Accurate customer LifeTime Value (LTV) predictions are of critical importance for evaluating the efficiency of customer management strategies, which could enhance advertising placement for better decision-making in ad systems. However, existing solutions for LTV prediction usually rely on determined historical sequence data, which are challenging to apply in Baidu ads due to two unique features including (i) uncertain behavior sequence of customers caused by dynamic advertising strategies, (ii) complex and long-tail distribution of LTV caused by continuous customer behaviors and unique business pattern of search and news feed ads in Baidu. To incorporate these new factors, we propose an Uncertain behavior Sequence modeling framework to predict customer LifeTime Value (USE-LTV), where we (i) utilize a transformer module to extract uncertain sequence features, and develop a dynamic weight mechanism to capture differentiated information under uncertain behavior sequence, (ii) design an continuous loss function tailored to the real-world long-tail exponential LTV distribution in Baidu ads. We extensively evaluate our method based on the industrial-scale real-world data from Baidu, one of the world's largest ads platform, demonstrate that USE-LTV achieves 11.64% of NMAE improvement for a-year LTV compared to the state-of-the-art method.

Motivated by the pursuit of greater efficiency, companies, such as Amazon and JD, are shifting toward a warehouse-distribution integration model to optimize logistics operations. In general full-link logistics scenarios, the collaboration between warehouses and sorting centers managed by different enterprises leads to data silos, posing challenges in securely sharing information and accurately predicting delivery times across the entire logistics network. Current delivery time prediction methods often overlook the heterogeneity of logistics networks and face data sharing constraints. We aim to address these issues by facilitating secure internode relationship analysis and leveraging distinct spatio-temporal characteristics to enhance efficiency. However, challenges remain in overcoming data isolation while maintaining protection and integrating diverse node characteristics for optimized modeling. To address these challenges, we propose the federated heterogeneous graph transformer (Fed-HGT) framework. This framework includes a federated training module that integrates local and central gradients by exchanging node representations and model parameters between logistics nodes and the central server. Additionally, it features a federated prediction module where local nodes compute time representations using their local data and transmit these to the central server. The central server then uses these representations to make accurate full-link delivery time predictions. Our method was evaluated on a dataset from a major e-commerce platform in China, demonstrating significant performance improvements over existing solutions.

A warehouse-distribution integration (WDI) e-commerce platform is an approach that combines warehousing and distribution processes, which is increasingly adopted in industry to enhance business efficiency. In the WDI e-commerce, one of the most important problems is to estimate the full-link delivery time for decision-making. Traditional methods designed for separate warehouse-distribution models struggle to address challenges in integrated systems. The difficulties stem from two main factors: (i) the contextual influence exerted by neighboring units within heterogeneous delivery networks, and (ii) the uncertainty in delivery times caused by dynamic and periodic temporal factors such as fluctuations in online sales volumes and the varying characteristics of different delivery units (e.g., warehouses and sorting centers). To address these challenges, we propose a novel full-link delivery time estimation framework calledHeterogeneousPeriodicSpatial-TemporalGraphTransformer (HPST-GT). First, we develop heterogeneous graph transformers to capture the hierarchical and diverse information of the warehouse-distribution network. Next, we design spatial-temporal transformers based on heterogeneous features to analyze the correlation between spatial and temporal information. Finally, we create a heterogeneous spatial-temporal graph prediction module to estimate full-link delivery time. Our method, evaluated on a one-month dataset from a leading e-commerce platform, surpasses current benchmarks across multiple performance metrics.

The emergence of e-commerce with logistics service provides great convenience to people’s lives. However, platform usually receive seller claim for some reasons (e.g., damaged packages). Thus, it is important to predict seller claim risk in logistics scenarios. Existing solution for seller claim risk predict are challenging to address this problem due to two unique features including (i) multi-side collaborative risk factor caused by data sharing constraints of e-commerce and logistics platform, (ii) industry-specific risk factor caused by the dynamic similarity of sellers. To incorporate these new factors, we propose a novel seller claim risk prediction framework (Fed-SCRP) via federated multi-view learning, where we (i) design a federated multi-view learning to deal with data isolation problem, (ii) develop a STG-SRIM model and a series of information union transformers to capture the hybrid semantic embedding of dynamic seller features and industry-specific risk factor. We conduct a comprehensive evaluation of our method using datasets from major Chinese e-commerce and logistics platforms. Experimental results demonstrate that our method outperforms state-of-the-art baselines in various metrics.

Warehouse-distribution integration has been adopted by many e-commerce retailers (e.g., Amazon, TAOBAO, and JD) as an efficient business mode. In warehouse-distribution integration e-commerce, one of the most important problems is to estimate the full-link delivery time for better decision-making. Existing solutions for traditional warehouse-distribution separation mode are challenging to address this problem due to two unique features in the integration mode including (i) contextual influence caused by neighbor units in heterogeneous delivery networks, (ii) uncertain delivery time caused by the dynamic temporal data (e.g., online sales volume) and heterogeneity of delivery units. To incorporate these new factors, we propose Heterogeneous Spatial-Temporal Graph Transformer (HST-GT), a novel full-link delivery time estimation method under the warehouse-distribution integration mode, where we (i) develop heterogeneous graph transformers to capture hierarchical heterogeneous information; and (ii) design a set of spatial-temporal transformers based on heterogeneous features to fully exploit the correlation of spatial and temporal information. We extensively evaluate our method based on one-month real-world data consisting of hundreds of warehouses and sorting centers, and millions of historical orders collected from one of the largest e-commerce retailers in the world. Experimental results demonstrate that our method outperforms state-of-the-art baselines in various metrics.