金融工程：AI识图关注银行、金融、公用事业、红利低波

• The report utilizes Convolutional Neural Networks (CNN) to model price-volume data and future prices, mapping learned features to industry theme sectors[1][75] - The latest configuration themes include banking, finance, utilities, and low volatility dividends, specifically covering indices such as the CSI Bank Index, CSI 800 Bank Index, SSE 180 Financial Stock Index, CSI All Share Utilities Index, and CSI Dividend Low Volatility Index[1][75][77] - The report provides detailed configuration information for the CNN industry themes, including specific dates and index codes[76] Quantitative Models and Construction Methods 1. Model Name: Convolutional Neural Network (CNN) - Construction Idea: Use CNN to model price-volume data and future prices, mapping learned features to industry theme sectors[1][75] - Detailed Construction Process: - Standardize price-volume data into charts for each stock within a window period - Apply CNN to these charts to learn features - Map the learned features to industry theme sectors - Evaluation: The model effectively identifies and maps features to relevant industry themes, providing actionable insights for sector allocation[1][75] Model Backtesting Results 1. CNN Model, Banking Theme, CSI Bank Index[76] 2. CNN Model, Banking Theme, CSI 800 Bank Index[76] 3. CNN Model, Financial Theme, SSE 180 Financial Stock Index[76] 4. CNN Model, Utilities Theme, CSI All Share Utilities Index[76] 5. CNN Model, Low Volatility Dividends Theme, CSI Dividend Low Volatility Index[76]