Hashemi, Mahdi
(2018)
Weighted machine learning for spatial-temporal data.
Doctoral Dissertation, University of Pittsburgh.
(Unpublished)
This is the latest version of this item.
Abstract
Sometimes not all training samples are equal in supervised machine learning due to their different accuracy, reliability, source, relevance, or other reasons. Non-weighted machine learning techniques are designed for equally important training samples: (a) the cost of misclassification is equal for training samples in parametric classification techniques, (b) residuals are equally important in parametric regression models, and (c) when voting in non-parametric classification and regression models, training samples either have equal weights or their weights are determined internally by kernels in the feature space, thus no external weights. In this thesis, we develop the weighted versions of Bayesian predictor, perceptron, multilayer perceptron, SVM, and decision tree and show how their results would be different from their non-weighted versions.
Applying machine learning techniques to spatial-temporal data poses the question that how the recorded location and time for training samples should contribute to the training and testing process. The prior knowledge of how spatial-temporal phenomena are autocorrelated cannot be properly captured by machine learning techniques which either ignore location and time altogether, or consider them as input features. Not to mention that the latter approach leads to increased sparseness of data in the feature space and more free parameters in the predictor; thus demanding for larger training datasets. We use the prior knowledge about the spatial-temporal autocorrelation to determine how relevant each training sample would be, given its spatial and temporal distances to the irresponsive (unlabeled) sample. Weighted machine learning techniques use this prior knowledge by taking the relevance of training samples with regard to the irresponsive sample into account as training samples’ weights. The proposed approach overcomes the aforementioned issues by enriching the training process with the prior knowledge about spatial-temporal autocorrelation. Because the spatial-temporal weight of training samples depends on the irresponsive sample’s location and time, the machine needs to be trained separately for each irresponsive sample. However, we show that in practice using only a small subset of training samples with largest spatial-temporal weights not only mitigates the training time but also results in the best accuracy in most cases.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
24 January 2018 |
| Date Type: |
Publication |
| Defense Date: |
25 September 2017 |
| Approval Date: |
24 January 2018 |
| Submission Date: |
25 November 2017 |
| Access Restriction: |
5 year -- Restrict access to University of Pittsburgh for a period of 5 years. |
| Number of Pages: |
131 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
School of Computing and Information > Information Science |
| Degree: |
PhD - Doctor of Philosophy |
| Thesis Type: |
Doctoral Dissertation |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
Machine learning, Spatial-temporal data, Analytical learning |
| Date Deposited: |
24 Jan 2018 16:27 |
| Last Modified: |
24 Jan 2023 06:15 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/33695 |
Available Versions of this Item
-
Weighted machine learning for spatial-temporal data. (deposited 24 Jan 2018 16:27)
[Currently Displayed]
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}