基於深度學習的台股指數預測模型結合新聞情緒分析

An FITX Stock Prediction Model Combining News Sentiment Analysis

鍾仕潁(SHIH-YING JHONG)

指導教授 ： 張瑞雄

國立臺北商業大學/管理學院/資訊與決策科學研究所/碩士(2022年)

摘要

近年來由於機器學習的發展，許多領域知識開始結合機器學習來做應用，其中深度學習以及強化學習更為當今主流。現今有許多深度學習結合股票預測相關的應用，加上近年自然語言處理技術的快速發展，因此本論文將研究藉由將情感分析的要素結合台股指數預模型，來增加台股指數預測模型的表現，同時本文也提出一種基於深度學習框架的台股指數預測模型，與情感分析的要素做結合，同時也將其他不同方法所訓練出的台股指數預測模型與情感分析的要素做結合，最後經研究發現，台股指數預測模型，加上情感分析的要素後，確實能增加模型的表現，同時本文所提出之模型相較於其他不同的預測模型，也具有較好的表現。

關鍵字

情感分析 ； 股價預測 ； 深度學習 ； 資料分析 ； 網路爬蟲

並列摘要

In recent years, due to the development of machine learning, many domain knowledge has begun to be applied in combination with machine learning, among which deep learning and reinforcement learning are more mainstream today.
Now, there are many applications related to deep learning combined with stock forecasting, and with the rapid development of natural language processing technology in recent years, this paper will study the combination of elements of sentiment analysis with the Taiwan stock index pre-model to increase the performance of the Taiwan stock index forecasting model. At the same time, this paper also proposes a Taiwan stock index prediction model based on the deep learning framework, which is combined with the elements of sentiment analysis, and also combines other Taiwan stock index prediction models using different methods with the elements of sentiment analysis.
Finally, the research found that the Taiwan stock index prediction model, together with the elements of sentiment analysis, can indeed increase the performance of the model. At the same time, the model proposed in this paper also has better performance than other prediction models.

並列關鍵字

sentiment analysis ； Stock prediction ； Deep learing ； Data analysis ； Web crawler

參考文獻

[1] 15 of The Best Sentiment Analysis Tools 取自: https://monkeylearn.com/blog/sentiment-analysis-tools/

[2] Akita, R., Yoshihara, A., Matsubara, T., & Uehara, K. (2016, June). Deep learning for stock prediction using numerical and textual information. In 2016 IEEE/ACIS 15th International Conference on Computer and Information Science (ICIS) (pp. 1-6). IEEE.

[3] Baccianella, S., Esuli, A., & Sebastiani, F. (2010, May). Sentiwordnet 3.0: An enhanced lexical resource for sentiment analysis and opinion mining. In Proceedings of the Seventh International Conference on Language Resources and Evaluation (LREC'10).

[4] Cambria, E., Poria, S., Hazarika, D., & Kwok, K. (2018, April). SenticNet 5: Discovering conceptual primitives for sentiment analysis by means of context embeddings. In Proceedings of the AAAI conference on artificial intelligence (Vol. 32, No. 1).

[5] Chen, C. C., Huang, H. H., & Chen, H. H. (2018, May). NTUSD-Fin: a market sentiment dictionary for financial social media data applications. In Proceedings of the 1st Financial Narrative Processing Workshop (FNP 2018)