Stock volatility is an important measure of financial risk. Due to the complexity and variability of financial markets, time series forecasting in the financial field is extremely challenging. This paper proposes a “model fusion learning algorithm” and a “feature reconstruction neural network” to forecast the future 10 min volatility of 112 stocks from different industries over the past three years. The results show that the model in this paper has higher fitting accuracy and generalization ability than the traditional model (CART, MLR, LightGBM, etc.). This study found that the “model fusion learning algorithm” can be well applied to financial data modeling; the “feature reconstruction neural network” can well-model data sets with fewer features.
With the development of the Internet, various forms of learning resources continue to flood the public, especially the promotion of video learning plat-forms. More and more people use their spare time to learn what they need. However, there is a general lack of intelligence in online learning platforms at present, which greatly reduced the utilization of online learning platforms and their educational advantages. The innovation of this paper is proposing and explaining that with the integration of big data, online education and ar-tificial intelligence, the contradiction of online education has turned into one between the lack of intelligence in online education and the demand of users. To solve this contradiction, this paper researches from the perspective of an algorithm. Course recommendation is the core algorithm of online ed-ucation. However, the current recommendation algorithm based on collabo-rative filtering has the disadvantages of cold start and useless recommenda-tion content. In this paper, Apriori and ACO algorithms in artificial intelli-gence are studied, and the proposal of an algorithmic framework named Posi-tion-Apriori-ACO brings forth new ideas in solving online education prob-lems. The Position-Apriori-ACO algorithm can effectively carry out course recommendation and learning path planning, and also provides a research di-rection for the intelligent development of online education.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.