As privacy concerns continue to grow, federated learning (FL) has gained significant attention as a promising privacy-preserving technology, leading to considerable advancements in recent years. Unlike traditional machine learning, which requires central data collection, FL keeps data localized on user devices. However, conventional FL assumes that all clients operate with identical model structures initialized by the server. In real-world applications, system heterogeneity is common, with clients possessing varying computational capabilities. This disparity can hinder training for resource-limited clients and result in inefficient resource use for those with greater processing power. To address this challenge, model-heterogeneous FL has been introduced, enabling clients to train models of varying complexity based on their hardware resources. This paper reviews state-of-the-art approaches in model-heterogeneous FL, analyzing their strengths and weaknesses, while identifying open challenges and future research directions. To the best of our knowledge, this is the first survey to specifically focus on model-heterogeneous FL.

Abstract Spark is one of the most popular big data analytical platforms. To save time, achieve high resource utilization, and remain cost-effective for Spark jobs, it is challenging but imperative for data scientists to configure suitable resource portions.In this paper, we investigate the proper parameter values that meet workloads’ performance requirements with minimized resource cost and resource utilization time. We propose SimCost , a simulation-based cost model, to predict the performance of jobs accurately. We achieve low-cost training by taking advantage of simulation framework , i.e., Monte Carlo simulation, which uses a small amount of data and resources to make a reliable prediction for larger datasets and clusters. Our method’s salient feature is that it allows us to invest low training costs while obtaining an accurate prediction. Through empirical experiments with 12 benchmark workloads, we show that the cost model yields less than 5% error on average prediction accuracy, and the recommendation achieves up to 6x resource cost saving.

Mobile apps have become an indispensable part of people’s daily lives. Users determine what apps to use and when and where to use them based on their tastes, interests, and personal demands, depending on their personality traits. This article aims to infer user profiles from their spatiotemporal mobile app usage behavior. Specifically, we first transform mobile app usage records into a heterogeneous graph. On the graph, nodes represent users, apps, locations, and time slots. Edges describe the co-occurrence of entities in usage records. We then develop a multi-relational heterogeneous graph attention network (MRel-HGAN), an end-to-end system for user profiling. MRel-HGAN first adopts a neighbor sampling strategy based on bootstrapping to sample heavily connected neighbors of a fixed size for each node. Next, we design a relational graph convolutional operation and a multi-relational attention operation. Through such modules, MRel-HGAN can generate node embedding by sufficiently leveraging the rich semantic information of the multi-relational structure in the mobile app usage graph. Experimental results on real-world mobile app usage datasets show the effectiveness and superiority of our MRel-HGAN in the user profiling task for attributes of gender and age.

Satellite imagery depicts the Earth’s surface remotely and provides comprehensive information for many applications, such as land use monitoring and urban planning. Existing studies on unsupervised representation learning for satellite images only take into account the images’ geographic information, ignoring human activity factors. To bridge this gap, we propose using the Point-of-Interest (POI) data to capture human factors and designing a contrastive learning-based framework to consolidate the representation of satellite imagery with POI information. Besides, we introduce a season-invariant representation learning model on satellite imagery, considering that human factors are mostly unchanging with respect to seasons. An attention model is designed at last to merge the representations from the geographic, seasonal, and POI perspectives adaptively. On the basis of real-world datasets collected from Beijing, 1 we evaluate our method for predicting socioeconomic indicators. The results show that the representation containing POI information outperforms the geographic representation in estimating commercial activity-related indicators. Our proposed attentional framework can estimate the socioeconomic indicators with R 2 of 0.874 and outperforms the baseline methods. Furthermore, we explore the differences in the representations of satellite images with varying socioeconomic statuses. Finally, we investigate the impact of geographic and POI perspective information in the representation learning process, as well as the effect of satellite imagery on various spatial resolutions.

Understanding economic development and designing government policies requires accurate and timely measurements of socioeconomic activities. In this paper, we show how to leverage city structural information and urban imagery like satellite images and street view images to accurately predict multi-level socioeconomic indicators. Our framework consists of four steps. First, we extract structural information from cities by transforming real-world street networks into city graphs (GeoStruct). Second, we design a contrastive learning-based model to refine urban image features by looking at geographic similarity between images, with images that are geographically close together having similar features (GeoCLR). Third, we propose using street segments as containers to adaptively fuse the features of multi-view urban images, including satellite images and street view images (GeoFuse). Finally, given the city graph with a street segment as a node and a neighborhood area as a subgraph, we jointly model street- and neighborhood-level socioeconomic indicator predictions as node and subgraph classification tasks. The novelty of our method is that we introduce city structure to organize multi-view urban images and model the relationships between socioeconomic indicators at different levels. We evaluate our framework on the basis of real-world datasets collected in multiple cities. Our proposed framework improves performance by over 10% when compared to state-of-the-art baselines in terms of prediction accuracy and recall.

Satellite imagery depicts the earth’s surface remotely and provides comprehensive information for many applications, such as land use monitoring and urban planning. Existing studies on unsupervised representation learning for satellite images only take into account the images’ geographic information, ignoring human activity factors. To bridge this gap, we propose using Point-of-Interest (POI) data to capture human factors and design a contrastive learning-based framework to consolidate the representation of satellite imagery with POI information. Also, we design an attention model that merges the representations from the geographic and POI perspectives adaptively. On the basis of real-world datasets collected from Beijing, we evaluate our method for predicting socioeconomic indicators. The results show that the representation containing POI information outperforms the geographic representation in estimating commercial activity-related indicators. Our proposed framework can estimate the socioeconomic indicators with an R2 of 0.874 and outperforms the baseline methods.

The prevalence of smartphones has promoted the popularity of mobile apps in recent years. Although significant effort has been made to understand mobile app usage, existing studies are based primarily on short-term datasets with limited time span, e.g., a few months. Therefore, many basic facts about the long-term evolution of mobile app usage are unknown. In this paper, we study how mobile app usage evolves over a long-term period. We first introduce an app usage collection platform named carat, from which we have gathered app usage records of 1,465 users from 2012 to 2017. We then conduct the first study on the long-term evolution processes on a macro-level, i.e., app-category, and micro-level, i.e., individual app. We discover that, on both levels, there is a growth stage enabled by the introduction of new technologies. Then there is a plateau stage caused by high correlations between app categories and a pareto effect in individual app usage, respectively. Additionally, the evolution of individual app usage undergoes an elimination stage due to fierce intra-category competition. Nevertheless, the diverseness of app-category and individual app usage exhibit opposing trends: app-category usage assimilates while individual app usage diversifies. Our study provides useful implications for app developers, market intermediaries, and service providers.

Bloom Filter is a space-efficient probabilistic data structure for checking the membership of elements in a set. Given multiple sets, a standard Bloom Filter is not sufficient when looking for the items to which an element or a set of input elements belong. An example case is searching for documents with keywords in a large text corpus, which is essentially a multiple set matching problem where the input is single or multiple keywords, and the result is a set of possible candidate documents. This article solves the multiple set matching problem by proposing two efficient Bloom Multifilters called Bloom Matrix and Bloom Vector, which generalize the standard Bloom Filter. Both structures are space-efficient and answer queries with a set of identifiers for multiple set matching problems. The space efficiency can be optimized according to the distribution of labels among multiple sets: Uniform and Zipf. Bloom Vector efficiently exploits the Zipf distribution of data for further space reduction. Indeed, both structures are much more space-efficient compared with the state-of-the-art, Bloofi. The results also highlight that a L ookup operation on Bloom Matrix is significantly faster than on Bloom Vector and Bloofi.

Spark is one of the prevalent big data analytical platforms. Configuring proper resource provision for Spark jobs is challenging but essential for organizations to save time, achieve high resource utilization, and remain cost-effective. In this paper, we study the challenge of determining the proper parameter values that meet the performance requirements of workloads while minimizing both resource cost and resource utilization time. We propose a simulation-based cost model to predict the performance of jobs accurately. We achieve low-cost training by taking advantage of simulation framework, i.e., Monte Carlo (MC) simulation, which uses a small amount of data and resources to make a reliable prediction for larger datasets and clusters. The salient feature of our method is that it allows us to invest low training cost while obtaining an accurate prediction. Through experiments with six benchmark workloads, we demonstrate that the cost model yields less than 7% error on average prediction accuracy and the recommendation achieves up to 5x resource cost saving.

We contribute by quantifying the effect of network latency and battery consumption on mobile app performance and retention, i.e., user's decisions to continue or stop using apps. We perform our analysis by fusing two large-scale crowdsensed datasets collected by piggybacking on information captured by mobile apps. We find that app performance has an impact in its retention rate. Our results demonstrate that high energy consumption and high latency decrease the likelihood of retaining an app. Conversely, we show that reducing latency or energy consumption does not guarantee higher likelihood of retention as long as they are within reasonable standards of performance. However, we also demonstrate that what is considered reasonable depends on what users have been accustomed to, with device and network characteristics, and app category playing a role. As our second contribution, we develop a model for predicting retention based on performance metrics. We demonstrate the benefits of our model through empirical benchmarks which show that our model not only predicts retention accurately, but generalizes well across application categories, locations and other factors moderating the effect of performance.

Popularity of mobile apps is traditionally measured by metrics such as the number of downloads, installations, or user ratings. A problem with these measures is that they reflect usage only indirectly. Indeed, retention rates, i.e., the number of days users continue to interact with an installed app, have been suggested to predict successful app lifecycles. We conduct the first independent and large-scale study of retention rates and usage trends on a dataset of app-usage data from a community of 339,842 users and more than 213,667 apps. Our analysis shows that, on average, applications lose 65% of their users in the first week, while very popular applications (top 100) lose only 35%. It also reveals, however, that many applications have more complex usage behaviour patterns due to seasonality, marketing, or other factors. To capture such effects, we develop a novel app-usage trend measure which provides instantaneous information about the popularity of an application. Analysis of our data using this trend filter shows that roughly 40% of all apps never gain more than a handful of users ( Marginal apps). Less than 0.1% of the remaining 60% are constantly popular ( Dominant apps), 1% have a quick drain of usage after an initial steep rise ( Expired apps), and 6% continuously rise in popularity ( Hot apps). From these, we can distinguish, for instance, trendsetters from copycat apps. We conclude by demonstrating that usage behaviour trend information can be used to develop better mobile app recommendations.

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%).

The Carat project started in 2012 has collected over 1.5 TB of data from over 850,000 mobile users all over the world. The project uses Apache Thrift to transmit data, and Apache Spark to run data analysis tasks, and the gist of the Carat analysis method has been published. While the Carat application code is open source, the data is much harder to share because of its size and privacy concerns. This paper outlines the challenges in sharing such a large-scale dataset with detailed information about smart devices, applications, and their users, and presents some solutions to these challenges.

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.

We propose crowd replication as a low-effort, easy to implement and cost-effective mechanism for quantifying the uses, activities, and sociability of public spaces. Crowd replication combines mobile sensing, direct observation, and mathematical modeling to enable resource efficient and accurate quantification of public spaces. The core idea behind crowd replication is to instrument the researcher investigating a public space with sensors embedded on commodity devices and to engage him/her into imitation of people using the space. By combining the collected sensor data with a direct observations and population model, individual sensor traces can be generalized to capture the behavior of a larger population. We validate the use of crowd replication as a data collection mechanism through a field study conducted within an exemplary metropolitan urban space. Results of our evaluation show that crowd replication accurately captures real human dynamics (0.914 correlation between indicators estimated from crowd replication and visual surveillance) and captures data that is representative of the behavior of people within the public space.

There is little information from independent sources in the public domain about mobile malware infection rates. The only previous independent estimate (0.0009%) [11], was based on indirect measurements obtained from domain-name resolution traces. In this paper, we present the first independent study of malware infection rates and associated risk factors using data collected directly from over 55,000 Android devices. We find that the malware infection rates in Android devices estimated using two malware datasets (0.28% and 0.26%), though small, are significantly higher than the previous independent estimate. Based on the hypothesis that some application stores have a greater density of malicious applications and that advertising within applications and cross-promotional deals may act as infection vectors, we investigate whether the set of applications used on a device can serve as an indicator for infection of that device. Our analysis indicates that, while not an accurate indicator of infection by itself, the application set does serve as an inexpensive method for identifying the pool of devices on which more expensive monitoring and analysis mechanisms should be deployed. Using our two malware datasets we show that this indicator performs up to about five times better at identifying infected devices than the baseline of random checks. Such indicators can be used, for example, in the search for new or previously undetected malware. It is therefore a technique that can complement standard malware scanning. Our analysis also demonstrates a marginally significant difference in battery use between infected and clean devices.

Nowadays Web search engines are experiencing significant performance challenges caused by a huge amount of Web pages and increasingly larger number of Web users. The key issue for addressing these challenges is to design a compact structure which can index Web documents with low space and meanwhile process keyword search very fast. Unfortunately, the current solutions typically separate the space optimization from the search improvement. As a result, such solutions either save space yet with search inefficiency, or allow fast keyword search but with huge space requirement. In this paper, to address the challenges, we propose a novel structure bitlist with both low space requirement and supporting fast keyword search. Specifically, based on a simple and yet very efficient encoding scheme, bitlist uses a single number to encode a set of integer document IDs for low space, and adopts fast bitwise operations for very efficient boolean-based keyword search. Our extensive experimental results on real and synthetic data sets verify that bitlist outperforms the recent proposed solution, inverted list compression [23, 22] by spending 36.71% less space and 61.91% faster processing time, and achieves comparable running time as [8] but with significantly lower space.

In recent years, the content-based publish/subscribe [12], [22] has become a popular paradigm to decouple information producers and consumers with the help of brokers. Unfortunately, when users register their personal interests to the brokers, the privacy pertaining to filters defined by honest subscribers could be easily exposed by untrusted brokers, and this situation is further aggravated by the collusion attack between untrusted brokers and compromised subscribers. To protect the filter privacy, we introduce an anonymizer engine to separate the roles of brokers into two parts, and adapt the k-anonymity and `-diversity models to the contentbased pub/sub. When the anonymization model is applied to protect the filter privacy, there is an inherent tradeoff between the anonymization level and the publication redundancy. By leveraging partial-order-based generalization of filters to track filters satisfying k-anonymity and ℓ-diversity, we design algorithms to minimize the publication redundancy. Our experiments show the proposed scheme, when compared with studied counterparts, has smaller forwarding cost while achieving comparable attack resilience.

Current smartphones have a storage capacity of several gigabytes. More and more information is stored on mobile devices. To meet the challenge of information organization, we turn to desktop search. Users often possess multiple devices, and synchronize (subsets of) information between them. This makes file synchronization more important. This paper presents Dessy, a desktop search and synchronization framework for mobile devices. Dessy supports synchronization of search results, individual files, and directory trees. It allows finding and synchronizing files that reside on remote computers, or the Internet. The contributions of this paper include an energy usage evaluation of the system. Dessy is closely integrated with the Syxaw file synchronizer, which provides efficient file and metadata synchronization, optimizing network usage.

The storage capacity of smartphones has reached tens of gigabytes, while the search functionality remains simple. We have designed a search and synchronization framework for mobile devices, called Dessy. Dessy has been designed with mobility and device constraints in mind. It requires only MIDP 2.0 Mobile Java with File Connection support, and Java 1.5 on desktop machines. This paper demonstrates the application in practice, using multiple devices and synchronizing files between a desktop computer, a laptop, smartphones, and the Internet. Smartphones and laptops are able to search for files hosted on other devices as well as on the Internet. Unnecessary search operations are avoided using Bloom filters.

With the advent of XML we have seen a renewed interest in methods for computing the difference between trees. Methods that include heuristic elements play an important role in practical applications due to the inherent complexity of the problem. We present a method for differencing XML as ordered trees based on mapping the problem to the domain of sequence alignment, applying simple and efficient heuristics in this domain, and transforming back to the tree domain. Our approach provides a method to quickly compute changes that are meaningful transformations on the XML tree level, and includes subtree move as a primitive operation. We evaluate the feasibility of our approach and benchmark it against a selection of existing differencing tools. The results show our approach to be feasible and to have the potential to perform on par with tools of a more complex design in terms of both output size and execution time.