An official website of the United States government
Context has been recognized as an important factor to consider in personalized recommender systems. Particularly in location-based services (LBSs), a fundamental task is to recommend to a mobile user where he/she could be interested to visit next at the right time. Additionally, location-based social networks (LBSNs) allow users to share location-embedded information with friends who often co-occur in the same or nearby points-of-interest (POIs) or share similar POI visiting histories, due to the social homophily theory and Tobler’s first law of geography. So, both the time information and LBSN friendship relations should be utilized for POI recommendation. Tensor completion has recently gained some attention in time-aware recommender systems. The problem decomposes a user-item-time tensor into low-rank embedding matrices of users, items and times using its observed entries, so that the underlying low-rank subspace structure can be tracked to fill the missing entries for time-aware recommendation. However, these tensor completion methods ignore the social-spatial context information available in LBSNs, which is important for POI recommendation since people tend to share their preferences with their friends, and near things are more related than distant things. In this paper, we utilize the side information of social networks and POI locations to enhance the tensor completion model paradigm for more effective time-aware POI recommendation. Specifically, we propose a regularization loss head based on a novel social Hausdorff distance function to optimize the reconstructed tensor. We also quantify the popularity of different POIs with location entropy to prevent very popular POIs from being over-represented hence suppressing the appearance of other more diverse POIs. To address the sensitivity of negative sampling, we train the model on the whole data by treating all unlabeled entries in the observed tensor as negative, and rewriting the loss function in a smart way to reduce the computational cost. Through extensive experiments on real datasets, we demonstrate the superiority of our model over state-of-the-art tensor completion methods.
more »
« less
Spatio-temporal prediction of social connections
It is long known that a user's mobility pattern can be affected by his social connections. Users tend to visit same locations visited by their friends. In this paper we investigate the inverse problem: How does a set of user trajectories reflect their social connections. To this end, we define the social connection prediction problem. Given two users, predict the probability that they are friends by mining their historical trajectories. A first approach to do so is to exam how often the two users visit the same location at the same time, which suffers from the problem that different locations/times may have different predictive power. We propose a comprehensive prediction model that is able to capture this difference between locations and time slots. To demonstrate its effectiveness, we trained the proposed model using the publicly available Foursquare dataset. The result shows the proposed model is able to predict existence of social connections between randomly selected users significantly more accurate comparing with the naive method.
more »
« less
- Award ID(s):
- 1637541
- PAR ID:
- 10036551
- Date Published:
- Journal Name:
- Fourth International ACM Workshop on Managing and Mining Enriched Geo-Spatial Data
- Page Range / eLocation ID:
- 1 to 6
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Mobile fitness tracking apps allow users to track their workouts and share them with friends through online social networks. Although the sharing of personal data is an inherent risk in all social networks, the dangers presented by sharing personal workouts comprised of geospatial and health data may prove especially grave. While fitness apps offer a variety of privacy features, at present it is unclear if these countermeasures are sufficient to thwart a determined attacker, nor is it clear how many of these services’ users are at risk. In this work, we perform a systematic analysis of privacy behaviors and threats in fitness tracking social networks. Collecting a month-long snapshot of public posts of a popular fitness tracking service (21 million posts, 3 million users), we observe that 16.5% of users make use of Endpoint Privacy Zones (EPZs), which conceal fitness activity near user-designated sensitive locations (e.g., home, office). We go on to develop an attack against EPZs that infers users’ protected locations from the remaining available information in public posts, discovering that 95.1% of moderately active users are at risk of having their protected locations extracted by an attacker. Finally, we consider the efficacy of state-of-the-art privacy mechanisms through adapting geo-indistinguishability techniques as well as developing a novel EPZ fuzzing technique. The affected companies have been notified of the discovered vulnerabilities and at the time of publication have incorporated our proposed countermeasures into their production systems.more » « less
-
User representation learning is vital to capture diverse user preferences, while it is also challenging as user intents are latent and scattered among complex and different modalities of user-generated data, thus, not directly measurable. Inspired by the concept of user schema in social psychology, we take a new perspective to perform user representation learning by constructing a shared latent space to capture the dependency among different modalities of user-generated data. Both users and topics are embedded to the same space to encode users' social connections and text content, to facilitate joint modeling of different modalities, via a probabilistic generative framework. We evaluated the proposed solution on large collections of Yelp reviews and StackOverflow discussion posts, with their associated network structures. The proposed model outperformed several state-of-the-art topic modeling based user models with better predictive power in unseen documents, and state-of-the-art network embedding based user models with improved link prediction quality in unseen nodes. The learnt user representations are also proved to be useful in content recommendation, e.g., expert finding in StackOverflow.more » « less
-
We study the problem of information sampling with a group of mobile robots from an unknown environment. Each robot is given a unique region in the environment for the sampling task. The objective of the robots is to visit a subset of locations in the environment such that the collected information is maximized, and consequently, the underlying information model matches as close to reality as possible. The robots have limited communication ranges, and therefore can only communicate when nearby one another. The robots operate in a stochastic environment and their control uncertainty is handled using factored Decentralized Markov Decision Processes (Dec-MDP). When two or more robots communicate, they share their past noisy observations and use a Gaussian mixture model to update their local information models. This in turn helps them to obtain a better Dec-MDP policy. Simulation results show that our proposed strategy is able to predict the information model closer to the ground truth version than compared to other algorithms. Furthermore, the reduction in the overall uncertainty is more than comparable algorithms.more » « less
-
We study the problem of information sampling with a group of mobile robots from an unknown environment. Each robot is given a unique region in the environment for the sampling task. The objective of the robots is to visit a subset of locations in the environment such that the collected information is maximized, and consequently, the underlying information model matches as close to reality as possible. The robots have limited communication ranges, and therefore can only communicate when nearby one another. The robots operate in a stochastic environment and their control uncertainty is handled using factored Decentralized Markov Decision Processes (Dec-MDP). When two or more robots communicate, they share their past noisy observations and use a Gaussian mixture model to update their local information models. This in turn helps them to obtain a better Dec-MDP policy. Simulation results show that our proposed strategy is able to predict the information model closer to the ground truth version than compared to other algorithms. Furthermore, the reduction in the overall uncertainty is more than comparable algorithms.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1145/3080546.3080551
