Scholarlink

The presence of complex time series and multimodal data in economics and finance necessitates the development of advanced analytical models capable of interpreting complex patterns and dynamics. Despite their fast development as LLM, the current AI systems cannot do that. This paper introduces how a novel computational architecture, NeuDen, which combines the strengths of evolving spiking neural networks (eSNNs) with the interpretability of evolving dynamic neuro-fuzzy systems (eDNFS), can be used to model high-dimensional temporal data in economics. The NeuDen model is applied to the analysis of remittance inflows, demonstrating its ability to not only model data but also uncover the underlying trends in intricate remittance data. This application highlights the model's potential as a tool for efficient modelling of economic data that also helps understand the dynamics of economic processes.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Predictive and Explainable Modelling in Economics on the Case Study of Remittance Prediction Using the NeuDen AI Computational Architecture

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Iman AbouHassan,
  • Nikola Kasabov,
  • Roumen Trifonov,
  • George Popov

摘要

The presence of complex time series and multimodal data in economics and finance necessitates the development of advanced analytical models capable of interpreting complex patterns and dynamics. Despite their fast development as LLM, the current AI systems cannot do that. This paper introduces how a novel computational architecture, NeuDen, which combines the strengths of evolving spiking neural networks (eSNNs) with the interpretability of evolving dynamic neuro-fuzzy systems (eDNFS), can be used to model high-dimensional temporal data in economics. The NeuDen model is applied to the analysis of remittance inflows, demonstrating its ability to not only model data but also uncover the underlying trends in intricate remittance data. This application highlights the model's potential as a tool for efficient modelling of economic data that also helps understand the dynamics of economic processes.