Massey Documents by Type

Permanent URI for this communityhttps://mro.massey.ac.nz/handle/10179/294

Browse

Has files: YesSubject: search.filters.subject.Dimensionality reduction

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    MOTL: enhancing multi-omics matrix factorization with transfer learning
    (BioMed Central Ltd, 2025-12-01) Hirst DP; Térézol M; Cantini L; Villoutreix P; Vignes M; Baudot A; Cosgrove A
    Joint matrix factorization is popular for extracting lower dimensional representations of multi-omics data but loses effectiveness with limited samples. Addressing this limitation, we introduce MOTL (Multi-Omics Transfer Learning), a framework that enhances MOFA (Multi-Omics Factor Analysis) by inferring latent factors for small multi-omics target datasets with respect to those inferred from a large heterogeneous learning dataset. We evaluate MOTL by designing simulated and real data protocols and demonstrate that MOTL improves the factorization of limited-sample multi-omics datasets when compared to factorization without transfer learning. When applied to actual glioblastoma samples, MOTL enhances delineation of cancer status and subtype.
  • Thumbnail Image
    Item
    Texture feature extraction and classification : a comparative study between traditional methods and deep learning : a thesis presented in partial fulfilment of the requirements for the degree of Master of Information Science in Computer Sciences at Massey University, Auckland, New Zealand
    (Massey University, 2020) Li, Jun
    Image classification has always been a core problem of computer vision. With the development of deep learning, it also provides a good solution for us to solve the problem of image feature extraction in image classification. In this thesis we used machine learning and convolutional neural network to study texture feature extraction and classification problems. We implemented a pipeline within the sklearn framework that utilized Local Binary Pattern (LBP) and Haralick as our feature descriptor and various classifiers (namely KNearest Neighbors, Linear Discriminant Analysis, Support Vector Machines, Multilayer Perceptron, Gaussian Naive Bayes, Random Forest, AdaBoost, Logistic Regression and Decision Tree) to evaluate the performance on some popular texture datasets (Brodatz dataset, four extended Outex datasets and VisTex dataset). We also employed Linear Discriminant Analysis as our dimension reduction schema to observe the changes in classification accuracy. We also took advantage of Keras with TensorFlow backend framework and built a pipeline that uses ImageNet-trained convolutional neural network models to train and analyze classifier, extract image feature information and make predictions on test dataset samples. This allowed us to compare the results between traditional methods and CNN based methods. It was found that the classification accuracy has been greatly improved with the CNN based method.