Search

Researchers have long used gravity models to analyze international trade patterns, identify export opportunities, and negotiate trade agreements. Recent research has emphasized the significance of relatedness and product complexity research in developing robust economic development strategies. This paper presents a novel approach, incorporating relatedness and product complexity as integral elements for interpreting export potential within gravity models powered by machine learning. Our approach stands out for its proficiency in accurately predicting bilateral trade values at a detailed product group level, providing valuable insights for policymakers and other stakeholders. The research leverages random forest machine learning models for predictions and incorporates relatedness and complexity to reveal new dimensions in international trade analysis. • Inspired by the rapidly evolving research on product complexity and relatedness, we explore the possibilities of incorporating the value of potential export flows when analyzing opportunities for diversifying exports. • Gravity models of international trade are commonly employed tools for estimating trade flows and analyzing export potential. However, traditionally, these models have been utilized to estimate aggregated trade or trade within a specific product category. • We demonstrate that machine learning methods can handle thousands of product groups in one integrated model, predicting export potential at the detailed 6-digit HS product group level. • Building on Hidalgo's work, we propose a 5 W economic complexity framework for assessing export diversification opportunities, covering what, when, worth, where, and who. • To pinpoint even better-informed priorities for export promotion, future research should integrate additional features such as trade barriers, expected market demand growth rates, and risks associated with destination markets. [ABSTRACT FROM AUTHOR]

Congestion is a bane of urban life that affects a large share of the population on a daily basis. Thus, congestion gets tremendous attention from city stakeholders, residents, and researchers. The key challenge to preventing congestion is to accurately predict the traffic status (e.g., speed) of a particular road segment which is greatly affected by many factors, such as spatial, temporal, and road conditions. Although several research studies have focused on preventing congestion, most prediction-based literature came short of accurate predictions regarding precision and time efficiency regarding large-scale datasets. This paper proposes a new hybrid approach called Grizzly. This approach utilizes an improved Sequence to Sequence Bi-directional Long Short Term Memory Neural Network model that integrates data pre-processing techniques such as normalization and embeddings to improve traffic prediction accuracy. Carried out experiments on two large-scale real-world datasets, namely PEMS-BAY and METR-LA, pinpointing that the proposed approach outperformed the pioneering competitors from time-series-based and hybrid neural network-based baselines in terms of the agreed-on evaluation criteria (precision and computation time). • Bidirectional LSTM-based approach has been developed for multistep speed prediction. • Seq-to-Seq architecture is relevant to capturing long-term temporal dependencies. • The combination of Embedding and Normalization improves prediction quality and cost. • Grizzly outperforms both classical baselines and hybrid Neural Network-based models. • Grizzly was evaluated on two large real-world datasets with three prediction tasks. [ABSTRACT FROM AUTHOR]