Check-in Frequency Research Articles

Modeling Long and Short Term User Preferences by Leveraging Multi-Dimensional Auxiliary Information for Next POI Recommendation

Next Point-of-Interest (POI) recommendation has shown great value for both users and providers in location-based services. Existing methods mainly rely on partial information in users’ check-in sequences, and are brittle to users with few interactions. Moreover, they ignore the impact of multi-dimensional auxiliary information such as user check-in frequency, POI category on user preferences modeling and the impact of dynamic changes in user preferences over different time periods on recommendation performance. To address the above limitations, we propose a novel method for next POI recommendation by modeling long and short term user preferences with multi-dimensional auxiliary information. In particular, the proposed model includes a static LSTM module to capture users’ multi-dimensional long term static preferences and a dynamic meta-learning module to capture users’ multi-dimensional dynamic preferences. Furthermore, we incorporate a POI category filter into our model to comprehensively simulate users’ preferences. Experimental results on two real-world datasets demonstrate that our model outperforms the state-of-the-art baseline methods in two commonly used evaluation metrics.

ITourSPOT: a context-aware framework for next POI recommendation in location-based social networks

ABSTRACT The rising prosperity of Location-based Social Networks (LBSNs) witnessed an explosion in the availability of geo-tagged social media data, which enables tremendous location-aware online services, especially next point of interest (POI) recommendation. However, previous next POI recommendation studies usually adopt fix-length time windows for user check-in sequence modeling, leading to a limited capacity in capturing fine-grained user temporal preferences that easily change over time. Besides, existing methods often directly leverage multi-modal contexts as auxiliary to alleviate the data sparsity issue, which fails to fully exploit the sequential patterns of contextual information for inferring user interest drift. To address the above challenges, we propose a novel framework named iTourSPOT which extends traditional collaborative filtering methods with a context-aware POI embedding architecture. For enhancing temporal interests modeling capacity, we associate the context feature extraction with varying-length sessions and incorporate check-in frequencies of POIs as prior knowledge to instruct the session representation learning of our model. Moreover, a collaborative sequence transduction model is designed for joint context sequence modeling and session-based POI recommendation. Experimental results on a real-world geo-tagged photo dataset clearly demonstrate the effectiveness of the proposed framework when compared with state-of-the-art baseline methods, especially in both sparse and cold-start scenarios.

Successive Trajectory Privacy Protection with Semantics Prediction Differential Privacy.

The publication of trajectory data provides critical information for various location-based services, and it is critical to publish trajectory data safely while ensuring its availability. Differential privacy is a promising privacy protection technology for publishing trajectory data securely. Most of the existing trajectory privacy protection schemes do not take into account the user’s preference for location and the influence of semantic location. Besides, differential privacy for trajectory protection still has the problem of balance between the privacy budget and service quality. In this paper, a semantics- and prediction-based differential privacy protection scheme for trajectory data is proposed. Firstly, trajectory data are transformed into a prefix tree structure to ensure that they satisfy differential privacy. Secondly, considering the influence of semantic location on trajectory, semantic sensitivity combined with location check-in frequency is used to calculate the sensitivity of each position in the trajectory. The privacy level of the position is classified by setting thresholds. Moreover, the corresponding privacy budget is allocated according to the location privacy level. Finally, a Markov chain is used to predict the attack probability of each position in the trajectory. On this basis, the allocation of the privacy budget is further adjusted and its utilization rate is improved. Thus, the problem of the balance between the privacy budget and service quality is solved. Experimental results show that the proposed scheme is able to ensure data availability while protecting data privacy.

On Inference of Sense of Place from Geo-Social Networks

The aim of the study is to investigate whether individuals report the places they are attached to in location-based services, and whether there is a relationship between the attachment scores of these places and their corresponding check-in frequency information. A survey is conducted to measure the degree of place attachment of individuals based on self-reported locations. Then their Foursquare log data is collected which includes their check-in and venue information. Our results show that the majority of the participants check in to locations that they are attached to. Attachment score is shown to be related to the check-in frequency. The tips left for venues include terms and phrases, suggesting place attachment.

Locus recommendation using probabilistic matrix factorization techniques

Introduction: The present paper is the outcome of the research “Locus Recommendation using Probabilistic
 Matrix Factorization Techniques” carried out in Manav Rachna International Institute of Research and Studies, India in the year 2019-20.
 
 Methodology: Matrix factorization is a model-based collaborative technique for recommending new items to
 the users. 
 
 Results: Experimental results on two real-world LBSNs showed that PFM consistently outperforms PMF.
 This is because the technique is based on gamma distribution to the model user and item matrix. Using gamma distribution is reasonable for check-in frequencies which are all positive in real datasets. However, PMF is based on Gaussian distribution that can allow negative frequency values as well.
 
 Conclusion: The motive of the work is to identify the best technique for recommending locations with the
 highest accuracy and allow users to choose from a plethora of available locations; the best and interesting
 location based on the individual’s profile.
 
 Originality: A rigorous analysis of Probabilistic Matrix Factorization techniques has been performed on popular LBSNs and the best technique for location recommendation has been identified by comparing the accuracy viz RMSE, Precision@N, Recall@N, F1@N of different models.
 
 Limitations: User’s contextual information like demographics, social and geographical preferences have
 not been considered while evaluating the efficiency of probabilistic matrix factorization techniques for POI Recommendations.
 

Exploiting Category-Level Multiple Characteristics for POI Recommendation

Point of interest (POI) recommendation has received significant attention in recent years, most existing studies exploit multiple auxiliary information to alleviate the problem of data sparsity, and consider the sequential features of user mobility. However, few studies consider the category-level characteristics derived from each user's historical check-in frequencies, and the characteristics between different latent factors. In this paper, we exploit category-level multiple characteristics to generate recommendation. First, we obtain the frequency characteristics by in-depth analysis of the historical check-in frequencies for different categories by each user, and propose a scheme including KL-divergence and text analysis algorithm. Then we propose a category-level sequential- and non-sequential influence-aware probabilistic generative model (CSNS), which models the characteristics (correlation and indeterminate decisiveness) between user latent behavior topics and latent sequence patterns. We design two stages to generate recommendations. In the first stage, CSNS and frequencies characteristics are exploited jointly to recommend the POI categories that users may visit. In the second stage, we depend on user profiles and poi features, and sort the candidate POI sets by combining the POI categories provided in the first stage. Comprehensive experiments on two real-world datasets demonstrate that our method outperforms the existing state-of-the-art POI recommendation models.

Role of Big Data in the Development of Smart City by Analyzing the Density of Residents in Shanghai

In recent decades, a large amount of research has been carried out to analyze location-based social network data to highlight their application. These location-based social network datasets can be used to propose models and techniques that can analyze and reproduce the spatiotemporal structures and symmetries in user activities as well as density estimations. In the current study, different density estimation techniques are utilized to analyze the check-in frequency of users in more detail from location-based social network dataset acquired from Sina-Weibo, also referred as Weibo, over a specific period in 10 different districts of Shanghai, China. The aim of this study is to analyze the density of users in Shanghai city from geolocation data of Weibo as well as to compare their density through univariate and bivariate density estimation techniques; i.e., point density and kernel density estimation (KDE) respectively. The main findings of the study include the following: (i) characteristics of users’ spatial behavior, the center of activity based on their check-ins, (ii) the feasibility of check-in data to explain the relationship between users and social media, and (iii) the presentation of evident results for regulatory or managing authorities for urban planning. The current study shows that the point density and kernel density estimation. KDE methods provide useful insights for modeling spatial patterns using geo-spatial dataset. Finally, we can conclude that, by utilizing the KDE technique, we can examine the check-in behavior in more detail for an individual as well as broader patterns in the population as a whole for the development of smart city. The purpose of this article is to figure out the denser places so that the authorities can divide the mobility of people from the same routes or at least they can control the situation from any further inconvenience.

Location recommendation privacy protection method based on location sensitivity division

Location-based recommendation services can provide users with convenient services, but this requires monitoring and collecting a large amount of location information. In order to prevent location information from being leaked after monitoring and collection, location privacy must be effectively protected. Therefore, this paper proposes a privacy protection method based on location sensitivity for location recommendation. This method uses location trajectories and check-in frequencies to set a threshold so as to classify location sensitivity levels. The corresponding privacy budget is then assigned based on the sensitivity to add Laplace noise that satisfies the differential privacy. Experimental results show that this method can effectively protect the user’s location privacy and reduce the impact of differential privacy noise on service quality.

A Comparative Study of Spatial and Temporal Preferences for Waterfronts in Wuhan based on Gender Differences in Check-In Behavior

The geographical location and check-in frequency of social platform users indicate their personal preferences and intentions for space. On the basis of social media data and gender differences, this study analyzes Weibo users’ preferences and the reasons behind these preferences for the waterfronts of the 21 major lakes within Wuhan’s Third Ring Road, in accordance with users’ check-in behaviors. According to the distribution characteristics of the waterfronts’ points of interest, this study explores the preferences of male and female users for waterfronts and reveals, through the check-in behaviors of Weibo users, the gender differences in the preference and willingness of these users to choose urban waterfronts. Results show that men and women check in significantly more frequently on weekends than on weekdays. Women are more likely than men to check in at waterfronts. Significant differences in time and space exist between male and female users’ preferences for different lakes.

A New Point-of-Interest Classification Model with an Extreme Learning Machine

With the increasing popularity of location-based social networks (LBSNs), an increasing number of people are sharing their locations with friends through check-in activities. Point-of-interest (POI) recommendation, in which new places are suggested to users, is one of the most important tasks in LBSNs. However, after recommendation, it is also of interest to consider whether a user will frequently visit a recommended POI, which may have significant implications regarding the user’s daily mobility behavior or personal preferences. Therefore, in this paper, we propose a new POI classification problem in which the POIs recommended to a user are divided into four classes according to the user’s predicted future check-in frequency: daily check-in POIs, weekly check-in POIs, monthly check-in POIs, and yearly check-in POIs. To solve this POI classification problem, we also propose a new POI classification model called POIC-ELM. In the POIC-ELM model, we first extract nine features related to three factors: each POI itself, the user’s personality, and the user’s social relationships. Then, we use these features to train a POI classifier based on an extreme learning machine (ELM), which is one of the most popular types of classifiers among state-of-the-art classification techniques. A series of experiments show that the effectiveness and efficiency of POIC-ELM are superior to those of other methods. The POIC-ELM model is a valid method for solving the POI classification problem.

Exploiting the roles of aspects in personalized POI recommender systems

The evolution of World Wide Web (WWW) and the smart-phone technologies have revolutionized our daily life. This has facilitated the emergence of many useful systems, such as Location-based Social Networks (LBSN) which have provisioned many factors that are crucial for selection of Point-of-Interests (POI). Some of the major factors are: (i) the location attributes, such as geo-coordinates, category, and check-in time, (ii) the user attributes, such as, comments, tips, reviews, and ratings made to the locations, and (iii) other information, such as the distance of the POI from user’s house/office, social tie between users, and so forth. Careful selection of such factors can have significant impact on the efficiency of POI recommendation. In this paper, we define and analyze the fusion of different major aspects in POI recommendation. Such a fusion and analysis is barely explored by other researchers. The major contributions of this paper are: (i) it analyzes the role of different aspects (e.g., check-in frequency, social, temporal, spatial, and categorical) in the location recommendation, (ii) it proposes two fused models—a ranking-based, and a matrix factorization-based, that incorporate all the major aspects into a single recommendation model, and (iii) it evaluates the proposed models against two real-world datasets.

Assessing spatiotemporal predictability of LBSN: a case study of three Foursquare datasets

Location-based social networks (LBSN) have provided new possibilities for researchers to gain knowledge about human spatiotemporal behavior, and to make predictions about how people might behave through space and time in the future. An important requirement of successfully utilizing LBSN in these regards is a thorough understanding of the respective datasets, including their inherent potential as well as their limitations. Specifically, when it comes to predictions, we must know what we can actually expect from the data, and how we could maximize their usefulness. Yet, this knowledge is still largely lacking from the literature. Hence, this work explores one particular aspect which is the theoretical predictability of LBSN datasets. The uncovered predictability is represented with an interval. The lower bound of the interval corresponds to the amount of regular behaviors that can easily be anticipated, and represents the correct predication rate that any algorithm should be able to achieve. The upper bound corresponds to the amount of information that is contained in the dataset, and represents the maximum correct prediction rate that cannot be exceeded by any algorithms. Three Foursquare datasets from three American cities are studied as an example. It is found that, within our investigated datasets, the lower bound of predictability of the human spatiotemporal behavior is 27%, and the upper bound is 92%. Hence, the inherent potentials of the dataset for predicting human spatiotemporal behavior are clarified, and the revealed interval allows a realistic assessment of the quality of predictions and thus of associated algorithms. Additionally, in order to provide further insight into the practical use of the dataset, the relationship between the predictability and the check-in frequencies are investigated from three different perspectives. It was found that the individual perspective provides no significant correlations between the predictability and the check-in frequency. In contrast, the same two quantities are found to be negatively correlated from temporal and spatial perspectives. Our study further indicates that the heavily frequented contexts and some extraordinary geographic features such as airports could be good starting points for effective improvements of prediction algorithms. In general, this research provides novel knowledge regarding the nature of the LBSN dataset and practical insights for a more reasonable utilization of the dataset.

Spatiotemporal Sequential Influence Modeling for Location Recommendations

Recommending to users personalized locations is an important feature of Location-Based Social Networks (LBSNs), which benefits users who wish to explore new places and businesses to discover potential customers. In LBSNs, social and geographical influences have been intensively used in location recommendations. However, human movement also exhibits spatiotemporal sequential patterns, but only a few current studies consider the spatiotemporal sequential influence of locations on users’ check-in behaviors. In this article, we propose a new gravity model for location recommendations, called LORE, to exploit the spatiotemporal sequential influence on location recommendations. First, LORE extracts sequential patterns from historical check-in location sequences of all users as a Location-Location Transition Graph (L 2 TG), and utilizes the L 2 TG to predict the probability of a user visiting a new location through the developed additive Markov chain that considers the effect of all visited locations in the check-in history of the user on the new location. Furthermore, LORE applies our contrived gravity model to weigh the effect of each visited location on the new location derived from the personalized attractive force (i.e., the weight) between the visited location and the new location. The gravity model effectively integrates the spatiotemporal, social, and popularity influences by estimating a power-law distribution based on (i) the spatial distance and temporal difference between two consecutive check-in locations of the same user, (ii) the check-in frequency of social friends, and (iii) the popularity of locations from all users. Finally, we conduct a comprehensive performance evaluation for LORE using three large-scale real-world datasets collected from Foursquare, Gowalla, and Brightkite. Experimental results show that LORE achieves significantly superior location recommendations compared to other state-of-the-art location recommendation techniques.