Loading...

Predicting Novelty Concepts in Data Streams

Soudani, Heydar | 2022

437 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 55328 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Beigy, Hamid
  7. Abstract:
  8. Many real-world environment challenges are not considered in laboratory-controlled models. Although different and powerful models have been developed for object detection and classification in diverse applications, many fail in the real world. One of the most important challenges is dealing with unknown data at the inference time. The second challenge is to change the characteristics of the data distribution over time, known as concept drift. These two important challenges are explored in the Data Stream environment, along with many of the events that a model may face in the real world. To address the challenges of learning in a data stream environment, this thesis first designs a comprehensive framework. Then, the adaptive learning models have been designed and implemented in the proposed framework with the help of Metric learning and Meta-learning algorithms, which are up-to-date models of machine learning. For further study, we also examined the proposed learning models in the form of OWR, incremental learning, and test-the-train, which are highly relevant to learning in a data stream environment. According to the results, the models designed in this study outperformed the existing methods in a range of different datasets and evaluation criteria.
  9. Keywords:
  10. Data Stream ; Incremental Learning ; Concept Drift ; Metalearning ; Metric Learning

 Digital Object List

 Bookmark

...see more