Denna studie handlar om elevers kunskap om evolutionära mekanismer och hur biologilärare utformar sin pedagogik för att stärka elevers begreppsmässiga förståelse. Studien består av fem semistrukturerade intervjuer med gymnasielärare i biologi vid två svenska gymnasieskolor. Studien lägger särskilt fokus på active learning-pedagogik och undersöker huruvida active learning kan främja elevers förståelse för evolutionära kärnbegrepp (variation, selektion och ärftlighet). Studiens resultat indikerar att de vanligaste missförstånden är att individer kan anpassa sig för att överleva, att begreppet fitness biologiska innebörd misstolkas samt en bristande förståelse för begreppet selektionstryck. Studien sätter förståelsen för evolutionära kärnbegrepp i relation till så kallade tröskelbegrepp som är utgör viktig förförståelse samt kognitiva biaser vilka utgör vanliga hinder för att kunna ta till sig kunskap om evolutionära mekanismer. Studien kontextualiserar active learning med närliggande didaktiska teoribildningar såsom Piagets begrepp ackommodera och det övre stegen i Blooms taxonomi. Ett resultat är att de lärare som själva har ett intresse av evolutionsundervisning i större utsträckning utvärderar sin undervisning och aktivt söker nya, mer illustrativa laborationer och övningar. Eftersom undervisningens utformning skiljer sig betydligt lärare emellan föreslås skapande av ett forum där evidensbaserade active learning- övningar kan tillgängliggöras för alla biologilärare. En active learning-laboration som baserat på forskning visat goda resultat beskrivs i detalj och jämförs ingående med en laboration genomförd av tre av de 5 intervjuade lärarna. Båda dessa laborationer utgör exempel på active learning-pedagogik, vilket skulle kunna utgöra en källa till inspiration som biologilärare kan använda för vidareutveckling av sin egen undervisning.

With the advancements in deep learning, object detection networks have become more robust. Nevertheless, a challenge with training deep networks is finding enough labelled training data for the model to perform well, due to constraints associated with acquiring relevant data. For this reason, active learning is used to minimize the cost by sampling the unlabeled samples that increase the performance the most. In the field of object detection, few works have been done in exploring effective hybrid active learning strategies that exploit the intermediate feature embeddings in neural networks. In this work, hybrid active learning methods are proposed and tested, using various uncertainty sampling techniques and the well-respected core-set method as the representative strategy. In addition, experiments are conducted with network embeddings to find a suitable strategy to model representation of all available samples. Experiments show mixed outcomes as to whether hybrid methods perform better than the core-set method used separately. / Med framstegen inom djupinlärning, har neurala nätverk för objektdetektering blivit mer robusta. En utmaning med att träna djupa neurala nätverk är att hitta en tillräcklig mängd träningsdata för att ett nätverk ska prestera bra, med tanke på de begränsningar som är förknippade med anskaffningen av relevant data. Av denna anledning används aktiv maskininlärning för att minimera kostnaden med att förvärva nya datapunkter, genom att göra kontinuerliga urval av de omärkta bilder som ökar prestandan mest. När det gäller objektsdetektering har få arbeten gjorts för att utforska effektiva hybridstrategier som utnyttjar de mellanliggande lagren som finns i ett neuralt nätverk. I det här arbetet föreslås och testas hybridmetoder i kontext av aktiv maskininlärning med hjälp av olika tekniker för att göra urval av datamängder baserade på osäkerhetsberäkningar men även beräkningar med hänsyn till representation (core-set-metoden). Dessutom utförs experiment med mellanliggande nätverksinbäddningar för att hitta en lämplig strategi för att modellera representation av alla tillgängliga bilder i datasetet. Experimenten visar blandade resultat när det gäller huruvida hybridmetoderna presterar bättre i jämförelse med seperata aktiv maskininlärning strategier där core-set metoden inte används.

Les systèmes de recommandation visent à présélectionner et présenter en premier les informations susceptibles d'intéresser les utilisateurs. Ceci a suscité l'attention du commerce électronique, où l'historique des achats des utilisateurs sont analysés pour prédire leurs intérêts futurs et pouvoir personnaliser les offres ou produits (appelés aussi items) qui leur sont proposés. Dans ce cadre, les systèmes de recommandation exploitent les préférences des utilisateurs et les caractéristiques des produits et des utilisateurs pour prédire leurs préférences pour des futurs items. Bien qu'ils aient démontré leur précision, ces systèmes font toujours face à de grands défis tant pour le monde académique que pour l'industrie : ces techniques traitent un grand volume de données qui exige une parallélisation des traitements, les données peuvent être également très hétérogènes, et les systèmes de recommandation souffrent du démarrage à froid, situation dans laquelle le système n'a pas (ou pas assez) d'informations sur (les nouveaux) utilisateurs/items pour proposer des recommandations précises. La technique de factorisation matricielle a démontré une précision dans les prédictions et une simplicité de passage à l'échelle. Cependant, cette approche a deux inconvénients : la complexité d'intégrer des données hétérogènes externes (telles que les caractéristiques des items) et le démarrage à froid pour un nouvel utilisateur. Cette thèse a pour objectif de proposer un système offrant une précision dans les recommandations, un passage à l'échelle pour traiter des données volumineuses, et permettant d'intégrer des données variées sans remettre en question l'indépendance du système par rapport au domaine d'application. De plus, le système doit faire face au démarrage à froid utilisateurs car il est important de fidéliser et satisfaire les nouveaux utilisateurs. Cette thèse présente quatre contributions au domaine des systèmes de recommandation: (1) nous proposons une implémentation d'un algorithme de recommandation de factorisation matricielle parallélisable pour assurer un meilleur passage à l'échelle, (2) nous améliorons la précision des recommandations en prenant en compte l'intérêt implicite des utilisateurs dans les attributs des items, (3) nous proposons une représentation compacte des caractéristiques des utilisateurs/items basée sur les filtres de bloom permettant de réduire la quantité de mémoire utile, (4) nous faisons face au démarrage à froid d'un nouvel utilisateur en utilisant des techniques d'apprentissage actif. La phase d'expérimentation utilise le jeu de données MovieLens et la base de données IMDb publiquement disponibles, ce qui permet d'effectuer des comparaisons avec des techniques existantes dans l'état de l'art. Ces expérimentations ont démontré la précision et l'efficacité de nos approches. / Recommender Systems aim at pre-selecting and presenting first the information in which users may be interested. This has raised the attention of the e-commerce, where the interests of users are analysed in order to predict future interests and to personalize the offers (a.k.a. items). Recommender systems exploit the current preferences of users and the features of items/users in order to predict their future preference in items.Although they demonstrate accuracy in many domains, these systems still face great challenges for both academia and industry: they require distributed techniques to deal with a huge volume of data, they aim to exploit very heterogeneous data, and they suffer from cold-start, situation in which the system has not (enough) information about (new) users/items to provide accurate recommendations. Among popular techniques, Matrix Factorization has demonstrated high accurate predictions and scalability to parallelize the analysis among multiple machines. However, it has two main drawbacks: (1) difficulty of integrating external heterogeneous data such as items' features, and (2) the cold-start issue. The objective of this thesis is to answer to many challenges in the field of recommender systems: (1) recommendation techniques deal with complex analysis and a huge volume of data; in order to alleviate the time consumption of analysis, these techniques need to parallelize the process among multiple machines, (2) collaborative filtering techniques do not naturally take into account the items' descriptions in the recommendation, although this information may help to perform more accurate recommendations, (3) users' and items' descriptions in very large dataset contexts can become large and memory-consuming; this makes data analysis more complex, and (4) the new user cold-start is particularly important to perform new users' recommendations and to assure new users fidelity. Our contributions to this area are given by four aspects: (1) we improve the distribution of a matrix factorization recommendation algorithm in order to achieve better scalability, (2) we enhance recommendations performed by matrix factorization by studying the implicit interest of the users in the attributes of the items, (3) we propose an accurate and low-space binary vector based on Bloom Filters for representing users/items through a high quantity of features in low memory-consumption, and (4) we cope with the new user cold-start in collaborative filtering by using active learning techniques. The experimentation phase uses the publicly available MovieLens dataset and IMDb database, what allows to perform fair comparisons to the state of the art. Our contributions demonstrate their performance in terms of accuracy and efficiency.

The critical shortage of registered nurses (RNs) in the United States has led to increased enrollment in nursing schools, but the number of graduates is still decreasing, as nursing students struggle and fail in upper division courses. There is a significant gap in knowledge concerning students' self-efficacy (SE) as a factor directly influencing students' academic performance. The problem examined in this correlational study was the impact of collaborative learning in an innovative classroom setting on Bachelor of Science in Nursing (BSN) students' SE and academic performance. Framed by Bandura's theory of SE, the research questions examined the relationship between students' SE scores at the beginning and end of the innovative course, and their end-of-course grade. The sample included 22 students from one nursing class (N = 22) in an undergraduate-level nursing program in Texas. Data sources included disaggregated student grades and an anonymous, online survey. Analyses included Chi-square and Pearson's r correlation of the data. Results indicated SE scores at the end of the course were higher than they were at the beginning of the course, which provided an initial understanding of the impact of the innovative learning environment on BSN students' academic performance, but were not statistically significant and could not, therefore, disprove the null hypothesis. This study indicates that student nursing courses could increase student self-efficacy, which would result in a positive impact in hospital and clinic support for United States citizens.

Active learning methodologies (ALM) are associated with student success, but little research on this topic has been pursued at the community college level. At a local community college, students in science, technology, engineering, and math (STEM) courses exhibited lower than average grades. The purpose of this study was to examine whether the use of ALM predicted STEM course grades while controlling for academic discipline, course level, and class size. The theoretical framework was Vygotsky's social constructivism. Descriptive statistics and multinomial logistic regression were performed on data collected through an anonymous survey of 74 instructors of 272 courses during the 2016 fall semester. Results indicated that students were more likely to achieve passing grades when instructors employed in-class, highly structured activities, and writing-based ALM, and were less likely to achieve passing grades when instructors employed project-based or online ALM. The odds ratios indicated strong positive effects (greater likelihoods of receiving As, Bs, or Cs in comparison to the grade of F) for writing-based ALM (39.1-43.3%, 95% CI [10.7-80.3%]), highly structured activities (16.4-22.2%, 95% CI [1.8-33.7%]), and in-class ALM (5.0-9.0%, 95% CI [0.6-13.8%]). Project-based and online ALM showed negative effects (lower likelihoods of receiving As, Bs, or Cs in comparison to the grade of F) with odds ratios of 15.7-20.9%, 95% CI [9.7-30.6%] and 16.1-20.4%, 95% CI [5.9-25.2%] respectively. A white paper was developed with recommendations for faculty development, computer skills assessment and training, and active research on writing-based ALM. Improving student grades and STEM course completion rates could lead to higher graduation rates and lower college costs for at-risk students by reducing course repetition and time to degree completion.

This thesis investigates automated annotation approaches to assist public authorities in Sweden in optimizing resource allocation and gaining valuable insights to enhance the preservation of high-quality welfare. The study uses pre-trained Swedish language models for the named entity recognition (NER) task of finding job titles in job advertisements from The Swedish Public Employment Service, Arbetsförmedlingen. Specifically, it evaluates the performance of the Swedish Bidirectional Encoder Representations from Transformers (BERT), developed by the National Library of Sweden (KB), referred to as KB-BERT. The thesis explores the impact of training data size on the models’ performance and examines whether active learning can enhance efficiency and accuracy compared to random sampling. The findings reveal that even with a small training dataset of 220 job advertisements, KB-BERT achieves a commendable F1-score of 0.770 in predicting job titles. The model’s performance improves further by augmenting the training data with an additional 500 annotated job advertisements, yielding an F1-score of 0.834. Notably, the highest F1-score of 0.856 is achieved by applying the active learning strategy of uncertainty sampling and the measure of mean entropy. The test data provided by Arbetsförmedlingen was re-annotated to evaluate the complexity of the task. The human annotator achieved an F1-score of 0.883. Based on these findings, it can be inferred that KB-BERT performs satisfactorily in classifying job titles from job ads.

I Sverige har utvecklingen gått från faktabaserad miljöundervisning till undervisning om hållbar utveckling som inte bara ska behandla fakta, utan även värderingar, känslor och de utmaningar som världen står inför. Komplexiteten i att undervisa om hållbar utveckling i kombination med att lärare förväntas inkludera det i sina undervisningsämnen sätter stor press på lärarna, inte minst för att många saknar eller tror att de saknar kunskap om hållbar utveckling och hur man kan undervisa om det. Detta bidrar till att lärare undervisar om hållbar utveckling i mindre grad än vad som förväntas från bland annat de globala målen och läroplanen. Med det som grund har den här studien syftat till att försöka minska glappet genom att ta fram ett övningsmaterial som kan användas ämnesövergripande på gymnasiet. Viktiga didaktiska aspekter och kunskapsaspekter togs fram från intervjuer med verksamma gymnasielärare respektive forskare inom polymerteknik. Från intervjuerna med lärarna framkom det att eleverna ska ges möjlighet att diskutera, se olika perspektiv, att undervisningen behandlar ämnen som eleverna kan relatera till. Det mest centrala som framkom i intervjuerna med forskarna var de konflikter som finns kring plast, som hur avfallshantering ska skötas, fördelning av markanvändning och hur ansvaret för nedskräpningen ska fördelas. Detta tillsammans med didaktisk forskning och aktuellvetenskap om plast blev resultatet tre stycken övningar med intressekonflikter som ett genomgående tema där eleverna ges möjlighet att utveckla flera viktiga kompetenser som behövs för att uppnå de globala målen. / In Sweden, the development has gone from a fact-based environmental education to teachingon sustainable development that will not only deal with facts, but also values, feelings and the challenges facing the world. The complexity of teaching about sustainable development combined with what teachers are expected to include in their subjects puts great pressure on them, because many lack or believe they lack knowledge about sustainable development and how to teach about it. This contributes to teachers teaching about sustainable development to a lesser extent than expected from the global goals and curriculum. With that as a basis, this study aims to reduce that gap by developing an education material that can be used across subjects in high school. Important didactic aspects and knowledge aspects derived from interviews with active high school teachers and researchers in polymer technology. From the interviews with the teachers, it emerged that the students should be given the opportunity to discuss, see different perspectives, and that the teachers should include areas that the students care about and can relate to. The most central that emerged in the interviews with the researchers were the conflicts that revolves around plastic, such as how waste management is to be handled, distribution of land use and how responsibility for littering should be distributed. This, together with didactic research and current science onplastic, resulted in three exercises with conflicts of interest as a pervasive theme where students are given the opportunity to develop several important skills needed to achieve theglobal goals.

In many relevant scenarios, there is an imbalance between abundant unlabeled data and scarce labeled data to train predictive models. Semi-Supervised Learning and Active Learning are two distinct approaches to deal with this issue. The first one directly uses the unlabeled data to improve model parameter learning, while the second performs a smart choice of unlabeled points to be sent to an annotator, or oracle, which can label these points and increase the labeled training set. In this context, Generative Models are highly appropriate, since they internally represent the data generating process, naturally benefiting from data samples independently of the presence of labels. This Thesis proposes Expectation-Maximization with Density-Weighted Entropy, a novel active semi-supervised learning framework tailored towards generative models. The method is theoretically explored and experiments are conducted to evaluate its application to Gaussian Mixture Models and Multinomial Mixture Models. Based on its partial success, several questions are raised and discussed as to identify possible improvements and decide which shortcomings need to be dealt with before the method is considered robust and generally applicable. / I många relevanta scenarier finns det en obalans mellan god tillgång på oannoterad data och sämre tillgång på annoterad data för att träna prediktiva modeller. Semi-Övervakad Inlärning och Aktiv Inlärning är två distinkta metoder för att hantera denna fråga. Den första använder direkt oannoterad data för att förbättra inlärningen av modellparametrar, medan den andra utför ett smart val av oannoterade punkter som ska skickas till en annoterare eller ett orakel, som kan annotera dessa punkter och öka det annoterade träningssetet. I detta sammanhang är Generativa Modeller mycket lämpliga eftersom de internt representerar data-genereringsprocessen och naturligt gynnas av dataexempel oberoende av närvaron av etiketter. Denna Masteruppsats föreslår Expectation-Maximization med Density-Weighted Entropy, en ny aktiv semi-övervakad inlärningsmetod som är skräddarsydd för generativa modeller. Metoden utforskas teoretiskt och experiment genomförs för att utvärdera dess tillämpning på Gaussiska Mixturmodeller och Multinomiala Mixturmodeller. Baserat på dess partiella framgång ställs och diskuteras flera frågor för att identifiera möjliga förbättringar och avgöra vilka brister som måste hanteras innan metoden anses robust och allmänt tillämplig.

One of the main aims of inquiry is to engage students as active, not passive, participants in science. The purpose of this study is to describe science educators’ and students’ views about inquiry-based instruction in order to better understand and improve implementation of evidence-based teaching strategies. Inquiry-based techniques have been shown to improve student understanding of scientific concepts, yet, there continue to be challenges in implementing these techniques. This research project utilizes Q Methodology, a research method that captures both common and disparate measures of subjectivity, to identify commonalities and defining viewpoints about inquiry-based teaching and learning. Three significantly different viewpoints were identified and each viewpoint represents differences in teaching styles and classroom environments. Additionally, consensus items reveal students and instructors highly value relating science to everyday life; however, a lack of importance is placed upon peer learning and use of open-ended questions.

This thesis brings a much-needed focus on the quality and scholarship of teaching as it pertains to educational and faculty development. The main purpose of this paper is to outline what more than 200 faculty members across a wide variety of disciplines have focused on over a three-year period to make significant (a 1.5 standard deviation increase or higher in online student ratings) and sustained improvements in their teaching. The top three factors of improvement include active/practical learning, teacher/student interactions, and clear expectations/learning outcomes. The researcher also discusses how institutions and faculty communities of practice, research, and faculty personality contribute to teaching performance. The findings of this research build upon the literature review on the scholarship of teaching. The researcher provides vignettes of faculty who have gone through a change process to improve their teaching, highlights important teaching areas for faculty to focus on in each college, provides practical application for change, and concludes by providing suggestions for future research. This thesis is full of hope and encouragement for all faculty and administrators, regardless of their personality, their current skill level at teaching, or the subject matter they teach.