Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Predicting social unrest events in South Africa using LSTM neural networks
This thesis demonstrates an approach to predict the count of social unrest events in South Africa. A comparison is made between traditional forecasting approaches and neural networks; the traditional forecast method selected being the Autoregressive Integrated Moving Average (ARIMA model). The type...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
Department of Statistical Sciences
2021
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613296071278592 |
|---|---|
| access_status_str | Open Access |
| author | Zambezi, Samantha |
| author2 | Nyirenda, Juwa |
| author_browse | Nyirenda, Juwa Zambezi, Samantha |
| author_facet | Nyirenda, Juwa Zambezi, Samantha |
| author_sort | Zambezi, Samantha |
| collection | Thesis |
| description | This thesis demonstrates an approach to predict the count of social unrest events in South Africa. A comparison is made between traditional forecasting approaches and neural networks; the traditional forecast method selected being the Autoregressive Integrated Moving Average (ARIMA model). The type of neural network implemented was the Long Short-Term Memory (LSTM) neural network. The basic theoretical concepts of ARIMA and LSTM neural networks are explained and subsequently, the patterns of the social unrest time series were analysed using time series exploratory techniques. The social unrest time series contained a significant number of irregular fluctuations with a non-linear trend. The structure of the social unrest time series suggested that traditional linear approaches would fail to model the non-linear behaviour of the time series. This thesis confirms this finding. Twelve experiments were conducted, and in these experiments, features, scaling procedures and model configurations are varied (i.e. univariate and multivariate models). Multivariate LSTM achieved the lowest forecast errors and performance improved as more explanatory features were introduced. The ARIMA model's performance deteriorated with added complexity and the univariate ARIMA produced lower forecast errors compared to the multivariate ARIMA. In conclusion, it can be claimed that multivariate LSTM neural networks are useful for predicting social unrest events. |
| format | Thesis |
| id | oai:open.uct.ac.za:11427/33986 |
| institution | University of Cape Town (South Africa) |
| language | eng |
| last_indexed | 2026-06-10T12:33:51.607Z |
| license_str | Not specified — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from UCTD — University of Cape Town Open Access Repository |
| publishDate | 2021 |
| publishDateRange | 2021 |
| publishDateSort | 2021 |
| publisher | Department of Statistical Sciences |
| publisherStr | Department of Statistical Sciences |
| record_format | dspace |
| source_str | UCTD — University of Cape Town Open Access Repository |
| spelling | oai:open.uct.ac.za:11427/33986 Predicting social unrest events in South Africa using LSTM neural networks Zambezi, Samantha Nyirenda, Juwa Statistical Sciences This thesis demonstrates an approach to predict the count of social unrest events in South Africa. A comparison is made between traditional forecasting approaches and neural networks; the traditional forecast method selected being the Autoregressive Integrated Moving Average (ARIMA model). The type of neural network implemented was the Long Short-Term Memory (LSTM) neural network. The basic theoretical concepts of ARIMA and LSTM neural networks are explained and subsequently, the patterns of the social unrest time series were analysed using time series exploratory techniques. The social unrest time series contained a significant number of irregular fluctuations with a non-linear trend. The structure of the social unrest time series suggested that traditional linear approaches would fail to model the non-linear behaviour of the time series. This thesis confirms this finding. Twelve experiments were conducted, and in these experiments, features, scaling procedures and model configurations are varied (i.e. univariate and multivariate models). Multivariate LSTM achieved the lowest forecast errors and performance improved as more explanatory features were introduced. The ARIMA model's performance deteriorated with added complexity and the univariate ARIMA produced lower forecast errors compared to the multivariate ARIMA. In conclusion, it can be claimed that multivariate LSTM neural networks are useful for predicting social unrest events. 2021-09-21T17:03:13Z 2021-09-21T17:03:13Z 2021 2021-09-21T17:02:30Z Master Thesis Masters MSc http://hdl.handle.net/11427/33986 eng application/pdf Department of Statistical Sciences Faculty of Science |
| spellingShingle | Statistical Sciences Zambezi, Samantha Predicting social unrest events in South Africa using LSTM neural networks |
| thesis_degree_str | Master's |
| title | Predicting social unrest events in South Africa using LSTM neural networks |
| title_full | Predicting social unrest events in South Africa using LSTM neural networks |
| title_fullStr | Predicting social unrest events in South Africa using LSTM neural networks |
| title_full_unstemmed | Predicting social unrest events in South Africa using LSTM neural networks |
| title_short | Predicting social unrest events in South Africa using LSTM neural networks |
| title_sort | predicting social unrest events in south africa using lstm neural networks |
| topic | Statistical Sciences |
| url | http://hdl.handle.net/11427/33986 |
| work_keys_str_mv | AT zambezisamantha predictingsocialunresteventsinsouthafricausinglstmneuralnetworks |