Principal AI Scientist, AISciences.ai, Monroe Township, NJ 08831, USA.
Tel.: 17187025746; email: raghu.etukuru@aisciences.ai (R.E.)
Manuscript submitted January 12, 2024; accepted January 31, 2024; published February 19, 2024.
Abstract—Traditional linear and simpler models often fail to capture the complex, multifaceted nature of real-world data, leading to inaccurate predictions. This research addresses this challenge by exploring the potential of complexity-conscious prediction, which seeks to incorporate the inherent intricacy within the data. The paper aims to demonstrate the significance of acknowledging and incorporating data complexity in forecasting models, especially in domains where accurate predictions are crucial for informed decision-making and can have a profound impact. By employing statistical methods to measure intricate patterns and by developing advanced deep learning models, such as Long Short-Term Memory (LSTM) networks and Generative Adversarial Networks (GANs), this research endeavors to achieve more accurate and reliable forecasts. Both LSTM and GAN models demonstrated remarkable capability in handling complex time series data, with MAPE values below 3.5%, indicating high accuracy. The GAN model, in particular, showed exceptional performance with a MAPE of less than 2% across all tested stocks, underscoring its advanced predictive capabilities. The findings suggest that deep learning models, especially GANs, substantially improve accuracy over traditional linear forecasting methods. This supports the thesis that integrating data complexity into predictive models through advanced deep learning techniques can significantly enhance forecast precision, thus providing a notable advantage in fields where accurate forecasting is crucial.
keywords—AI-Driven forecasting, complexity-conscious prediction, forecasting accuracy, generative adversarial networks, intricate patterns, time series.
Cite: Raghurami Etukuru, "Enhancing Forecasting Accuracy through Artificial Intelligence-Driven Complexity-Conscious Prediction Modeling," Journal of Advances in Artificial Intelligence vol. 2, no. 1, pp. 60-78, 2024.
Copyright © 2024 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (
CC BY 4.0).
