Exploiting label correlations is important to multi-label classification. Previous methods capture the high-order label correlations mainly by transforming the label matrix to a latent label space with low-rank matrix factorization. However, the label matrix is generally a full-rank or approximate full-rank matrix, making the low-rank factorization inappropriate. Besides, in the latent space, the label correlations will become implicit. To this end, we propose a simple yet effective method to depict the high-order label correlations explicitly, and at the same time maintain the high-rank of the label matrix. Moreover, we estimate the label correlations and infer model parameters simultaneously via the local geometric structure of the input to achieve mutual enhancement. Comparative studies over twelve benchmark data sets validate the effectiveness of the proposed algorithm in multi-label classification. The exploited high-order label correlations are consistent with common sense empirically.Our code is publicly available athttps://github.com/Chongjie-Si/HOMI.

Multi-label ensemble can achieve superior performance on multi-label learning problems by integrating a number of base classifiers. In existing multi-label ensemble methods, the base classifiers are usually trained with the same original features; it is difficult for each base classifier to capture label-relevant or label subset-relevant information. Meanwhile, the manually designed integrating strategies cannot automatically distinguish the importance of the base classifiers, which also lack flexibility and scalability. In order to resolve these problems, this paper proposes a new multi-label ensemble framework, named Attention-aware Multi-label Ensemble with Label Subset-specIfic FeaTures (AME-LSIFT). It utilizes$c$-means clustering to produce Label Subset-specIfic FeaTures (LSIFT), constructs a neural network based model for each label subset, and integrates the base models with a dynamic and automatic attention-aware mechanism. Moreover, an objective function that considers both the label subset accuracy and ensemble accuracy is developed for training the proposed AME-LSIFT. Experiments conducted on ten benchmark datasets demonstrate the superior performance of the proposed method compared with state-of-the-art approaches.

Utilizing large-scale GPS data to improve taxi services has become a popular research problem in the areas of data mining, intelligent transportation, geographical information systems, and the Internet of Things. In this paper, we utilize a large-scale GPS data set generated by over 7,000 taxis in a period of one month in Nanjing, China, and propose TaxiRec: a framework for evaluating and discovering the passenger-finding potentials of road clusters, which is incorporated into a recommender system for taxi drivers to seek passengers. In TaxiRec, the underlying road network is first segmented into a number of road clusters, a set of features for each road cluster is extracted from real-life data sets, and then a ranking-based extreme learning machine (ELM) model is proposed to evaluate the passenger-finding potential of each road cluster. In addition, TaxiRec can use this model with a training cluster selection algorithm to provide road cluster recommendations when taxi trajectory data is incomplete or unavailable. Experimental results demonstrate the feasibility and effectiveness of TaxiRec.

Utilizing large-scale GPS data to improve taxi services becomes a popular research problem in the areas of data mining, intelligent transportation, and the Internet of Things. In this paper, we utilize a large-scale GPS data set generated by over 7,000 taxis in a period of one month in Nanjing, China, and propose TaxiRec; a framework for discovering the passenger-finding potentials of road clusters, which is incorporated into a recommender system for taxi drivers to hunt passengers. In TaxiRec, we first construct the road network by defining the nodes and road segments. Then, the road network is divided into a number of road clusters through a clustering process on the mid points of the road segments. Afterwards, a set of features for each road cluster is extracted from real-life data sets, and a ranking-based extreme learning machine (ELM) model is proposed to evaluate the passenger-finding potential of each road cluster. Experimental results demonstrate the feasibility and effectiveness of the proposed framework.

Location-based business review (LBBR) sites (e.g., Yelp) provide us a possibility to recommend new points of interest (POIs) for users. The geographical position and category of POIs have been considered as two major factors in modeling users' preferences. However, it is argued that the user's visiting behaviors are also affected by the attributes of POIs, which reflect the basic features of the POIs. Besides, a user may have different preference levels on the same POI with regard to different criteria. To this end, we propose a new personalized POI recommendation framework using Multi-Criteria Decision Making (MCDM). Firstly, preference models are built for the user's geographical, category, and attribute preferences. Then, an MCDM-based recommendation framework is designed to iteratively combine the user's preferences on the three criteria and select the top-N POIs as a recommendation list. Experimental results show that our framework not only outperforms the state-of-the-art POI recommendation techniques, but also provides a better trade-off mechanism for MCDM than the weighted sum approach.

Exploring massive mobile data for location-based services (LBS) becomes one of the key challenges in mobile data mining. In this paper, we propose a framework that uses large-scale cell tower data dumps and extracts points-of-interest (POIs) from a social network web site called Weibo, and provides new LBS based on these two data sets, i.e., predicting the existence of POIs and the number of POIs in a certain area. We use Voronoi diagram to divide a city area into non-overlapping regions, and a k-means clustering algorithm to aggregate neighboring cell towers into region groups. A supervised learning algorithm is adopted to build up a model between the number of connections of cell towers and the POIs in different region groups, where a classification or regression model is used to predict the POI existence or the number of POIs, respectively. We studied 12 state-of-the-art classification and regression algorithms, and the experimental results demonstrate the feasibility and effectiveness of the proposed framework.