Personalized Point-of-interest Recommendation Research Articles

UFC: A Unified POI Recommendation Framework

With the popularity of location-based social networks, personalized points-of-interest (POIs) recommendation has become an essential online service, providing a wide variety of user preferred check-in locations, namely POIs. However, the sparsity of user-POI matrix makes it difficult to recommend unvisited POIs to users. The availability of the user, social and geographical information offers an unprecedented opportunity to address this problem. In this paper, our goal is to accurately recommend POIs to users with the proper order. To handle this, we propose a practical POI recommendation framework UFC by incorporating User preference, Friend importance and Check-in correlation. UFC combines three fundamental factors and derives an overall prediction score of each user for any POIs. User preference is personalized based on collaboration filtering. As for the friend importance, we argue that intimate friends and distant friends share common influence and thus merge these two different factors to model friend importance. We study the geographical correlation of user check-in behavior in depth and employ a probabilistic model to characterize check-in correlation. Extensive experimental results exhibit a significant improvement in our proposed method compared to other state-of-art POI recommendation algorithms.

IMCRec: A multi-criteria framework for personalized point-of-interest recommendations

User preferences on geographical positions and categories of points-of-interest (POIs) have been exploited for POI recommendations. However, in reality, the users’ visiting behaviors are also affected by the attributes of POIs, which reflect the important features of the POIs. Integrating the geographical, category and attribute criteria for POI recommendations suffers challenges of (1) modeling user preferences on multiple attributes with different values, (2) integrating conflicting multiple criteria for POI recommendations, and (3) learning personalized weights from one’s check-in histories with heterogeneous data types. To address these challenges, we propose a new personalized POI recommendation framework, called iMCRec, which recommends POIs by integrating user preferences on geographical, category and attribute criteria with personalized weights. In iMCRec, preference models are first built for individual user’s geographical, category, and attribute preferences. Especially, we propose an attribute preference model by considering both preferences on values of each attribute and importance of different attributes. A sophisticated collaborative filtering method is also developed to fuse the opinion of similar users under the three criteria separately. To learn the personalized weights on the three criteria, a weight learning strategy is proposed. We then develop a fast Multi-Criteria Decision Making (MCDM) algorithm, called FastMCDM, to integrate the three conflicting criteria and efficiently generate top-N POIs as recommendations. Finally, we evaluate the performance of our iMCRec through extensive experiments using two real-world datasets collected from Yelp. Experimental results show that iMCRec not only performs better than the state-of-the-art POI recommendation techniques, but it is also more flexible in dealing with the scale problem and more effective in learning personalized weights than other multi-criteria-based techniques.

Exploring Spatial and Mobility Pattern’s Effects for Collaborative Point-of-Interest Recommendation

In recent years, researches on the mining of user check-in behaviors for point-of-interest(POI) recommendations has attracted a lot of attention. Personalized POI recommendation is a significant task in location-based social networks(LBSNs) because it helps target users explore their surrounding environment and greatly benefits the business in real life. Although a personalized POI recommendation system can significantly facilitate users' outdoor activities, it faces many challenging problems, such as the hardness to model human mobility and the difficulty to address data sparsity. Moreover, geographical influence on users should be personalized, but current studies only model the geographical influence on all users' check-in behaviors in a universal way. In this paper, we design a novel and effective personalized POI recommendation system. First, our system mines the target user's active area based on his or her check-in history, and designs a personalized user spatial similarity calculation method based on the target user's active area. Secondly, our system takes into account three features of the human mobility pattern: spatial, temporal, and sequential properties. Furthermore, our system designs a novel personalized user mobility pattern similarity calculation method based on the features of human mobility pattern. Finally, a recommendation list is generated based on the idea of collaborative filtering. Compared with the state-of-the-art POI recommendation approaches, the experimental results demonstrate that our system achieves much better performance.

Combining Geographical and Social Influences with Deep Learning for Personalized Point-of-Interest Recommendation

Personalized point-of-interest (POI) recommendation is important to location-based social networks (LBSNs) for helping users to explore new places and for helping third-party services to launch targeted advertisements. Discovering effective features or representations from check-in data is the key to POI recommendation. Deep learning is a representation-learning method with multiple levels for discovering intrinsic features to better represent user preferences. We analyzed users’ check-in behavior in detail and developed a deep learning model to integrate geographical and social influences for POI recommendation tasks. We used a semi-restricted Boltzmann machine to model the geographical similarity and a conditional layer to model the social influence. Experiments with real-world LBSNs showed that our method performed better than other state-of-the-art methods. Theoretically, our study contributes to the effective usage of data science and analytics for social recommender system design. In practice, our results can be used to improve the quality of personalized POI recommendation services for websites and applications.

A Common Topic Transfer Learning Model for Crossing City POI Recommendations.

With the popularity of location-aware devices (e.g., smart phones), large amounts of location-based social media data (i.e., user check-in data) are generated, which stimulate plenty of works on personalized point of interest (POI) recommendations using machine learning techniques. However, most of the existing works could not recommend POIs in a new city to a user where the user and his/her friends have never visited before. In this paper, we propose a common topic transfer learning graphical model-the common-topic transfer learning model (CTLM)-for crossing-city POI recommendations. The proposed model separates the city-specific topics (or features) of each city from the common topics (or features) shared by all cities, to enable the users' real interests in the source city to be transferred to the target city. By doing so, the ill-matching problem between users and POIs from different cities can be well addressed by preventing the real interests of users from being influenced by the city-specific features. Furthermore, we incorporate the spatial influence into our proposed model by introducing the regions' accessibility. As a result, the co-occurrence patterns of users and POIs are modeled as the aggregated result from these factors. To evaluate the performance of the CTLM, we conduct extensive experiments on Foursquare and Twitter datasets, and the experimental results show the advantages of CTLM over the state-of-the-art methods for the crossing-city POI recommendations.

Exploiting Spatial and Temporal for Point of Interest Recommendation

An increasing number of users have been attracted by location-based social networks (LBSNs) in recent years. Meanwhile, user-generated content in online LBSNs like spatial, temporal, and social information provides an ever-increasing chance to study the human behavior movement from their spatiotemporal mobility patterns and spawns a large number of location-based applications. For instance, one of such applications is to produce personalized point of interest (POI) recommendations that users are interested in. Different from traditional recommendation methods, the recommendations in LBSNs come with two vital dimensions, namely, geographical and temporal. However, previously proposed methods do not adequately explore geographical influence and temporal influence. Therefore, fusing geographical and temporal influences for better recommendation accuracy in LBSNs remains potential. In this work, our aim is to generate a top recommendation list of POIs for a target user. Specially, we explore how to produce the POI recommendation by leveraging spatiotemporal information. In order to exploit both geographical and temporal influences, we first design a probabilistic method to initially detect users’ spatial orientation by analyzing visibility weights of POIs which are visited by them. Second, we perform collaborative filtering by detecting users’ temporal preferences. At last, for making the POI recommendation, we combine the aforementioned two approaches, that is, integrating the spatial and temporal influences, to construct a unified framework. Our experimental results on two real-world datasets indicate that our proposed method outperforms the current state-of-the-art POI recommendation approaches.

RecNet: a deep neural network for personalized POI recommendation in location-based social networks

ABSTRACTHow to exploit various features of users and points of interest (POIs) for accurate POI recommendation is important in location-based social networks (LBSNs). In this paper, a novel POI recommendation framework, named RecNet, is proposed, which is developed based on a deep neural network (DNN) to incorporate various features in LBSNs and learn their joint influence on user behavior. More specifically, co-visiting, geographical and categorical influences in LBSNs are exploited to alleviate the data sparsity issue in POI recommendation and are converted to feature vector representations of POIs and users via feature embedding. Moreover, the embedded POIs and users are fed into a DNN pairwise to adaptively learn high-order interactions between features. Our method is evaluated on two publicly available LBSNs datasets and experimental results show that RecNet outperforms state-of-the-art algorithms for POI recommendation.

Augmented Collaborative Filtering for Sparseness Reduction in Personalized POI Recommendation

As mobile device penetration increases, it has become pervasive for images to be associated with locations in the form of geotags. Geotags bridge the gap between the physical world and the cyberspace, giving rise to new opportunities to extract further insights into user preferences and behaviors. In this article, we aim to exploit geotagged photos from online photo-sharing sites for the purpose of personalized Point-of-Interest (POI) recommendation. Owing to the fact that most users have only very limited travel experiences, data sparseness poses a formidable challenge to personalized POI recommendation. To alleviate data sparseness, we propose to augment current collaborative filtering algorithms along from multiple perspectives. Specifically, hybrid preference cues comprising user-uploaded and user-favored photos are harvested to study users’ tastes. Moreover, heterogeneous high-order relationship information is jointly captured from user social networks and POI multimodal contents with hypergraph models. We also build upon the matrix factorization algorithm to integrate the disparate sources of preference and relationship information, and apply our approach to directly optimize user preference rankings. Extensive experiments on a large and publicly accessible dataset well verified the potential of our approach for addressing data sparseness and offering quality recommendations to users, especially for those who have only limited travel experiences.

A Time-Aware Personalized Point-of-Interest Recommendation via High-Order Tensor Factorization

Recently, location-based services (LBSs) have been increasingly popular for people to experience new possibilities, for example, personalized point-of-interest (POI) recommendations that leverage on the overlapping of user trajectories to recommend POI collaboratively. POI recommendation is yet challenging as it suffers from the problems known for the conventional recommendation tasks such as data sparsity and cold start, and to a much greater extent. In the literature, most of the related works apply collaborate filtering to POI recommendation while overlooking the personalized time-variant human behavioral tendency. In this article, we put forward a fourth-order tensor factorization-based ranking methodology to recommend users their interested locations by considering their time-varying behavioral trends while capturing their long-term preferences and short-term preferences simultaneously. We also propose to categorize the locations to alleviate data sparsity and cold-start issues, and accordingly new POIs that users have not visited can thus be bubbled up during the category ranking process. The tensor factorization is carefully studied to prune the irrelevant factors to the ranking results to achieve efficient POI recommendations. The experimental results validate the efficacy of our proposed mechanism, which outperforms the state-of-the-art approaches significantly.

A temporal-aware POI recommendation system using context-aware tensor decomposition and weighted HITS

The popularity of location-based social networks (LBSN) provides us with a new perspective for understanding people's travel behaviours and enables a lot of location-based services, such as point of interest (POI) recommendation. However, personalized POI recommendation is very challenging, as the user-location matrix is very sparse for traditional collaborative filtering (CF)-based POI recommendation approaches. The problem becomes even more challenging when people travel to a new city. In addition, temporal influence plays an important role in POI recommendation, for most users tend to visit different kinds of POIs at different time in a day, e.g., visiting a food-related POI at noon and visiting a nightlife spot at night. To the end, we propose a novel POI recommendation system, which consists of two components: context-aware tensor decomposition (CTD) for user preferences modelling and weighted HITS (Hypertext Induced Topic Search)-based POI rating (WHBPR). We model user preferences with a three-dimension tensor (user-category-time). Supplementing the missing entries of the tensor through CTD with the aid of other three matrices, we recover user preferences of different time slots. WHBPR incorporates the impacts of user preferences and social opinions on POI rating. We evaluated our method using the real Foursquare datasets, verifying the advantages of our method beyond other baselines.

SEM-PPA: A semantical pattern and preference-aware service mining method for personalized point of interest recommendation

Point-of-Interest (POI) recommendation has received increasing attention in Location-based Social Networks (LBSNs). It involves user behavior analysis, movement pattern model and trajectory sequence prediction, in order to recommend personalized services to target user. Existing POI recommendation methods are confronted with three problems: (1) they only consider the location information of users' check-ins, which causes data sparsity; (2) they fail to consider the order of users' visited locations, which is valuable to reflect the interest or preference of users; (3) users cannot be recommended the suitable services when they move into the new place. To address the above issues, we propose a semantical pattern and preference-aware service mining method called SEM-PPA to make full use of the semantic information of locations for personalized POI recommendation. In SEM-PPA, we firstly propose a novel algorithm to classify the locations into different types for location identification; then we construct the user model for each user from four aspects, which are location trajectory, semantic trajectory, location popularity and user familiarity; in addition, a potential friends discovery algorithm based on movement pattern is proposed. Finally, we conduct extensive experiments to evaluate the recommendation accuracy and recommendation effectiveness on two real-life datasets from GeoLife and Beijing POI. Experimental results show that SEM-PPA can achieve better recommendation performance in particular for sparse data and recommendation accuracy in comparison with other methods.

Effective successive POI recommendation inferred with individual behavior and group preference

The prevalence of smartphones and mobile social networks allow the users to share their location-based life experience much easier. The large amount of data generated in related location-based social networks provides informative cues on user's behaviors and preferences to support personalized location-based services, like point-of-interest (POI) recommendation. Yet achieving accurate personalized POI recommendation is challenging as the data available for each user is highly sparse. In addition, the computational complexity is high due to the large number of users. In this paper, a novel methodology for personalized successive POI recommendation is proposed. First, the preferred successive category of location is predicted using a third-rank tensor computed based on the partially observed transitions between the categories of user's successive locations where the missing transitions are uncovered by inferring the group preference. The group is achieved according to users' demographics and frequently visited locations. Then, a bipartite graph is constructed based on the recommended categories for each user. To obtain the personalized ranking of locations, a distance weighted HITS algorithm is proposed so that the location authority score is updated iteratively according to the visiting frequency of the group and some distance constraints. The proposed two-step approach with the category prediction incorporated aims to boost the location prediction performance via the smoothing and at the same time reduce the complexity. Experimental results obtained based on the real-world location-based social network data show that the proposed approach outperforms the existing state-of-the-art methods by a large margin.

Author Topic Model based Collaborative Filtering for Personalized POI Recommendation

From social media has emerged continuous needs for automatic travel recommendations. Collaborative filtering (CF) is the most well-known approach. However, existing approaches generally suffer from various weaknesses. For example , sparsity can significantly degrade the performance of traditional CF. If a user only visits very few locations, accurate similar user identification becomes very challenging due to lack of sufficient information for effective inference. Moreover, existing recommendation approaches often ignore rich user information like textual descriptions of photos which can reflect users' travel preferences. The topic model (TM) method is an effective way to solve the “sparsity problem,” but is still far from satisfactory. In this paper, an author topic model-based collaborative filtering (ATCF) method is proposed to facilitate comprehensive points of interest (POIs) recommendations for social users. In our approach, user preference topics, such as cultural, cityscape, or landmark, are extracted from the geo-tag constrained textual description of photos via the author topic model instead of only from the geo-tags (GPS locations). Advantages and superior performance of our approach are demonstrated by extensive experiments on a large collection of data.

A Framework of Personalized Travelling Information Services for Thailand

Planning each comprehensive trip is regarded as a complicated and time-consuming task which includes the process starting from searching specific tourist information to planning an itinerary in unfamiliar areas. In response to the decision making challenges that tourists may encounter in planning a trip or discovering a destination, this paper proposes an integrated framework for a system supporting tourists in Thailand. The aim of the framework is to provide a service that considers data from a variety of sources, including static and social data, and then recommends points of interest (POI) and itineraries for POI, based on the requirements and interests of the user. The framework consists of three components: a personalized POI recommendation engine, an itinerary planner, and a mobile application. Together these form the foundation of a personalized travelling information system for Thailand. This paper outlines the basic framework and provides a discussion on the potential issues encountered.