Hybrid Human and Machine Learning Algorithms to Forecast the European Stock Market
This paper explores the power of news sentiment to predict financial returns, particularly the returns of a set of European stocks. Building on past decision support work going back to the Delphi method, this paper describes a text analysis expert weighting algorithm that aggregates the responses of...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Hindawi-Wiley
2023-01-01
|
| Series: | Complexity |
| Online Access: | http://dx.doi.org/10.1155/2023/5847887 |
| _version_ | 1797833703517323264 |
|---|---|
| author | Germán G. Creamer Yasuaki Sakamoto Jeffrey V. Nickerson Yong Ren |
| author_facet | Germán G. Creamer Yasuaki Sakamoto Jeffrey V. Nickerson Yong Ren |
| author_sort | Germán G. Creamer |
| collection | DOAJ |
| description | This paper explores the power of news sentiment to predict financial returns, particularly the returns of a set of European stocks. Building on past decision support work going back to the Delphi method, this paper describes a text analysis expert weighting algorithm that aggregates the responses of both humans and algorithms by dynamically selecting the best answer according to previous performance. The proposed system is tested through an experiment in which ensembles of experts, crowds, and machines analyzed Thomson Reuters news stories and predicted the returns of the relevant stocks mentioned right after the stories appeared. In most cases, the expert weighting algorithm was better than or as good as the best algorithm or human. The algorithm’s capacity to dynamically select the best answers from humans and machines results in an evolving collective intelligence: the final decision is an aggregation of the best automated individual answers, some of which come from machines and some from humans. Additionally, this paper shows that the groups of humans, algorithms, and expert weighting algorithms have associated with them, particularly, news topics that these groups are good at making predictions from. |
| first_indexed | 2024-04-09T14:28:29Z |
| format | Article |
| id | doaj.art-a09ad3b5e63e4402be162d10e6fb9ef1 |
| institution | Directory Open Access Journal |
| issn | 1099-0526 |
| language | English |
| last_indexed | 2024-04-09T14:28:29Z |
| publishDate | 2023-01-01 |
| publisher | Hindawi-Wiley |
| record_format | Article |
| series | Complexity |
| spelling | doaj.art-a09ad3b5e63e4402be162d10e6fb9ef12023-05-04T00:00:04ZengHindawi-WileyComplexity1099-05262023-01-01202310.1155/2023/5847887Hybrid Human and Machine Learning Algorithms to Forecast the European Stock MarketGermán G. Creamer0Yasuaki Sakamoto1Jeffrey V. Nickerson2Yong Ren3Stevens Institute of TechnologyStevens Institute of TechnologyStevens Institute of TechnologyStevens Institute of TechnologyThis paper explores the power of news sentiment to predict financial returns, particularly the returns of a set of European stocks. Building on past decision support work going back to the Delphi method, this paper describes a text analysis expert weighting algorithm that aggregates the responses of both humans and algorithms by dynamically selecting the best answer according to previous performance. The proposed system is tested through an experiment in which ensembles of experts, crowds, and machines analyzed Thomson Reuters news stories and predicted the returns of the relevant stocks mentioned right after the stories appeared. In most cases, the expert weighting algorithm was better than or as good as the best algorithm or human. The algorithm’s capacity to dynamically select the best answers from humans and machines results in an evolving collective intelligence: the final decision is an aggregation of the best automated individual answers, some of which come from machines and some from humans. Additionally, this paper shows that the groups of humans, algorithms, and expert weighting algorithms have associated with them, particularly, news topics that these groups are good at making predictions from.http://dx.doi.org/10.1155/2023/5847887 |
| spellingShingle | Germán G. Creamer Yasuaki Sakamoto Jeffrey V. Nickerson Yong Ren Hybrid Human and Machine Learning Algorithms to Forecast the European Stock Market Complexity |
| title | Hybrid Human and Machine Learning Algorithms to Forecast the European Stock Market |
| title_full | Hybrid Human and Machine Learning Algorithms to Forecast the European Stock Market |
| title_fullStr | Hybrid Human and Machine Learning Algorithms to Forecast the European Stock Market |
| title_full_unstemmed | Hybrid Human and Machine Learning Algorithms to Forecast the European Stock Market |
| title_short | Hybrid Human and Machine Learning Algorithms to Forecast the European Stock Market |
| title_sort | hybrid human and machine learning algorithms to forecast the european stock market |
| url | http://dx.doi.org/10.1155/2023/5847887 |
| work_keys_str_mv | AT germangcreamer hybridhumanandmachinelearningalgorithmstoforecasttheeuropeanstockmarket AT yasuakisakamoto hybridhumanandmachinelearningalgorithmstoforecasttheeuropeanstockmarket AT jeffreyvnickerson hybridhumanandmachinelearningalgorithmstoforecasttheeuropeanstockmarket AT yongren hybridhumanandmachinelearningalgorithmstoforecasttheeuropeanstockmarket |