白玉培 章芷洋 唐艇宗 張艾嘉 李琳玉
摘 要:21世紀以來,AI人工智能以及計算機技術、通訊技術的快速發(fā)展,人工智能的核心技術機器學習逐漸成為研究的熱點。在當前這個大數據的時代,機器學習的應用前景十分廣闊,更多的行業(yè)選擇使用機器學習進行數據的處理。汽車工業(yè)在發(fā)達及發(fā)展中國家中的經濟占比中較重,主要汽車生產國的外匯交易數據在商業(yè)實踐與國際貿易中具有重要意義。同時,金融交易領域存在大量的歷史數據,而這些數據資源為預測外匯價格以及跌漲趨勢具有重要的作用。文章針對主要汽車生產國外匯交易數據進行了預處理和特征轉換,并對比分析了支持向量機、隨機森林、以及XGBoost模型對外匯交易數據評估的預測能力。研究結果表明XGBoost要優(yōu)于傳統(tǒng)的支持向量機和隨機森林。
關鍵詞:外匯交易 支持向量機 隨機森林 XGBoost
Research on Foreign Exchange Data of Main Automobile Production based on Machine Learning
Bai Yupei Zhang Zhiyang Tang Tingzong Zhang Aijia Li Linyu
Abstract:Since the 21st century, with the rapid development of AI, computer technology and communication technology, machine learning, the core technology of AI, have gradually become research hotspots. In the current era of big data, machine learning has a very broad application prospect. More industries choose to use machine learning for data processing. The automobile industry accounts for a large proportion of the economy in developed and developing countries. The foreign exchange data of major automobile producing countries are of great significance in business practice and international trade. At the same time, there are a lot of historical data in the field of financial transactions, and these data resources play an important role in predicting foreign exchange prices and the trend of decline and rise. This paper preprocesses and transforms the main foreign exchange trade data of automobile production, and analyzes the prediction ability of support vector machine, random forest and XGBoost model. The results show that XGBoost is better than traditional support vector machine and random forest.
Key words:foreign exchange trading, support vector machine, random forest, XGBoost