose future directions involving memory-augmented policies, hybrid planning integrations, and richer sensor fusion to further enhance autonomy and reliability in practical applications