Leveraging Spatiotemporal Graph Neural Networks for Multi-Store Sales Forecasting

The recent study on the optimization of deep learning models for predicting dynamic market behavior highlights the effectiveness of a hybrid sequence model in forecasting e-commerce transactions. This model utilizes multi-scale temporal convolutions, gated recurrent modules, and time-aware self-attention to accurately predict daily demand for individual SKUs over multiple horizons. The research is based on the UCI Online Retail II dataset and employs rigorous evaluation metrics to ensure reliability.

A new dataset named MultiBanAbs has been developed to facilitate Bangla abstractive summarization, comprising over 54,000 articles and summaries from diverse sources including blogs and newspapers. This initiative addresses the limitations of existing summarization systems that primarily focus on news articles, which often do not reflect the varied nature of real-world Bangla texts.

Recent research has introduced Liquid Time-Constant Networks (LTCs) and Closed-form Continuous-time Networks (CfCs) to model retinal ganglion cell activity in tiger salamanders, demonstrating lower mean absolute error (MAE) and faster convergence compared to traditional convolutional models and LSTMs.

A recent study utilized the METABRIC breast cancer cohort to explore the effectiveness of copula-based fusion of clinical and genomic machine learning risk scores for predicting 5-year cancer-specific mortality. By modeling the joint relationship between clinical variables and genomic data, researchers aimed to enhance risk stratification beyond traditional linear methods.

A recent study has introduced a bidirectional Long Short-Term Memory (LSTM) neural network designed to classify light curves of transient astronomical objects using data from the Photometric LSST Astronomical Time-series Classification Challenge (PLAsTiCC). The study reorganized the original fourteen object classes into five categories to mitigate class imbalance, achieving high performance in certain classifications while facing challenges with others.

A new study has introduced a synthetic data generation framework aimed at identifying pre-injury patterns in triathletes by analyzing lifestyle, recovery, and load dynamics features. This framework generates physiologically plausible athlete profiles and simulates individualized training programs while considering factors such as sleep quality and stress levels, which are often overlooked in injury prediction models.

An AI-based framework has been developed to predict feed intake for individual beef cattle and pen-level aggregation, utilizing data from 19 experiments conducted at the Nancy M. Cummings Research Extension & Education Center in Carmen, ID, alongside environmental data from AgriMet Network weather stations. This framework aims to leverage big data generated by electronic feeding systems to enhance precision livestock farming practices.

A new paper presents a hybrid framework for portfolio optimization that combines Long Short-Term Memory (LSTM) forecasting with Proximal Policy Optimization (PPO) reinforcement learning. This innovative approach aims to enhance portfolio management by leveraging deep learning to predict market trends and dynamically adjust asset allocations across various financial instruments, including U.S. and Indonesian equities, U.S. Treasuries, and cryptocurrencies.