Multi-view region embedding has become an important technique to be pursued due to the increasing availability of diverse urban sensing data. Existing methods typically adopt attention-based fusion or contrastive alignment to integrate multiple data sources such as human mobility and points-of-interest (POIs) into unified region representations. However, these approaches often fail to effectively balance cross-view commonality modeling with the preservation of view-specific characteristics. To address this limitation, we propose ComSRE, a novel region embedding framework that explicitly disentangles shared and distinctive components in multi-view region representations. ComSRE integrates three key modules: (1) a Multi-view Representation Learning module that fuses complementary information across views, (2) a View-to-Commonality Contrastive Alignment module that aligns the representation of each view to a shared commonality anchor to enhance cross-view consistency, and (3) a Multi-view Differential Orthogonality module that isolates distinctive signals unique to each view and promotes their independence through orthogonality constraints. Extensive experiments on three real-world urban datasets demonstrate that ComSRE consistently outperforms state-of-the-art methods across multiple downstream tasks, achieving superior predictive accuracy and representation quality.

Predicting human mobility remains a fundamental challenge, especially when individuals deviate from routine patterns due to exploration, disruptions, or rare events. While sequential models like Transformers excel at capturing regular movement patterns, their performance often degrades under nonroutine scenarios involving rare or unfamiliar transitions. To address this, we propose ROAM (Routine-Oriented Adaptive Mobility Predictor), a novel framework that jointly models human mobility from both sequential and relational perspectives, enabling adaptive handling of both routine and nonroutine behaviors during prediction. ROAM combines a sequential encoder that captures historically frequent transitions with a complementary graph-based relational reasoning module that encodes both user-specific and group-level mobility structures. To dynamically integrate these views, we introduce a hierarchical confidence-aware gating mechanism that adaptively balances sequential and relational predictions based on their internal reliability. Extensive experiments on real-world mobility datasets show that ROAM consistently outperforms state-of-the-art baselines in next location prediction. Further analysis reveals that the combination of sequential and relational reasoning substantially improves robustness, particularly under out-of-routine scenarios.

A recent trend in urban computing involves utilizing multi-modal data for urban region embedding, which can be further expanded in a variety of downstream urban sensing tasks. Many previous studies rely on multi-graph embedding techniques and follow a two-stage paradigm: first building a k-nearest neighbor graph based on fixed region correlations for each view, and then blending multi-view information in a posterior stage to learn region representations. However, multi-graph construction and multi-graph representation learning are not associated in most existing two-stage studies, and the relationship between them is not leveraged, which can provide complementary information to each other. In this paper, we unify these two stages into one by constructing learnable weighted complete graphs of regions and propose a new one-stage Region Embedding method with Adaptive region correlation Discovery (READ). Specifically, READ comprises three modules, including a disentangled region feature learning module utilizing a city-context Transformer to encode regions' semantic and mobility features, and an adaptive weighted multi-graph construction module that builds multiple complete graphs with learnable weights based on disentangled features of regions. In addition, we propose a multi-graph representation learning module to yield effective region representations that integrate information from multiple graphs. We conduct thorough experiments on three downstream tasks to assess READ. Experimental results demonstrate that READ considerably outperforms state-of-the-art baseline methods in urban region embedding.

Learning urban region embeddings has substantially advanced urban analysis, but their typical focus on individual cities leads to disparate embedding spaces, hindering cross-city knowledge transfer and the reuse of downstream task predictors. To tackle this issue, we present Consistency Region Embedding (CoRE), a unified framework integrating region embedding learning with cross-city latent space alignment. CoRE first embeds regions from two cities into separate latent spaces, followed by the alignment of latent space manifolds and fine-grained individual regions from both cities. This ensures compatible and comparable embeddings within aligned latent spaces, enabling predictions of various socioeconomic indicators without ground truth labels by migrating knowledge from label-rich cities. Extensive experiments show CoRE outperforms competitive baselines, confirming its effectiveness for cross-city knowledge transfer via aligned latent spaces.

Image Super-Resolution (ISR) has seen significant progress with the introduction of remarkable generative models. However, challenges such as the trade-off issues between fidelity and realism, as well as computational complexity, have also posed limitations on their application. Building upon the tremendous success of autoregressive models in the language domain, we propose VARSR, a novel visual autoregressive modeling for ISR framework with the form of next-scale prediction. To effectively integrate and preserve semantic information in low-resolution images, we propose using prefix tokens to incorporate the condition. Scale-aligned Rotary Positional Encodings are introduced to capture spatial structures and the diffusion refiner is utilized for modeling quantization residual loss to achieve pixel-level fidelity. Image-based Classifier-free Guidance is proposed to guide the generation of more realistic images. Furthermore, we collect large-scale data and design a training process to obtain robust generative priors. Quantitative and qualitative results show that VARSR is capable of generating high-fidelity and high-realism images with more efficiency than diffusion-based methods. Our codes are released at https://github.com/quyp2000/VARSR.

Next location prediction is a critical task in human mobility modeling, with broad applications in personalized recommendation, urban planning, and location-based services. Recently, researchers have used prompt-based large language models (LLMs) to improve next location prediction with pre-trained knowledge. However, they face inherent challenges in bridging the gap between textual prompts for semantic contextual understanding and human mobility data for transition pattern modeling. In this paper, we introduce SILO, a framework designed for Semantic Integration in LOcation prediction via LLMs. We first construct a hybrid semantic space that seamlessly integrates ID-based embeddings, text-derived semantics, and auxiliary contextual information, enabling comprehensive modeling of sequential mobility patterns alongside contextual nuances. We then propose user-centric prompts that specify the prediction task for LLMs while embedding user context within a special token. Further, we utilize LLMs as the prediction backbone to process both user-specific prompts and hybrid ID-context embeddings of location sequences. To enhance predictive performance, we finally introduce a dual-logits strategy, combining sequential transition logits with user profile-guided semantic preference logits. Extensive experiments on two large-scale real-world mobility datasets demonstrate that SILO significantly outperforms state-of-the-art baselines, validating its effectiveness in modeling complex mobility patterns through semantic integration using LLMs.

As a crucial component of intelligent transportation systems (ITS), accurate traffic flow prediction has attracted considerable attention. Numerous popular models, such as recurrent neural networks (RNN) and graph convolution networks (GCN), have been extensively applied to this task. However, due to the constraints of the topology of urban road network and the law of dynamic change with time, the single-layer network model cannot capture the dynamic correlation between the spatial–temporal and the traditional road network characteristics for traffic flow. In this article, we propose a spatial–temporal twolayer synchronous network (ST2SNet) for traffic flow prediction. First, we introduce a novel network structure comprising a spatial–temporal fusion layer and a situation awareness layer, where temporal and spatial correlations are fused synchronously using a transformer architecture. Next, we incorporate the residual gated GCN structure and a 2-D convolution network to effectively integrate road network information and temporal trends for spatial–temporal traffic flow prediction. Finally, we evaluate ST2SNet on two benchmark datasets, PeMSD4, and PeMSD8. Compared to baseline methods, our model achieves MAE improvements ranging from 4.00% to 48.30% on PeMSD4 and from 4.88%to 54.13%on PeMSD8. These outstanding experimental results demonstrate that ST2SNet significantly enhances prediction performance.

Recently, learning urban region representations utilizing multi-modal data (information views) has become increasingly popular, for deep understanding of the distributions of various socioeconomic features in cities. However, previous methods usually blend multi-view information in a posteriors stage, falling short in learning coherent and consistent representations across different views. In this paper, we form a new pipeline to learn consistent representations across varying views, and propose the multi-view Contrastive Prediction model for urban Region embedding (ReCP), which leverages the multiple information views from point-of-interest (POI) and human mobility data. Specifically, ReCP comprises two major modules, namely an intra-view learning module utilizing contrastive learning and feature reconstruction to capture the unique information from each single view, and inter-view learning module that perceives the consistency between the two views using a contrastive prediction learning scheme. We conduct thorough experiments on two downstream tasks to assess the proposed model, i.e., land use clustering and region popularity prediction. The experimental results demonstrate that our model outperforms state-of-the-art baseline methods significantly in urban region representation learning.

Next location prediction is a crucial task in human mobility modeling, and is pivotal for many downstream applications like location-based recommendation and transportation planning. Although there has been a large body of research tackling this problem, the usefulness of user preference and temporal regularity remains underrepresented. Specifically, previous studies usually neglect the explicit user preference information entailed from human trajectories and fall short in utilizing the arrival time of next location, as a key determinant on next location. To address these limitations, we propose a Multi-Context aware Location Prediction model (MCLP) to predict next locations for individuals, where it explicitly models user preference and the next arrival time as context. First, we utilize a topic model to extract user preferences for different types of locations from historical human trajectories. Second, we develop an arrival time estimator to construct a robust arrival time embedding based on the multi-head attention mechanism. The two components provide pivotal contextual information for the subsequent prediction. Finally, we utilize the Transformer architecture to mine sequential patterns and integrate multiple contextual information to predict the next locations. Experimental results on two real-world mobility datasets show that our proposed MCLP outperforms baseline methods.

Human mobility flow prediction forecasts the number of passengers coming into (inflows) or leaving from (outflows) every region of a city. It is crucial for many applications such as mobile marketing or route optimization. People have proposed deep learning methods to predict human mobility flows. However, existing methods neglect the hierarchical nature of human mobility behaviors. Each of us is unique, and we live beyond our neighborhoods. In the process of cross-regional activities, humans migrate in the hierarchical structure of buildings, neighborhoods, regions, cities, and countries. In this article, we propose a predictive framework, hierarchical fuzzy C-means (Hierarchical-FCM)-residual networks (ResNets), to capture the hierarchical structure of human mobility for prediction. First, we offer a Hierarchical-FCM clustering algorithm that is trained to learn the relationship between proximity and road network hierarchically on large human mobility data. Second, we design a fusion model to incorporate the knowledge learned from hierarchical clusters into deep ResNets to improve prediction accuracy. We compare our framework with 25 existing state-of-the-art models, ranging from traditional time-series and machine learning predictors, such as ARIMA and RNN, to the latest deep learning methods designed for human mobility prediction, such as ST-ResNets and DeepST. Our framework outperforms all existing models, by a margin of 2%–68% in terms of prediction accuracy, showing its effectiveness. We are among the first to incorporate the hierarchical structure of human mobility into location clustering for human mobility flow prediction. We empirically show that incorporating such knowledge significantly improves prediction performance.

A crucial area of research in Human-AI Interaction focuses on understanding how the integration of AI into social systems influences human behavior, for example, how news-feeding algorithms affect people’s voting decisions. But little attention has been paid to how human behavior shapes AI performance. We fill this research gap by introducingroutinenessto measure human behavior for the AI system, which assesses the degree of routine in a person’s activity based on their past activities. We apply the proposedroutinenessmetric to two extensive human behavior datasets: the human mobility dataset with over 700 million data samples and the social media dataset with over 3.8 million data samples. Our analysis revealsroutinenesscan effectively detect behavioral changes in human activities. The performance of AI algorithms is profoundly determined by humanroutineness, which provides valuable guidance for the selection of AI algorithms.

Human mobility prediction is crucial for epidemic control, urban planning, and traffic forecasting systems. We observe urban traffic flow prediction has a hierarchical structure, in which human mobility prediction should consider not only the spatial and the temporal relationships, but also the high-level mobility trend between individuals and regions. In this paper, we propose a human mobility clustering algorithm based on trend iteration of spectral clustering (TISC) to incorporate the high-level human mobility trend between individuals and regions. We integrate our TISC clustering algorithm with two existing urban traffic flow predictive models: namely, deep spatio-temporal residual network (ST-ResNet) and deep spatio-temporal 3D network (ST-3DNet). By adapting our TISC clustering algorithm, the prediction accuracy of both algorithms has been improved significantly (30.96$\%$for ST-ResNet and 24.66$\%$for ST-3DNet). We also compare the TISC-based predictive framework with 26 state-of-the-art human mobility prediction algorithms. We observe that our TISC algorithm considerably outperforms all 26 methods, reducing the predictive error from 6.93% to 69.55$\%$.

Recent work on knowledge graph completion (KGC) focuses on acquiring embeddings of entities and relations in knowledge graphs. These embedding methods necessitate that all test entities be present during the training phase, resulting in a time-consuming retraining process for out-of-knowledge-graph (OOKG) entities. To tackle this predicament, current inductive methods employ graph neural networks (GNNs) to represent unseen entities by aggregating information of the known neighbors, and enhance the performance with additional information, such as attention mechanisms or logic rules. Nonetheless, Two key challenges continue to persist: (i) identifying inter-rule correlations to further facilitate the inference process, and (ii) capturing interactions among rule mining, rule inference, and embedding to enhance both rule and embedding learning.

Blind image quality assessment (BIQA) aims to auto-matically evaluate the perceived quality of a single image, whose performance has been improved by deep learning-based methods in recent years. However, the paucity of labeled data somewhat restrains deep learning-based BIQA methods from unleashing their full potential. In this paper, we propose to solve the problem by a pretext task customized for BIQA in a self-supervised learning manner, which enables learning representations from orders of mag-nitude more data. To constrain the learning process, we propose a quality-aware contrastive loss based on a simple assumption: the quality of patches from a distorted image should be similar, but vary from patches from the same image with different degradations and patches from different images. Further, we improve the existing degradation process and form a degradation space with the size of roughly 2 × 107. After pretrained on ImageNet using our method, models are more sensitive to image quality and perform significantly better on downstream BIQA tasks. Experimental results show that our method obtains remarkable improvements on popular BIQA datasets.

Categories of Point of Interest (POI) facilitate location-based services from many aspects like location search and POI recommendation. However, POI categories are often incomplete and new POIs are being consistently generated, this rises the demand for semantic annotation for POIs, i.e., labeling the POI with a semantic category. Previous methods usually model sequential check-in information of users to learn POI features for annotation. However, users' check-ins are hardly obtained in reality, especially for those newly created POIs. In this context, we present a Spatial-Textual POI Annotation (STPA) model for static POIs, which derives POI categories using only the geographic locations and names of POIs. Specifically, we design a GCN-based spatial encoder to model spatial correlations among POIs to generate POI spatial embeddings, and an attention-based text encoder to model the semantic contexts of POIs to generate POI textual embeddings. We finally fuse the two embeddings and preserve multi-view correlations for semantic annotation. We conduct comprehensive experiments to validate the effectiveness of STPA with POI data from AMap. Experimental results demonstrate that STPA substantially outperforms several competitive baselines, which proves that STPA is a promising approach for annotating static POIs in map services.

Multivariate time series forecasting has wide applications such as traffic flow prediction, supermarket commodity demand forecasting and etc., and a large number of forecasting models have been developed. Given these models, a natural question has been raised: what theoretical limits of forecasting accuracy can these models achieve? Recent works of urban human mobility prediction have made progress on the maximum predictability that any algorithm can achieve. However, existing approaches on maximum predictability on the multivariate time series fully ignore the interrelationship between multiple variables. In this article, we propose a methodology to measure the upper limit of predictability for multivariate time series with multivariate constraint relations. The key of the proposed methodology is a novel entropy, named Multivariate Constraint Sample Entropy (McSE), to incorporate the multivariate constraint relations for better predictability. We conduct a systematic evaluation over eight datasets and compare existing methods with our proposed predictability and find that we get a higher predictability. We also find that the forecasting algorithms that capture the multivariate constraint relation information, such as GNN, can achieve higher accuracy, confirming the importance of multivariate constraint relations for predictability.

Urban human mobility prediction is forecasting how people move in cities. It is crucial for many smart city applications including route optimization, preparing for dramatic shifts in modes of transportation, or mitigating the epidemic spread of viruses such as COVID-19. Previous research propose the maximum predictability to derive the theoretical limits of accuracy that any predictive algorithm could achieve on predicting urban human mobility. However, existing maximum predictability only considers the sequential patterns of human movements and neglects the contextual information such as the time or the types of places that people visit, which plays an important role in predicting one's next location. In this paper, we propose new theoretical limits of predictability, namely Context-Transition Predictability, which not only captures the sequential patterns of human mobility, but also considers the contextual information of human behavior. We compare our Context-Transition Predictability with other kinds of predictability and find that it is larger than these existing ones. We also show that our proposed Context-Transition Predictability provides us a better guidance on which predictive algorithm to be used for forecasting the next location when considering the contextual information. Source code is at https://github.com/zcfinal/ContextTransitionPredictability.

Time series prediction has wide applications ranging from stock price prediction, product demand estimation to economic forecasting. In this article, we treat the taxi and Uber demand in each location as a time series, and reduce the taxi and Uber demand prediction problem to a time series prediction problem. We answer two key questions in this area. First, time series have different temporal regularity. Some are easy to be predicted and others are not. Given a predictive algorithm such as LSTM (deep learning) or ARIMA (time series), what is the maximum prediction accuracy that it can reach if it captures all the temporal patterns of that time series? Second, given the maximum predictability, which algorithm could approach the upper bound in terms of prediction accuracy? To answer these two question, we use temporal-correlated entropy to measure the time series regularity and obtain the maximum predictability. Testing with 14 million data samples, we find that the deep learning algorithm is not always the best algorithm for prediction. When the time series has a high predictability a simple Markov prediction algorithm (training time 0.5s) could outperform a deep learning algorithm (training time 6 hours). The predictability can help determine which predictor to use in terms of the accuracy and computational costs. We also find that the Uber demand is easier to be predicted compared the taxi demand due to different cruising strategies as the former is demand driven with higher temporal regularity.

Multivariate time series forecasting has wide applications such as traffic flow prediction, supermarket commodity demand forecasting and etc. In literature, Due to the complex temporal patterns and inter-dependencies among multivariate time series, a large number of forecasting models have been developed. However, one question still remains unclear: how these models perform on a certain forecasting task, and there is lack of comprehensive performance comparison of these models on different tasks. To this end, in this paper, we conduct a systematic evaluation of eight representative forecasting models over eight multivariate time series datasets, and have the following findings: 1) When the datasets exhibit strong periodic patterns, deep learning models perform best. Otherwise on the datasets in a non-periodic manner, the statistical models such as ARIMA perform best. 2) For the long term prediction involving a high horizon value, the direct prediction strategy could lead to lower errors than the recursive one, but at the cost of higher training time. 3) For the multivariate time series explicitly involving graphic inter-dependencies among the multivariates, e.g., the road network topology in the spatio-temporal time series of traffic volumes in multiple routes, the Graph Convolution Network can incorporate the graphic inter-dependencies into their forecasting models for smaller prediction errors.

Recent years have witnessed unprecedented amounts of telecommunication (Telco) data generated by Telco radio and core equipment. For example, measurement records (MRs) are generated to report the connection states, e.g., received signal strength at the mobile device, when mobile devices give phone calls or access data services. Telco historical data (e.g., MRs) have been widely analyzed to understand human mobility and optimize the applications such as urban planning and traffic forecasting. The key of these applications is to precisely localize outdoor mobile devices from these historical MR data. Previous works calculate the location of a mobile device based on each single MR sample, ignoring the sequential and temporal locality hidden in the consecutive MR samples. To address the issue, we propose a deep neural network (DNN)-based localization framework namely DeepLoc to ensemble a recently popular sequence learning model LSTM and a CNN. Without skillful feature design and post-processing steps, DeepLoc can generate a smooth trajectory consisting of accurately predicted locations. Extensive evaluation on 6 datasets collected at three representative areas (core business, urban and suburban areas in Shanghai, China) indicates that DeepLoc greatly outperforms 10 counterparts.

We contribute by developing GeoMatch as a novel, scalable, and efficient big-data pipeline for large-scale map matching on Apache Spark. GeoMatch improves existing spatial big data solutions by utilizing a novel spatial partitioning scheme inspired by Hilbert space-filling curves. Thanks to the partitioning scheme, GeoMatch can effectively balance operations across different processing units and achieve significant performance gains. We demonstrate the effectiveness of GeoMatch through rigorous and extensive benchmarks that consider data sets containing large-scale urban spatial data sets ranging from 166, 253 to 3.78 billion location measurements. Our results show over 17-fold performance improvements compared to previous works while achieving better processing accuracy than current solutions (97.48%).

Cities are inherently dynamic. Interesting patterns of behavior typically manifest at several key areas of a city over multiple temporal resolutions. Studying these patterns can greatly help a variety of experts ranging from city planners and architects to human behavioral experts. Recent technological innovations have enabled the collection of enormous amounts of data that can help in these studies. However, techniques using these data sets typically focus on understanding the data in the context of the city, thus failing to capture the dynamic aspects of the city. The goal of this work is to instead understand the city in the context of multiple urban data sets. To do so, we define the concept of an "urban pulse" which captures the spatio-temporal activity in a city across multiple temporal resolutions. The prominent pulses in a city are obtained using the topology of the data sets, and are characterized as a set of beats. The beats are then used to analyze and compare different pulses. We also design a visual exploration framework that allows users to explore the pulses within and across multiple cities under different conditions. Finally, we present three case studies carried out by experts from two different domains that demonstrate the utility of our framework.

In big cities, taxi service is imbalanced. In some areas, passengers wait too long for a taxi, while in others, many taxis roam without passengers. Knowledge of where a taxi will become available can help us solve the taxi demand imbalance problem. In this paper, we employ a holistic approach to predict taxi demand at high spatial resolution. We showcase our techniques using two real-world data sets, yellow cabs and Uber trips in New York City, and perform an evaluation over 9,940 building blocks in Manhattan. Our approach consists of two key steps. First, we use entropy and the temporal correlation of human mobility to measure the demand uncertainty at the building block level. Second, to identify which predictive algorithm can approach the theoretical maximum predictability, we implement and compare three predictors: the Markov predictor (a probability-based predictive algorithm), the Lempel-Ziv-Welch predictor (a sequence-based predictive algorithm), and the Neural Network predictor (a predictive algorithm that uses machine learning). The results show that predictability varies by building block and, on average, the theoretical maximum predictability can be as high as 83%. The performance of the predictors also vary: the Neural Network predictor provides better accuracy for blocks with low predictability, and the Markov predictor provides better accuracy for blocks with high predictability. In blocks with high maximum predictability, the Markov predictor is able to predict the taxi demand with an 89% accuracy, 11% better than the Neural Network predictor, while requiring only 0.03% computation time. These findings indicate that the maximum predictability can be a good metric for selecting prediction algorithms.

Understanding urban human mobility is crucial for epidemic control, urban planning, traffic forecasting systems and, more recently, various mobile and network applications. Nowadays, a variety of urban human mobility data have been gathered and published. Pervasive GPS data can be collected by mobile phones. A mobile operator can track people's movement in cities based on their cellular network location. This urban human mobility data contains rich knowledge about locations and can help in addressing many urban challenges such as traffic congestion or air pollution problems. In this article, we survey recent literature on urban human mobility from a data mining view: from the data collection and cleaning, to the mobility models and the applications. First, we summarize recent public urban human mobility data sets and how to clean and preprocess such data. Second, we describe recent urban human mobility models and predictors, e.g., the deep learning predictor, for predicting urban human mobility. Third, we describe how to evaluate the models and predictors. We conclude by considering how applications can utilize the mobility models and predictive tools for addressing city challenges.

Many applications, such as product promotion advertisement and traffic congestion notification, benefit from opportunistic content exchange in Delay Tolerant Networks (DTNs). An important requirement of such applications is timely delivery. However, the intermittent connectivity of DTNs may significantly delay content exchange, and cannot guarantee timely delivery. The state-of-the-arts capture mobility patterns or social properties of mobile devices. Such solutions do not capture patterns of delivered content in order to optimize content delivery. Without such optimization, the content demanded by a large number of subscribers could follow the same forwarding path as the content by only one subscriber, leading to traffic congestion and packet drop. To address the challenge, in this paper, we develop a solution framework, namely Ameba, for timely delivery. In detail, we first leverage content properties to derive an optimal routing hop count of each content to maximize the number of needed nodes. Next, we develop node utilities to capture interests, capacity and locations of mobile devices. Finally, the distributed forwarding scheme leverages the optimal routing hop count and node utilities to deliver content towards the needed nodes in a timely manner. Illustrative results verify that Ameba achieves comparable delivery ratio as Epidemic but with much lower overhead.

We introduce CoSense, a collaborative sensing platform for mobile devices that opportunistically distributes sensing tasks between familiar devices in close proximity. We use empirical energy measurements together with data collected from everyday transportation behaviour to demonstrate that our solution can significantly reduce power consumption while maintaining the best possible sensing accuracy.

Many applications, such as product promotion advertisement and traffic congestion notification, benefit from the opportunistic content exchange in Delay Tolerant Networks (DTNs). An important requirement of such applications is timely delivery. However, the intermittent connectivity of DTNs may significantly delay content exchange and cannot guarantee timely delivery. The state-of-the-arts capture the mobility patterns or social properties of mobile devices. However, there is little optimization in terms of the delivered content. Without such optimization, the content demanded by a large number of subscribers could follow the same forwarding path as the content by only one subscriber. To address the challenge, in this paper, we separate content routing from content forwarding. For content routing, we leverage content properties to derive an optimal routing hop count for each content in order to maximize the number of nodes which receive demanded content. Next, for timely forwarding, we develop node utilities to capture interests and mobility patterns of mobile devices for the selection of content carriers. The distributed greedy relay scheme, Ameba, leverages the optimal routing hop count and developed utilities to timely relay content to the needed nodes as fast as possible. Illustrative results show that Ameba is able to achieve comparable delivery ratio as the Epidemic but with much lower overhead.