Global ETD Search

11	När en kyss inte är en kyss : En diskursanalys av hur en svensk domstol konstruerar våld och sexuella övergrepp Knutsson Fröjd, Kajsa January 2013 (has links) Tidigare forskning har visat samband mellan språkanvändning och den uppfattning som kvinnor själva och samhället har om våld och övergrepp. Kvinnor som har blivit utsatta för våld och övergrepp möter ofta misstro och skuldbeläggande från omgivningen, vilket leder till att de inte blir trodda och att förövaren i många fall inte döms. I detta arbete undersöks hur en viss språkanvändning framställer våld och övergrepp på ett sätt som bland annat skuldbelägger offret, och vilka konsekvenser den språkanvändningen får. Metoden för arbetet är diskursanalys, och består av en textgranskning av domar från en svensk tingsrätt. Resultaten visar att rätten beskriver våld och övergrepp på ett sätt som leder till konsekvenser som att våldet döljs, förövarens ansvar minskas samt att offret skuldbeläggs. Resultatet leder till förslag på hur rätten istället ska beskriva våld och övergrepp, för att undvika ovan nämnda konsekvenser. Resultatet mynnar också ut i ett förslag till förändring av svensk lagstiftning, men det konstateras även att denna lagändring möjligtvis måste föregås av en samhällelig förändring i stort. / Earlier research has demonstrated a link between language and the view that women themselves and society have of violence and abuse. Women who have experienced violence and abuse often face suspicion and blame from society, which leads to that they are not believed and that the perpetrator in many cases, is not convicted. This study focuses on how the language used in a Swedish court describes violence and abuse in ways that leads to the blaming of the victim, and the further consequences of that specific use of language. The method used in this thesis is discourse analysis, and the empirical study consists of a text review of court judgements by a Swedish court. Findings are that the court describes violence and abuse in a way that leads to that violence is hidden, the perpetrator's responsibility is reduced and the victim gets blamed. The result leads to suggestions on how the court could describe violence and abuse differently, in order to avoid these negative consequences. The result also points towards a proposal for a change in Swedish law, although this change in the law might possibly be preceded by a change in society in general. domestic violence the ideal victim mutual terms improper use of language diskursanalys av domslut kvinnomisshandel det idealiska offret ömsesidiga termer inkorrekt språkanvändning
12	Characterizing Feature Influence and Predicting Video Popularity on YouTube / En karakterisering av olika egenskapers inverkan och förutsägelse av videopopularitet på YouTube Abdihakim, Ali January 2021 (has links) YouTube is an online video sharing platform where users can distribute and consume video and other types of content. The rapid technological advancement along with the proliferation och technological gadgets has led to the phenomenon of viral videos where videos and content garner hundreds of thousands if not million of views in a short span of time. This thesis looked at the reason for these viral content, more specifically as it pertains to videos on YouTube. This was done by building a predictor model using two different approaches and extracting important features that causes video popularity. The thesis further observed how the subsequent features impact video popularity via partial dependency plots. The knn model outperformed logistic regression model. The thesis showed, among other things that YouTube channel and title were the most important features followed by comment count, age and video category. Much research have been done pertaining to popularity prediction, but less on deriving important features and evaluating their impact on popularity. Further research has to be conduced on feature influence, which is paramount to comprehend the causes for content going viral. / YouTube är en online-plattform där användare kan distribuera och konsumera video och andra typer av innehåll. Den snabba tekniska utvecklingen tillsammans med spridningen av mobila plattformar har lett till fenomenet virala videor där videor får hundratusentals, om inte miljontals, visningar på kort tid. I arbetet undersöktes orsaken till virala videor på YouTube. Det gjordes genom att bygga två modeller för att förutspå videopopularitet och därefter analysera viktiga egenskaper som orsakar denna. Resultaten visade att Knn- modellen ger bättre resultat än logistisk regression. Arbetet visade bland annat att YouTube-kanalen och titeln var de viktigaste egenskaperna som driver popularitet, följt av antal kommentarer på en video, videons ålder och videons kategori. Vidare forskning är dock nödvändig inom detta område. Mycket forskning har gjorts för att förutsäga populariteten hos videor, men mindre fokus har lagts på att analysera deras viktiga egenskaper och utvärdera deras inverkan på populariteten. Viral Viralt Computer Sciences Datavetenskap (datalogi)
13	Discovering Implant Terms in Medical Records Jerdhaf, Oskar January 2021 (has links) Implant terms are terms like "pacemaker" which indicate the presence of artifacts in the body of a human. These implant terms are key to determining if a patient can safely undergo Magnetic Resonance Imaging (MRI). However, to identify these terms in medical records is time-consuming, laborious and expensive, but necessary for taking the correct precautions before an MRI scan. Automating this process is of great interest to radiologists as it ideally saves time, prevents mistakes and as a result saves lives. The electronic medical records (EMR) contain the documented medical history of a patient, including any implants or objects that an individual would have inside their body. Information about such objects and implants are of great interest when determining if and how a patient can be scanned using MRI. This information is unfortunately not easily extracted through automatic means. Due to their sparse presence and the unusual structure of medical records compared to most written text, makes it very difficult to automate using simple means. By leveraging the recent advancements in Artificial Intelligence (AI), this thesis explores the ability to identify and extract such terms automatically in Swedish EMRs. For the task of identifying implant terms in medical records a generally trained Swedish Bidirectional Encoder Representations from Transformers (BERT) model is used, which is then fine-tuned on Swedish medical records. Using this model a variety of approaches are explored two of which will be covered in this thesis. Using this model a variety of approaches are explored, namely BERT-KDTree, BERT-BallTree, Cosine Brute Force and unsupervised NER. The results show that BERT-KDTree and BERT-BallTree are the most rewarding methods. Results from both methods have been evaluated by domain experts and appear promising for such an early stage, given the difficulty of the task. The evaluation of BERT-BallTree shows that multiple methods of extraction may be preferable as they provide different but still useful terms. Cosine brute force is deemed to be an unrealistic approach due to computational and memory requirements. The NER approach was deemed too impractical and laborious to justify for this study, yet is potentially useful if not more suitable given a different set of conditions and goals. While there is much to be explored and improved, these experiments are a clear indication that automatic identification of implant terms is possible, as a large number of implant terms were successfully discovered using automated means. AI Machine Learning Medical Records Patient Records Medical Record Electronic Records Electronic Medical Records BERT EMR Implant Terms Implants Term Terms Term Discovery Artificial Intelligence Word Similarity Word Similarity word-similarity embeddings word embeddings word-embeddings transformers KDTREE BALLTREE NER AI Artificiel Intelligens Maskininlärning Patient Journal Medicinsk Journal Elektronisk Medicinsk Journal Termer BERT KDTREE BALLTREE NER liknande ord transformers EMR

Page generated in 0.0217 seconds