Sökning: "active machine learning"

Visar resultat 1 - 5 av 124 uppsatser innehållade orden active machine learning.

  1. 1. "The Uphill AI Contract Challenge The Intra-Active Task: Reimagining Contracts"

    Magister-uppsats, Göteborgs universitet/Juridiska institutionen

    Författare :Filip Seiborg Wikström; [2024-02-16]
    Nyckelord :AI; Contract Law; New Materialism; Karen Barad; Intra-Action; Spacetimemattering; Ethico-Onto-Epistem-Ology; Cartesian-Newtonian paradigms; Antimethodology; Agency; Machine Learning;

    Sammanfattning : The traditional contract theories are insufficient to handle the challenges Artificial Intelligence (AI) is currently causing and will continue to cause to contract law. These challenges involve problems concerning the subject/object divide, agency, the embedding of legal code into interactive programming code, and ethical aspects concerning the transfer of power away from lawyers. LÄS MER

  3. 2. Analyzing the performance of active learning strategies on machine learning problems

    Uppsats för yrkesexamina på avancerad nivå, Uppsala universitet/Avdelningen för systemteknik

    Författare :Vendela Werner; [2023]
    Nyckelord :computer science; bioinformatics; machine learning; active learning; artificial intelligence; supervised learning; Astrazeneca; maskininlärning; artificiell intelligens; datorvetenskap; active learning; bioinformatik; supervised learning;

    Sammanfattning : Digitalisation within industries is rapidly advancing and data possibilities are growing daily. Machine learning models need a large amount of data that are well-annotated for good performance. To get well-annotated data, an expert is needed, which is expensive, and the annotation itself could be very time-consuming. LÄS MER

  4. 3. Active learning for text classification in cyber security

    Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Amanda Carp; [2023]
    Nyckelord :Interactive machine learning; Active learning; Cost-effective active learning; Cyber environment; Interaktiv maskininlärning; Aktiv inlärning; Kostnadseffektiv aktiv inlärning; Cyberdomänen;

    Sammanfattning : In the domain of cyber security, machine learning promises advanced threat detection. However, the volume of available unlabeled data poses challenges for efficient data management. This study investigates the potential for active learning, a subset of interactive machine learning, to reduce the effort required for manual data labelling. LÄS MER

  5. 4. Utilizing Transformers with Domain-Specific Pretraining and Active Learning to Enable Mining of Product Labels

    Uppsats för yrkesexamina på avancerad nivå, Uppsala universitet/Avdelningen för beräkningsvetenskap

    Författare :Erik Norén; [2023]
    Nyckelord :Machine Learning; Natural Language Processing; NLP; BERT; Active Learning; AL; Pharmacovigilance; UMC; Uppsala Monitoring Centre; Structured Product Label; SPL; Transformers; Domain-Specific Pretraining; NER; Named Entity Recognition; Adverse Drug Reaction; ADR;

    Sammanfattning : Structured Product Labels (SPLs), the package inserts that accompany drugs governed by the Food and Drugs Administration (FDA), hold information about Adverse Drug Reactions (ADRs) that exists associated with drugs post-market. This information is valuable for actors working in the field of pharmacovigilance aiming to improve the safety of drugs. LÄS MER

  7. 5. Designing micro exertion game for 2-minute active breaks with ML-agent-based skeletal tracking act the Godot Engine

    Kandidat-uppsats, Linköpings universitet/Institutionen för datavetenskap

    Författare :Mohamad Shafeah; Rayan Bayat; [2023]
    Nyckelord :Exergame; mining exergame; 2-min-exergame.;

    Sammanfattning : This thesis presents the design and development of an innovative exergame aimed atcombating the adverse effects of sedentary lifestyles, particularly prevalent among officeworkers and gamers. The game, built using the open-source Godot game engine andGoogle's MediaPipe machine learning framework, utilizes motion capture technology totranslate real-world player movements into in-game actions. LÄS MER