The paper introduces a novel method called Session-based Recommendation with Graph Neural Networks (SR-GNN) for session-based recommendation systems. Traditional methods often struggle with accurate user representations and complex item transitions within sessions. SR-GNN addresses these issues by modeling session sequences as graph-structured data, where each session is represented as a graph. GNNs capture complex item transitions within sessions, which are difficult to reveal using conventional sequential methods. Each session is then represented as a combination of global preference and current interest using an attention network. Extensive experiments on real datasets demonstrate that SR-GNN outperforms state-of-the-art methods in session-based recommendation tasks. The main contributions of the work include a novel approach to modeling session sequences as graphs and using GNNs to capture complex item transitions, as well as a session embedding strategy that combines long-term preference and current interests.The paper introduces a novel method called Session-based Recommendation with Graph Neural Networks (SR-GNN) for session-based recommendation systems. Traditional methods often struggle with accurate user representations and complex item transitions within sessions. SR-GNN addresses these issues by modeling session sequences as graph-structured data, where each session is represented as a graph. GNNs capture complex item transitions within sessions, which are difficult to reveal using conventional sequential methods. Each session is then represented as a combination of global preference and current interest using an attention network. Extensive experiments on real datasets demonstrate that SR-GNN outperforms state-of-the-art methods in session-based recommendation tasks. The main contributions of the work include a novel approach to modeling session sequences as graphs and using GNNs to capture complex item transitions, as well as a session embedding strategy that combines long-term preference and current interests.