Point-of-interest Recommendation Research Articles

MvStHgL: Multi-view Hypergraph Learning with Spatial-temporal Periodic Interests for Next POI Recommendation

Providing potential next point-of-interest (POI) suggestions for users has become a prominent task in location-based social networks, which receives more and more attention from the industry and academia, and it remains challenging due to highly dynamic and personalized interactions in user movements. Currently, state-of-the-art works develop various graph- and sequential-based learning methods to model user-POI interactions and transition regularities. However, there are still two significant shortcomings in these works: (1) Ignoring personalized spatial- and temporal-aspect interactive characteristics capable of exhibiting periodic interests of users; (2) Insufficiently leveraging the sequential patterns of interactions for beyond-pairwise high-order collaborative signals among users’ sequences. To jointly address these challenges, we propose a novel multi-view hypergraph learning with spatial-temporal periodic interests for next POI recommendation (MvStHgL). In the local view, we attempt to learn the POI representation of each interaction via jointing periodic characteristics of spatial and temporal aspects. In the global view, we design a hypergraph by regarding interactive sequences as hyperedges to capture high-order collaborative signals across users, for further POI representations. More specifically, the output of POI representations in the local view is used for the initialized embedding, and the aggregation and propagation in the hypergraph are performed by a novel Node-to-Hypergraph-to-Node scheme. Furthermore, the captured POI embeddings are applied to achieve sequential dependency modeling for next POI prediction. Extensive experiments on three real-world datasets demonstrate that our proposed model outperforms the state-of-the-art models.

Higher-order embedded learning for heterogeneous information networks and adaptive POI recommendation

Traditional POI recommendation results tend to be homogeneous due to data sparsity, and cannot adapt to the dynamic switching of user locations. Regarding the issues, we propose an adaptive Point of Interest (POI) recommendation algorithm that integrates the user social attributes and POI natural attributes. On the one hand, to fully utilize the complex node interactions and semantic information of heterogeneous networks, an effective higher-order embedding learning framework based on multi-head attention mechanism and multi-task shared features, called UI-HEHo, is presented. It divides hierarchical embedding learning into inter-order and intra-order aggregation to learn richer and more comprehensive representations. Based on UI-HEHo framework, a POI-user social attribute model is constructed to obtain the relational distance between interest points. On the other hand, a POI-natural attributes model based on user preference activity range prediction is constructed to further draw the influence between interest points. Furthermore, a tensor-based dynamic update scheme is designed to reduce the high cost of model updates. Finally, the Radviz graph is utilized to obtain the final recommendation results. Experimental results demonstrate our proposed approach significantly improves the POI recommendation performance, especially achieves 20.7% and 23.8% improvement on the Foursquare and Yelp datasets in terms of the newly defined diversity and novelty metrics.

Recency-based spatio-temporal similarity exploration for POI recommendation in location-based social networks

Point-of-interest (POI) recommendation is one of the primary tasks of location-based social networks (LBSNs). With user data in bulk, extracting useful information and addressing issues such as data sparsity and cold-start problems looming large in collaborative filtering become difficult. One of the plausible solutions is to incorporate contextual information into the recommendation process. In this article, we propose a Recency-based Spatio-Temporal Similarity Exploration (RSTSE) for POI recommendation that utilizes the recency-based trust estimation among the prospective neighbors of the target user. The trust level is categorized into two heads: direct trust, which can be extracted from the peer group information of the user, and indirect trust, which is measured based on venue popularity, temporal recency, radial proximity, and transitivity. The approach consists of two phases. In the incipient phase, POIs are extracted based on the preferences of potential neighbors, including the users who are recognized peers, the users with similar visiting histories in the spatial and temporal context, and the users with friend-of-friend relations. The telic phase involves Neural Collaborative Filtering (NCF) to capture the linear and non-linear user–POI interactions better. RSTSE has been evaluated on three real-world datasets, namely, Gowalla, Foursquare, and Weeplaces, and the results suggest efficacy over other state-of-the-art approaches.

TAU: Trajectory Data Augmentation with Uncertainty for Next POI Recommendation

Next Point-of-Interest (POI) recommendation has been proven effective at utilizing sparse, intricate spatial-temporal trajectory data to recommend subsequent POIs to users. While existing methods commonly alleviate the problem of data sparsity by integrating spatial-temporal context information, POI category features, and social relationships, they largely overlook the fact that the trajectory sequences collected in the datasets are often incomplete. This oversight limits the model’s potential to fully leverage historical context. In light of this background, we propose Trajectory Data Augmentation with Uncertainty (TAU) for Next POI Recommendation. TAU is a general graph-based trajectory data augmentation method designed to complete user mobility patterns by marrying uncertainty estimation into the next POI recommendation task. More precisely, TAU taps into the global transition pattern graph to identify sets of intermediate nodes located between every pair of locations, effectively leveraging edge weights as transition probabilities. During trajectory sequence construction, TAU selectively prompts intermediate nodes, chosen based on their likelihood of occurrence as pseudo-labels, to establish comprehensive trajectory sequences. Furthermore, to gauge the certainty and impact of pseudo-labels on the target location, we introduce a novel confidence-aware calibration strategy using evidence deep learning (EDL) for improved performance and reliability. The experimental results clearly indicate that our TAU method achieves consistent performance improvements over existing techniques across two real-world datasets, verifying its effectiveness as the state-of-the-art approach to the task.

Learning Time Slot Preferences via Mobility Tree for Next POI Recommendation

Next Point-of-Interests (POIs) recommendation task aims to provide a dynamic ranking of POIs based on users' current check-in trajectories. The recommendation performance of this task is contingent upon a comprehensive understanding of users' personalized behavioral patterns through Location-based Social Networks (LBSNs) data. While prior studies have adeptly captured sequential patterns and transitional relationships within users' check-in trajectories, a noticeable gap persists in devising a mechanism for discerning specialized behavioral patterns during distinct time slots, such as noon, afternoon, or evening. In this paper, we introduce an innovative data structure termed the ``Mobility Tree'', tailored for hierarchically describing users' check-in records. The Mobility Tree encompasses multi-granularity time slot nodes to learn user preferences across varying temporal periods. Meanwhile, we propose the Mobility Tree Network (MTNet), a multitask framework for personalized preference learning based on Mobility Trees. We develop a four-step node interaction operation to propagate feature information from the leaf nodes to the root node. Additionally, we adopt a multitask training strategy to push the model towards learning a robust representation. The comprehensive experimental results demonstrate the superiority of MTNet over eleven state-of-the-art next POI recommendation models across three real-world LBSN datasets, substantiating the efficacy of time slot preference learning facilitated by Mobility Tree.

Where To Go at the Next Timestamp

The next Point of Interest (POI) recommendation is the core technology of smart city. Current state-of-the-art models attempt to improve the accuracy of the next POI recommendation by incorporating temporal and spatial intervals or by partitioning the POI coordinates into grids. However, they all overlook a detail that in real life, people always want to know where to go at an exact time point or after a specific time interval instead of aimlessly asking where to go next. Moreover, due to individual preferences, different users may visit different places at the same timestamp. Therefore, utilizing timestamp queries can enhance the personalized recommendation capability of the model and mitigate overfitting risks. These implies that using timestamp can achieve more precise recommendations. To the best of our knowledge, we are the first to use the next timestamp for next POI recommendation. In particular, we propose a Time-Stamp Cross Attention Network (TSCAN). TSCAN is a two-layer cross-attention network. The first layer, Time Stamp Cross Attention Block (TSCAB), uses cross-attention between the next timestamp and historical timestamps, and multiplies the attention scores on corresponding POI to predict the next POI that is most related to the history. The other layer, Cross Time Interval Aware Block (CTIAB), applies the time intervals between the next timestamp and historical timestamps to the POI obtained by TSCAB and historical POIs, allowing temporally adjacent POIs to have a greater similarity. Our model not only has a significant improvement in accuracy but also achieves the goal of personalized recommendation, effectively alleviating overfitting. We evaluate the proposed model with three real-world LBSN datasets, and show that TSCAN outperforms the state-of-the-art next POI recommendation models by 5~9%. TSCAN can not only recommend the next POI, but also recommend the possible POI to visit at any specific timestamp in the future.

Learning user preferences from Multi-Contextual Sequence influences for next POI recommendation

<abstract> <p>The recommendation of the next Point of Interest (POI) has attracted significant attention within the domain of POI recommendations in recent years. Existing methods for next POI recommendation are built on the original check-in sequences of users. Despite effectiveness, the original check-in sequences mix the influences of different contextual factors, which inevitably weakens the model ability of learning user preferences from the complex contextual information. To overcome this issue, we propose a novel Multi-Contextual Sequence-based Attention Network (MCSAN) for next POI recommendations. MCSAN first develops a new con-textual influence-based sampling strategy, which can transform the original check-in sequences into a series of contextual information-aware subsequences. Moreover, the constructed subsequences meticulously capture the impacts of various contextual information from the original check-in sequences. Then, MCSAN leverage the attention-based neural network to learn the representations of POIs from the generated subsequences. Finally, MCSAN develops a new feature fusion method that extracts user preferences from the learned POI presentations adaptively. Extensive experiments conducted on real-world datasets indicate the effectiveness of our proposed MCSAN for the next POI recommendation task, compared to recent representative methods.</p> </abstract>

FAGRec: Alleviating data sparsity in POI recommendations via the feature-aware graph learning

<abstract> <p>Point-of-interest (POI) recommendation has attracted great attention in the field of recommender systems over the past decade. Various techniques, such as those based on matrix factorization and deep neural networks, have demonstrated outstanding performance. However, these methods are susceptible to the impact of data sparsity. Data sparsity is a significant characteristic of POI recommendation, where some POIs have limited interaction records and, in extreme cases, become cold-start POIs with no interaction history. To alleviate the influence of data sparsity on model performance, this paper introduced FAGRec, a POI-recommendation model based on the feature-aware graph. The key idea was to construct an interaction graph between POIs and their initial features. This allows the transformation of POI features into a weighted aggregation of initial features. Different POIs can share the learned representations of initial features, thereby mitigating the issue of data sparsity. Furthermore, we proposed attention-based graph neural networks and a user preference estimation method based on delayed time factors for learning representations of POIs and users, contributing to the generation of recommendations. Experimental results on two real-world datasets demonstrate the effectiveness of FAGRec in the task of POI recommendation.</p> </abstract>

Exploiting User Mobility for Time-aware POI Recommendation in Social Networks

Location-Based Social Networks (LBSNs) have recently gained increasing popularity. Timeaware Point-of-Interest(POI) recommendation is one of the most important location-aware services, which can recommend locations to a target user at the specific time based on check-in history. However, current techniques ignore user’s mobility within a region, which plays a vital role in the POI chosen decision. Most of existing algorithms fail to capture the latent relations behind the temporal factors and geographical influence. In this paper, we propose a novel hybrid Time-aware POI Recommendation model based on User Mobility: TPR-UM, which improves the POI recommendation accuracy greatly. More specifically, by introducing the implicit region factor, we capture users’ mobility through collective user actions and geographical properties of locations. We generate the check-in patterns based on different time intervals and different regions to exploit the latent relations between temporal factors and geographical influence. Finally, we conduct comprehensive experiments on two real-world datasets, Gowalla and Foursquare. The experimental results demonstrate that our hybrid method is effective and outperforms other state-of-the-art algorithms in terms of precision, recall and Fβ measurement.

Sampling-based epoch differentiation calibrated graph convolution network for point-of-interest recommendation

In location-based social networks, calibrating a point-of-interest (POI) recommendation system is as important as its accuracy for improving user satisfaction. POI recommendation calibration is primarily classified as categorical or geographical calibration. Categorical calibration ensures that the recommended items are distributed proportionally among the past interest categories of the target user. When a target user checks 80 Chinese, 10 Japanese, and 10 French restaurants, a recommendation list with a ratio of 8:1:1 for Chinese, Japanese, and French restaurants can be reasonably expected. In addition to categorical calibration, geographical calibration has been proposed to increase user interest in the recommended results. Users have a high probability of revisiting locations in their subareas. Therefore, the POIs recommended in multiple subareas of interest are more suitable than those from one small and frequently visited subarea. However, improving the calibration and accuracy are conflicting tasks. To achieve high calibration while maintaining accuracy, previous studies proposed reranking-based techniques to rerank the candidate list and return POIs with high calibration. However, optimizing the calibration by reranking is independent of the basic-candidate-item generation model, resulting in a suboptimal system. To tackle the problem, we propose a novel sampling-based differentiation technique to merge the task of improving calibration into the GCN model training process and directly generate the final recommendation list. The model is flexible and can be applied to different domains, where a domain can be a subarea or category. In a three-layer GCN, the layer one represents the historical check-ins of the user, whereas layer three includes the candidate POIs from which the target user aggregates information. We trained the model to make the distribution of the POI domains at layer three approximated the distribution at layer one. Experimental results on Philadelphia and Tucson datasets confirmed that the proposed method outperforms all state-of-the-art GCN+ geo-reranking and GCN+ MCF baselines, improving Recall@ 5 from 0.0394 to 0.0412 (4.57%) and Jensen–Shannon measure (JS)@ 5 from 0.5931 to 0.6734 (13.54%) on the Philadelphia dataset and improving Recall@ 5 from 0.0495 to 0.0517 (4.40%) and JS@ 5 from 0.5869 to 0.6598 (12.42%) on the Tucson dataset for categorical calibration. The model was also tested in the geographical domain and a similar trend was observed.

Siamese learning based on graph differential equation for Next-POI recommendation

Next Point-of-Interest (POI) recommendation is highly challenging in its ill-posedness of data sparsity and elusive motives. Many models, including sequence- and graph-based, have been proposed to alleviate these problems. However, they still contain drawbacks: they either need a more accurate depiction of users’ complex trajectories or novel perspectives on temporal-variant features in the discretized sequences of visits. To better cope with the two challenges, we proposed a novel time-continuous model, namely POIGDE. Our model explicitly exploits continuous variation of users’ interests by solving a graph differential equation (GDE) on users’ interaction behaviors. To maintain an invariant distribution while solving GDEs, we utilize a time-serial graph along with an interval-aware attention mechanism to learn the dynamics of interest transference. This novel update mechanism helps to track the original data distribution by confining the update process in a linear combination manner. Under this restriction, our model can obtain item representations for prediction, which share the same distribution as the actual POIs with which users interact. This practice also applies Siamese Learning in the POI recommendation to directly learn from the representation of ground-truth labels. By comparison with positive samples, the inferred probability of POIs being visited in the future can be obtained based on similarity (i.e., mean square error) between high-dimensional representations of the corresponding POIs. Our model outperforms state-of-the-art models on three real-world datasets by 0.16-23.44%, showing its potential and prospects in the POI recommendation domain.

Learning Hierarchical Spatial Tasks with Visiting Relations for Next POI Recommendation

Sparsity is an established problem for the next Point-of-Interest (POI) recommendation task, where it hinders effective learning of user preferences from the User-POI matrix. However, learning multiple hierarchically related spatial tasks, and visiting relations between users and POIs, can help to alleviate this sparsity problem. In this article, we propose our Hierarchical Multi-Task Graph Recurrent Network (HMT-GRN) approach, which alleviates the sparsity problem by learning different User-Region matrices of lower sparsities in a multi-task setting. We then perform a Hierarchical Beam Search (HBS) on the different region and POI distributions to hierarchically reduce the search space with increasing spatial granularity and predict the next POI. Our HBS provides efficiency gains by reducing the search space, resulting in speedups of 5 to 7 times over an exhaustive approach. In addition, we propose a selectivity layer to predict if the next POI has been visited before by the user to balance between personalization and exploration. Further, we propose a novel Joint Triplet Loss Learning (JTLL) module to learn visited and unvisited relations between users and POIs for the recommendation task. Experimental results on two real-world Location-Based Social Network (LBSN) datasets show that our proposed approach significantly outperforms baseline and the state-of-the-art methods.

A Diffusion Model for POI Recommendation

Next Point-of-Interest (POI) recommendation is a critical task in location-based services that aim to provide personalized suggestions for the user’s next destination. Previous works on POI recommendation have laid focus on modeling the user’s spatial preference. However, existing works that leverage spatial information are only based on the aggregation of users’ previous visited positions, which discourages the model from recommending POIs in novel areas. This trait of position-based methods will harm the model’s performance in many situations. Additionally, incorporating sequential information into the user’s spatial preference remains a challenge. In this article, we propose Diff-POI : a Diffu sion-based model that samples the user’s spatial preference for the next POI recommendation. Inspired by the wide application of diffusion algorithm in sampling from distributions, Diff-POI encodes the user’s visiting sequence and spatial character with two tailor-designed graph encoding modules, followed by a diffusion-based sampling strategy to explore the user’s spatial visiting trends. We leverage the diffusion process and its reverse form to sample from the posterior distribution and optimized the corresponding score function. We design a joint training and inference framework to optimize and evaluate the proposed Diff-POI. Extensive experiments on four real-world POI recommendation datasets demonstrate the superiority of our Diff-POI over state-of-the-art baseline methods. Further ablation and parameter studies on Diff-POI reveal the functionality and effectiveness of the proposed diffusion-based sampling strategy for addressing the limitations of existing methods.

NRDL: Decentralized user preference learning for privacy-preserving next POI recommendation

Predicting where a user goes next in terms of his or her previously visited points of interest (POIs) is significant for facilitating users’ daily lives. Simultaneously, it must be acknowledged that the check-in and trajectory information of the user is absolutely disclosed to others in location-based social networks when recommending the next POIs. Therefore, how to achieve an accurate next POI recommendation on the premise of privacy preservation is a critical challenge. To address this challenge, we propose decentralized user preference learning for privacy-preserving next POI recommendation, called NRDL. First, to capture the user’s next POI preference, we model the user’s real-time demand representation by POI profile, POI category, absolute time, transition time and distance between previously visited POIs, which is input into an attention-based recurrent neural network (RNN) model for embedding. Second, to perform privacy preservation, we develop a decentralized learning framework that can achieve user preference learning by raw data on each user’s side. Learning on each user’s side can make the privacy data of the user not be revealed to platforms or others, and learning from raw data can guarantee the value of check-ins and further accuracy. Finally, we evaluate the proposed model on two widely used Gowalla and Foursquare datasets, and the improvements over the state-of-the-art model are 25.00% and 9.95% at recall@1 and NDCG@1 on Gowalla as well as 22.51% and 10.59% on Foursquare.

Temporal-geographical attention-based transformer for point-of-interest recommendation

Point-of-Interest (POI) recommendation is one of the most important tasks in the field of social network analysis. Many efforts have been proposed to enhance the model performance for the POI recommendation task in recent years. Existing studies have revealed that the temporal factor and geographical factor are two crucial contextual factors which influence user decisions. However, they only learn representations of POIs and users from the single contextual factor and fuse the learned representations in the final stage, which ignores the interactions of different contextual factors, leading to learning suboptimal representations of POIs and users. To overcome this gap, we propose a novel Temporal-Geographical Attention-based Transformer (TGAT) for the POI recommendation task. Specifically, TGAT develops a hybrid sequence sampling strategy that samples the sequence of POIs from the different contextual factor POI graphs generated by the users’ check-in records. In this way, the interactions of different contextual factors can be care-fully pre-served. Then TGAT conducts a Transformer-based neural network backbone to learn representations of POIs from the sampling sequences. In addition, a weighted aggregation strategy is proposed to fuse the representations learned from different context factors. The extensive experimental results on real-world datasets have demonstrated the effectiveness of TGAT.

A BiLSTM-attention-based point-of-interest recommendation algorithm

Abstract Aiming at the problem that users’ check-in interest preferences in social networks have complex time dependences, which leads to inaccurate point-of-interest (POI) recommendations, a location-based POI recommendation model using deep learning for social network big data is proposed. First, the original data are fed into an embedding layer of the model for dense vector representation and to obtain the user’s check-in sequence (UCS) and space-time interval information. Then, the UCS and spatiotemporal interval information are sent into a bidirectional long-term memory model for detailed analysis, where the UCS and location sequence representation are updated using a self-attention mechanism. Finally, candidate POIs are compared with the user’s preferences, and a POI sequence with three consecutive recommended locations is generated. The experimental analysis shows that the model performs best when the Huber loss function is used and the number of training iterations is set to 200. In the Foursquare dataset, Recall@20 and NDCG@20 reach 0.418 and 0.143, and in the Gowalla dataset, the corresponding values are 0.387 and 0.148.

MARAN: Supporting awareness of users’ routines and preferences for next POI recommendation based on spatial aggregation

Next point-of-interest (POI) recommendation has emerged as an essential task in recommender systems with the rapid development of location-based social networks (LBSNs). It has a wide range of applications in smart cities for building personalized scenarios. Current research typically uses sequential relationships to mine user preferences; however, it fails to sufficiently explore the spatial dependence of check-ins and the multi-perspective information they contain. To this end, this study proposes a Multiple Active Region Aware Network (MARAN), a novel routine-aware model for the next POI recommendation that simultaneously captures the user’s routine regularity and short-term preference changes from check-in records. The key to MARAN is its ability to decompose sophisticated user behavior into two parts. One is a stable routine part characterized by central-based graphs built from historical trajectories based on spatial aggregation. The other is an unstable preference part that obtains the user’s recent changes from short-term trajectories. Moreover, a neighborhood-aware negative sampler based on adjacent areas was designed to alleviate spatial sparsity, that is, the imbalance between positive and negative samples during model training. Experiments on two real-world datasets demonstrated that MARAN outperformed state-of-the-art methods.

Pre-Training Across Different Cities for Next POI Recommendation

The Point-of-Interest (POI) transition behaviors could hold absolute sparsity and relative sparsity very differently for different cities. Hence, it is intuitive to transfer knowledge across cities to alleviate those data sparsity and imbalance problems for next POI recommendation. Recently, pre-training over a large-scale dataset has achieved great success in many relevant fields, like computer vision and natural language processing. By devising various self-supervised objectives, pre-training models can produce more robust representations for downstream tasks. However, it is not trivial to directly adopt such existing pre-training techniques for next POI recommendation, due to the lacking of common semantic objects (users or items) across different cities . Thus in this paper, we tackle such a new research problem of pre-training across different cities for next POI recommendation. Specifically, to overcome the key challenge that different cities do not share any common object, we propose a novel pre-training model named CATUS , by transferring the cat egory-level u niversal tran s ition knowledge over different cities. Firstly, we build two self-supervised objectives in CATUS : next category prediction and next POI prediction , to obtain the universal transition-knowledge across different cities and POIs. Then, we design a category-transition oriented sampler on the data level and an implicit and explicit transfer strategy on the encoder level to enhance this transfer process. At the fine-tuning stage, we propose a distance oriented sampler to better align the POI representations into the local context of each city. Extensive experiments on two large datasets consisting of four cities demonstrate the superiority of our proposed CATUS over the state-of-the-art alternatives. The code and datasets are available at https://github.com/NLPWM-WHU/CATUS.

Contrastive graph learning long and short-term interests for POI recommendation

Modeling users’ short-term dynamic and long-term static interests to enhance Point-of-Interests (POI) recommendation performance has shown lots of advantages. Since users’ check-in records can be viewed as a graph network, methods based on Graph Neural Networks (GNNs) have recently shown promising applicability for POI recommendation. However, existing GNN-based works have the following shortcomings: (1) ignoring the impact of complex higher-order relationships between user-POI dynamics over time; and (2) ignoring the difference in POI importance that cannot effectively capture the imbalances of geographical influence among POIs. To address these challenges, we propose a novel Self-supervised Long-and Short-term model (SLS-REC) for POI recommendation. Specifically, we first design a spatio-temporal Hawkes attention hypergraph neural network to capture the spatial dependence and temporal evolution in users’ short-term dynamic interests. Then we introduce a dynamic propagation mechanism of GNNs to learn the geographic influences underlying geographic imbalances among POIs. In addition, the contrastive learning framework over a fine-grained node dropout strategy is applied to maximize the mutual information of long and short-term interest representations. Finally, we adaptively unify the recommendation and self-supervised task with an attention-based mechanism to optimize the proposed SLS-REC model for POI recommendation. Experiments on real-world datasets show that the proposed model significantly outperforms state-of-the-art methods.

Conversation-Based Adaptive Relational Translation Method for Next POI Recommendation With Uncertain Check-Ins.

The uncertain check-ins bring challenges for current static next point-of-interest (POI) recommendation methods. Fortunately, the conversation-based recommendation has been shown the merit of integrating immediate user preference for more accurate recommendations. We, therefore, propose a conversation-based adaptive relational translation (CART) approach for the next POI recommendation over uncertain check-ins. It is equipped with recommender and conversation modules to interactively acquire users' immediate preferences and make dynamic recommendations. Specifically, the recommender built upon the adaptive relational translation method performs location prediction via modeling both users' historical sequential behaviors and the immediate preference received from conversations; the conversation module aims to achieve successful recommendations in fewer conversation turns by learning a conversational strategy, whereby the recommender can be updated via the user response. Extensive experiments on four real-world datasets show the superiority of our proposed CART over the state of the arts.