Cognitive Investment Advisor Based on Adaptive Neuro-Fuzzy Intelligence: CogANFIS

Investment decision-making is fundamentally a cognitive process, especially for individual investors who face complex choices under uncertainty. This study presents a cognitive investment advisor that combines an adaptive neuro-fuzzy inference system (ANFIS) with behavioral clustering techniques (K-Means and self-organizing maps). This system (CogANFIS) models three key dimensions of investor input: demographic characteristics, financial status, and psychological/cognitive characteristics. The system predicts the most appropriate investment type, even for non-professional respondents, based on previous investment tendencies. The proposed framework is continuously adapted through feedback loops. Thus, it enhances the cognitive harmony between human characteristics and machine intelligence in financial recommendations. Empirical evaluation using real-world data from 1542 respondents, along with simulated scenarios in 27 rules, demonstrated strong predictive performance ($\mathbf{R M S E} \boldsymbol{=} \mathbf{0 . 7 2}$). This research advances cognitive-adaptive frameworks for human-machine collaboration by integrating behavioral dimensions and expert feedback mechanisms into an integrated neuro-fuzzy architecture. Thus, it enhances personalized financial decision-making in dynamic market environments.