Machine Learning Algorithms for Automated Artifact Classification in Large Digital Datasets

EasyChair Preprint 14244

Abstract

The exponential growth of digital data presents unique challenges and opportunities for the classification of artifacts within large datasets. Traditional methods of classification, often manual and labor-intensive, struggle to keep pace with the volume and diversity of data. Machine learning (ML) offers a robust solution by automating the classification process, enhancing accuracy, and reducing the time required for data analysis.

This abstract explores the application of machine learning algorithms to the automated classification of artifacts in large digital datasets. It reviews various ML techniques, including supervised learning, unsupervised learning, and deep learning, each offering unique strengths for different types of data and classification tasks. Supervised learning algorithms, such as Support Vector Machines (SVM), Decision Trees, and Neural Networks, are highlighted for their effectiveness in scenarios where labeled training data is available. Unsupervised methods, including clustering algorithms like K-means and hierarchical clustering, are discussed for their ability to identify patterns in unlabeled data. Deep learning approaches, particularly Convolutional Neural Networks (CNNs), are noted for their superior performance in image and text classification tasks.

The abstract also addresses the challenges associated with artifact classification using ML, such as the need for large, annotated datasets, the handling of noisy or incomplete data, and the interpretability of complex models. Moreover, it examines recent advancements in transfer learning and data augmentation techniques, which mitigate these challenges by improving model generalization and efficiency.

Keyphrases: Computer Science, Digital Archaelogy, Machine Learning Algorithm, Mmachine learning, Technology, computing

@booklet{EasyChair:14244,
  author    = {Favour Olaoye and Chris Bell and Peter Broklyn},
  title     = {Machine Learning Algorithms for Automated Artifact Classification in Large Digital Datasets},
  howpublished = {EasyChair Preprint 14244},
  year      = {EasyChair, 2024}}