A collaborative paper titled Investor Sentiment from Images: A Few-shot Learning Investigation by Professor Xianming Sun, Executive Deputy Director of the Base, and Associate Professor Xiaohang Ren has been published in Journal of Accounting Literature, a prestigious academic journal.

Abstract: This research aims to extract emotional features from New York Times news images (2018–2023) using few-shot learning approaches. Leveraging machine learning, it offers a systematic investigation into how image-driven emotions affect investor behavior in the US equity market and contribute to the prediction of market movements.Design/methodology/approach:This study employs the DeepEMD model to extract emotional features from 181,233 news images, constructing a daily sentiment index based on visual media. By defining sentiment thresholds, the study develops differentiated strategies for positive and negative emotional signals. In addition, it integrates four machine learning models – AdaBoost, Support Vector Machine, ExtraTrees and Random Forest – alongside a traditional linear regression model to forecast the prices of various US stock market indices.Findings:This study finds that news image sentiment has a significant impact on financial markets. Positive sentiment strategies applied to serious news topics are associated with higher returns, whereas negative sentiment in entertainment-related content signals potential opportunities for contrarian investment. Moreover, the influence of image-based sentiment on the market exhibits a delayed effect of approximately 2–3 days, with particularly strong predictive power for small-cap stocks. Compared with the traditional linear models, machine learning approaches demonstrate superior performance in capturing the nonlinear dynamics between sentiment and market behavior, offering novel analytical tools for behavioral finance research and sentiment-driven anomaly-based investment strategies.Originality/value:This study integrates visual data analysis into the domain of behavioral finance, highlighting the distinctive role of image-based sentiment in uncovering market anomalies and informing investment strategies.

Keywords: Few-shot learning/ Investor sentiment/ Image classification/ Financial forecasting

Link: https://erj.ajcass.com/#/issue?id=122073&year=2025&issue=12

Author profile

Xianming Sun is an professor and department head of the Financial Engineering Department at Zhongnan University of Economics and Law, as well as the Executive Deputy Director of the Innovation and Talent Base for Digital Technology and Finance. His research focuses on financial engineering, financial technology, and related fields, with his research findings published in journals such as Journal of Economic Dynamics and Control, Journal of Economic Behavior & Organization, Journal of Futures Markets, Accounting and Finance, and Energy Economics. He has been honored as a "Wenlan Young Scholar" at Zhongnan University of Economics and Law and has led research projects funded by the National Natural Science Foundation of China and the Fundamental Research Funds for the Central Universities. His related research has won the Excellent Paper Award at the 16th China Finance Engineering Annual Conference.

Xiaohang Ren is a specially appointed associate professor at the Business School of Central South University, a highly cited Chinese scholar of Elsevier, and a top 2% scientist in the world. He mainly focuses on research in energy finance, financial risks, and financial measurement. He has published more than 100 papers in authoritative domestic and foreign journals such as Journal of the American Statistical Association, Transportation Research Part A, Energy Economics, Quantitative Finance, Review of Quantitative Finance and Accounting, International Review of Financial Analysis, Technological Forecasting and Social Change, Business Strategy and the Environment, Pacific-Basin Finance Journal, Applied Energy, Resources Policy, Renewable & Sustainable Energy Review, Energy, Journal of Management Sciences in China (English Edition), Systems Engineering - Theory & Practice, and Chinese Journal of Management Science. Among them, more than 30 papers are ESI hot articles or highly cited papers. He serves as the lead editor of Sustainable Communities (Taylor & Francis), associate editor of Humanities & Social Sciences Communications (the only social science and humanities journal affiliated with Nature), and guest editor of SSCI journals such as Climate Change Economics, Economic Change and Restructuring.