Tourism destination image perception aims to depict the urban tourism image from the perspective of the perception of tourists, which, therefore, sheds new light on the advancement and innovation of urban tourism. The model proposed in this study can effectively describe the image perception of a tourism destination, with its research conclusions providing a vital referential basis for the sustainable development of urban tourism. Combined with LDA, we construct the research framework of tourism destination image perception and then take the online comments of popular scenic spots in Wuhan on Ctrip Travel as an example. The results show that four aspects are included in tourists’ perception of the city image of Wuhan: experience, history culture, leisure service, and tourist destination. Among them, the social network of the experience dimension is most closely related. In addition, emotion analysis illustrates that tourists’ emotional tendencies tend to be positive under the four perceptual dimensions.

Because the participants are not limited by age-, gender-, race-, or geography-related barriers, recently, massive open online courses (MOOC) have witnessed remarkable growth in number of online self-learners, courses providers and online platforms. MOOC learners usually share some learning experiences and release millions of course-related comments in discussion forum. On the one hand, these comments could reflect the learners’ attitudes toward the online courses. On the other hand, semantic knowledge hidden in these comments would assist other learners to choose the appropriate courses and help instructors to improve their courses’ attraction. Recently, few research works focus on evaluating the courses through reviews mining. Thus, this paper constructs a curriculum evaluation system based on MOOC reviews, which quantifies the curriculum from different topics. Firstly, we employ latent dirichlet allocation (LDA) to mine the reviews generated by students, and obtain a topic-word distribution matrix and a comment-topic distribution matrix which can describe the topics of the course comments. Next, the emotion values of the comments in each topic are calculated by the auto-encoder and Bi-LSTM text classification model. We utilize the emotions and the quantified scores of the courses on different topics to establish a comprehensive curriculum evaluation system. The experimental results show that there are five main indicators abstracted from students’ reviews, which are instructor, course content, course assessment, MOOC platform, and hot courses. Moreover, comment texts of each course under different evaluation indicators are objectively and accurately converted into numerical marks, which can provide the students and educators with reliable references.

The rapid growth of location-based services (LBSs) has greatly enriched people’s urban lives and attracted millions of users in recent years. Location-based social networks (LBSNs) allow users to check-in at a physical location and share daily tips on points of interest (POIs) with their friends anytime and anywhere. Such a check-in behavior can make daily real-life experiences spread quickly through the Internet. Moreover, such check-in data in LBSNs can be fully exploited to understand the basic laws of humans’ daily movement and mobility. This paper focuses on reviewing the taxonomy of user modeling for POI recommendations through the data analysis of LBSNs. First, we briefly introduce the structure and data characteristics of LBSNs, and then we present a formalization of user modeling for POI recommendations in LBSNs. Depending on which type of LBSNs data was fully utilized in user modeling approaches for POI recommendations, we divide user modeling algorithms into four categories: pure check-in data-based user modeling, geographical information-based user modeling, spatiotemporal information-based user modeling, and geosocial information-based user modeling. Finally, summarizing the existing works, we point out the future challenges and new directions in five possible aspects.

Imbalance learning is a hot topic in the data mining and machine learning domains. Data-level, algorithm-level and ensemble solutions are the three main methods proposed thus far to address imbalance learning. To alleviate the issues of data explosion and feature selection in multilayer perceptron based on simultaneous two-sample representation(S2SMLP), in this paper, firstly, spectral clustering is exploited to select majority samples so as to construct a smaller training dataset for the classifier. We divide all majority samples into many clusters through spectral clustering, extract different numbers of representative samples from a cluster according to the size of each cluster, the average distance between the minority class and all samples of the cluster, then construct the training dataset of the classifier by combining these extracted samples from the majority class and all minority samples. Secondly, we propose a novel feature selection method based on the pairwise samples distance constraint, which considers the class labels of paired samples, select the features which push two similar samples closer together and pull two different samples farther apart. Finally, we conduct extensive experiments on 44 two-class imbalanced datasets and four high-dimensional DNA microarray datasets. The experimental results demonstrate that our proposed algorithms outperform some state-of-the-art algorithms in terms of $F\textrm {-measure},G\textrm {-mean}$ and AUC.