Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Estimating South African consumer sentiment using social media data
Thesis (MCom)--Stellenbosch University, 2024.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | en_ZA |
| Published: |
Stellenbosch : Stellenbosch University
2025
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613885480042496 |
|---|---|
| access_status_str | Open Access |
| author | Hugo, Lise-Marie |
| author2 | Nagar, Priyanka |
| author_browse | Hugo, Lise-Marie Nagar, Priyanka |
| author_facet | Nagar, Priyanka Hugo, Lise-Marie |
| author_sort | Hugo, Lise-Marie |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description |
Thesis (MCom)--Stellenbosch University, 2024. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/131675 |
| institution | Stellenbosch University (South Africa) |
| language | en_ZA |
| last_indexed | 2026-06-10T12:43:14.822Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2025 |
| publishDateRange | 2025 |
| publishDateSort | 2025 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/131675 Estimating South African consumer sentiment using social media data Hugo, Lise-Marie Nagar, Priyanka Stellenbosch University. Faculty of Economic and Management Sciences. Dept. of Statistical & Actuarial Science. Consumer behavior -- South Africa Social media -- Economic aspects -- South Africa Machine learning -- Social networks -- South Africa Data mining -- South Africa UCTD Thesis (MCom)--Stellenbosch University, 2024. ENGLISH SUMMARY: This study investigates an alternative measurement of South African consumer confidence based on Twitter data, aiming to understand the implications of social media data in text classification, examine the relationship between social media activity and consumer confidence, and propose an alternative, timely measurement for South African consumer confidence as opposed to the FNB/BER Consumer Confidence Index (CCI). To obtain a representative sample of the South African consumer population, both a random sampling and a hashtag-selection technique were employed and explored. This was followed by the proposal of a hybrid approach combining both techniques. To investigate the relationship between social media activity and consumer confidence, three text classification models were identified and compared, where the random forest model outperformed XGBoost and the support vector machine (SVM). The random forest model was compared to the personal consumption expenditure index (PCE) alongside combinations of the random forest model, bag-of-words estimates using VaderSentiment and the FNB/BER CCI. Additionally, a relationship between social media activity, consumer confidence and consumer expenditure was found, indicating a perceived advantage over the CCI. This study demonstrates that social media data can be effectively utilised to gauge consumer confidence in South Africa. The hybrid approach combining random sampling and hashtag-selection techniques, along with the sentiment models, provided a robust and timely alternative to traditional consumer confidence measurements, FNB/BER CCI. These findings highlight the potential of leveraging social media analytics for economic indicators. AFRIKAANSE OPSOMMING: Geen opsomming beskikbaar. Masters 2025-02-05T09:19:08Z 2025-02-05T09:19:08Z 2024-12 Thesis https://scholar.sun.ac.za/handle/10019.1/131675 en_ZA Stellenbosch University x, 67 pages : illustrations, includes annexures application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Consumer behavior -- South Africa Social media -- Economic aspects -- South Africa Machine learning -- Social networks -- South Africa Data mining -- South Africa UCTD Hugo, Lise-Marie Estimating South African consumer sentiment using social media data |
| title | Estimating South African consumer sentiment using social media data |
| title_full | Estimating South African consumer sentiment using social media data |
| title_fullStr | Estimating South African consumer sentiment using social media data |
| title_full_unstemmed | Estimating South African consumer sentiment using social media data |
| title_short | Estimating South African consumer sentiment using social media data |
| title_sort | estimating south african consumer sentiment using social media data |
| topic | Consumer behavior -- South Africa Social media -- Economic aspects -- South Africa Machine learning -- Social networks -- South Africa Data mining -- South Africa UCTD |
| url | https://scholar.sun.ac.za/handle/10019.1/131675 |
| work_keys_str_mv | AT hugolisemarie estimatingsouthafricanconsumersentimentusingsocialmediadata |