With the popularity of smart mobile devices, location-based services (LBS) have been widely applied. Predicting geographical locations from text holds significant value for smart cities and personalized travel. Existing research primarily focuses on the retrieval or prediction of labeled locations, such as cities or points of interest (POIs). However, in scenarios like autonomous driving navigation and autonomous logistics delivery, it is necessary to precisely predict the coordinates of unlabeled locations, for example, 200 meters northwest of a certain location. Consequently, we introduce a new task to infer fine-grained unlabeled locations from text. This task is particularly challenging because of the ambiguous text and the semantic gap between geographic and textual modalities. In this paper, we aim to construct an end-to-end fine-grained location prediction model to accurately predict the unlabeled locations mentioned in texts. First, we encode the geographic coordinates and transform the location prediction problem into a geographic encoding generation problem. Second, we propose a multi-scale cross-modal loss (MCL) to learn the implicit mapping between geographic and textual modalities. Lastly, we design a multi-task prediction model ULP to predict the coordinates of unlabeled locations. We conducted experiments on two real-world datasets, and the results show that our proposed method outperforms existing state-of-the-art retrieval-based methods.

The next point of interest (POI) recommendation task recommends POIs to users that they may be interested in next time based on their historical trajectories. This task holds value for both users and businesses. However, it has consistently faced the issue of cold-start caused by sparse user check-in data. Existing research mainly focuses on knowledge transfer among cities within the same data source, but these data are very rare. The abundance of available third-party data presents opportunities to improve cold-start performance, but it is not easy. This third-party data contain numerous entities, such as POIs and users, which have different representations and distributions across different data domains, making knowledge transfer difficult. To address these challenges, we propose the Tree-Based Cross-domain Knowledge Transfer (TCKT) model. First, we construct a multi-granularity Geographical Frequency Tree (GF-Tree), transforming the POI recommendation problem into a path generation problem. Second, we design a pre-training model to mine general user behavior patterns and spatio-temporal features among POIs from large-scale third-party data. Finally, we propose a dual-channel domain adaptation model to facilitate cross-domain knowledge transfer and improve cold-start performance. Experimental results on three public datasets demonstrate that our method outperforms state-of-the-art (SOTA) baseline methods.

This study is inspired by the current image restoration technology. If we regard the users participating in the rumor as image pixels, similar to social networks, the recovery of pixel data is affected by the pixels themselves and neighbor pixels, then the prediction of user behavior in the rumor diffusion can be regarded as the process of image restoration for pixel-blurred user behavior images. We first propose a diffusion2pixel algorithm that transforms the user relationship network of topic diffusion into image pixel matrix. To cope with the diversity and complexity of the diffusion feature space, the user relationship network is reduced to a low-rank dense vectorization by representation learning before being pixelated by cutting and diffusion. Second, considering the competitive relationship between rumor and anti-rumor, transition matrix of rumor mutual influences is established by evolutionary game theory. A mutual influence model of rumor and anti-rumor is then proposed. Finally, we combine the transition matrix of rumor mutual influence into a simple prediction method Graph-CNN of rumor and anti-rumor topic diffusion based on dynamic iteration mechanism. Experiments confirmed the proposed model can effectively predict the group diffusion trends of rumor, and reflects the competitive relationship between rumor and anti-rumor.

With the increasing network service content, innovative methods are required for developing optimized network service for e-commerce companies. Accordingly, this study focuses on designing a framework containing personalization, interest group identification, and recommendation mechanisms. The primary contribution of this paper is to propose a recommendation model based on data compensation and dynamic user interest grouping. First, to address the problem of sparse user rating data, homeostasis compensation is performed on native data to more realistically restore the preference relationship between users and items by introducing the advantages of generative adversarial network in learning data distribution and enhancing data samples. Second, to address the problem of user interest generalization, information entropy is introduced to measure the user interest feature space. In addition, the time window marking method is used to further quantify the users’ dynamic interest group around the users’ interest drift. Finally, considering tensor decomposition characteristics in data dimension transformation and data compression, a score prediction model based on the “user-item-interest group” tensor decomposition is constructed. Simultaneously, a time decay function is introduced in the construction of the tensor to dynamically fit the user behavior and further improve prediction accuracy. Experiments show that the proposed framework can effectively improve the recommendation accuracies resulting from both sparse scoring data and dynamic user interest division.