Predicting the Shanghai Composite Index Using Chinese TikTok Self-Media Data and Machine Learning Model in China
The generation and application of new self-media provide new ways to acquire information access for Internet users. It also provides a large amount of quality data for the accurate prediction of the Shanghai composite index. In this paper, we combined various machine learning and deep learning model...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2024-01-01
|
| Series: | Discrete Dynamics in Nature and Society |
| Online Access: | http://dx.doi.org/10.1155/ddns/7201831 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841555857660182528 |
|---|---|
| author | Li Zhiming Han Huijian Li Zongwei Zhang Rui |
| author_facet | Li Zhiming Han Huijian Li Zongwei Zhang Rui |
| author_sort | Li Zhiming |
| collection | DOAJ |
| description | The generation and application of new self-media provide new ways to acquire information access for Internet users. It also provides a large amount of quality data for the accurate prediction of the Shanghai composite index. In this paper, we combined various machine learning and deep learning models with the search data of Chinese TikTok, which is related to the Shanghai composite index, to predict the Shanghai composite index. In addition, we compared and analyzed the prediction results of several machine learning and deep learning models in the short term, medium term, and long term. The results showed that the support vector regression model had the lowest mean absolute percentage error and the highest prediction accuracy in the short, medium, and long term, and the strongest robustness compared with other models. This was followed by random forest regression, which outperformed the remaining five benchmark prediction models (convolutional neural network, LSTM, GRU neural network, radial basis function neural network, extreme learning machine, and transformer model) in terms of prediction accuracy and robustness. The prediction results provide an innovative exploration of the prediction of the Shanghai composite index using self-media network search data. The prediction method provides a new research idea for macroeconomic prediction and forecasting and also enriches the theoretical research of machine learning methods in the field of macroeconomic index prediction. |
| format | Article |
| id | doaj-art-be9349481d0242d2be02bc909c3f5d0c |
| institution | Kabale University |
| issn | 1607-887X |
| language | English |
| publishDate | 2024-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Discrete Dynamics in Nature and Society |
| spelling | doaj-art-be9349481d0242d2be02bc909c3f5d0c2025-01-08T00:00:04ZengWileyDiscrete Dynamics in Nature and Society1607-887X2024-01-01202410.1155/ddns/7201831Predicting the Shanghai Composite Index Using Chinese TikTok Self-Media Data and Machine Learning Model in ChinaLi Zhiming0Han Huijian1Li Zongwei2Zhang Rui3Management Science and Engineering SchoolComputer Science SchoolData Analysis DepartmentComputer Science SchoolThe generation and application of new self-media provide new ways to acquire information access for Internet users. It also provides a large amount of quality data for the accurate prediction of the Shanghai composite index. In this paper, we combined various machine learning and deep learning models with the search data of Chinese TikTok, which is related to the Shanghai composite index, to predict the Shanghai composite index. In addition, we compared and analyzed the prediction results of several machine learning and deep learning models in the short term, medium term, and long term. The results showed that the support vector regression model had the lowest mean absolute percentage error and the highest prediction accuracy in the short, medium, and long term, and the strongest robustness compared with other models. This was followed by random forest regression, which outperformed the remaining five benchmark prediction models (convolutional neural network, LSTM, GRU neural network, radial basis function neural network, extreme learning machine, and transformer model) in terms of prediction accuracy and robustness. The prediction results provide an innovative exploration of the prediction of the Shanghai composite index using self-media network search data. The prediction method provides a new research idea for macroeconomic prediction and forecasting and also enriches the theoretical research of machine learning methods in the field of macroeconomic index prediction.http://dx.doi.org/10.1155/ddns/7201831 |
| spellingShingle | Li Zhiming Han Huijian Li Zongwei Zhang Rui Predicting the Shanghai Composite Index Using Chinese TikTok Self-Media Data and Machine Learning Model in China Discrete Dynamics in Nature and Society |
| title | Predicting the Shanghai Composite Index Using Chinese TikTok Self-Media Data and Machine Learning Model in China |
| title_full | Predicting the Shanghai Composite Index Using Chinese TikTok Self-Media Data and Machine Learning Model in China |
| title_fullStr | Predicting the Shanghai Composite Index Using Chinese TikTok Self-Media Data and Machine Learning Model in China |
| title_full_unstemmed | Predicting the Shanghai Composite Index Using Chinese TikTok Self-Media Data and Machine Learning Model in China |
| title_short | Predicting the Shanghai Composite Index Using Chinese TikTok Self-Media Data and Machine Learning Model in China |
| title_sort | predicting the shanghai composite index using chinese tiktok self media data and machine learning model in china |
| url | http://dx.doi.org/10.1155/ddns/7201831 |
| work_keys_str_mv | AT lizhiming predictingtheshanghaicompositeindexusingchinesetiktokselfmediadataandmachinelearningmodelinchina AT hanhuijian predictingtheshanghaicompositeindexusingchinesetiktokselfmediadataandmachinelearningmodelinchina AT lizongwei predictingtheshanghaicompositeindexusingchinesetiktokselfmediadataandmachinelearningmodelinchina AT zhangrui predictingtheshanghaicompositeindexusingchinesetiktokselfmediadataandmachinelearningmodelinchina |