A Clustering Refinement Approach for Revealing Urban Spatial Structure from Smart Card Data

Facilitated by rapid development of the data-intensive techniques together with communication and sensing technology, we can take advantage of smart card data collected through Automatic Fare Collection (AFC) systems to establish connections between public transit and urban spatial structure. In this paper, with a case study on Shenzhen metro system in China, we investigate the agglomeration pattern of passenger flow among subway stations. Specifically, leveraging inbound and outbound passenger flows at subway stations, we propose a clustering refinement approach based on cluster member stability among multiple clusterings produced by isomorphic or heterogeneous clusterers. Furthermore, we validate and elaborate five clusters of subway stations in terms of regional functionality and urban planning by comparing station clusters with reference to government planning policies and regulations of Shenzhen city. Additionally, outlier stations with ambiguous functionalities are detected using proposed clustering refinement framework.

Multi-View Multiple Clusterings Using Deep Matrix Factorization

Multi-view clustering aims at integrating complementary information from multiple heterogeneous views to improve clustering results. Existing multi-view clustering solutions can only output a single clustering of the data. Due to their multiplicity, multi-view data, can have different groupings that are reasonable and interesting from different perspectives. However, how to find multiple, meaningful, and diverse clustering results from multi-view data is still a rarely studied and challenging topic in multi-view clustering and multiple clusterings. In this paper, we introduce a deep matrix factorization based solution (DMClusts) to discover multiple clusterings. DMClusts gradually factorizes multi-view data matrices into representational subspaces layer-by-layer and generates one clustering in each layer. To enforce the diversity between generated clusterings, it minimizes a new redundancy quantification term derived from the proximity between samples in these subspaces. We further introduce an iterative optimization procedure to simultaneously seek multiple clusterings with quality and diversity. Experimental results on benchmark datasets confirm that DMClusts outperforms state-of-the-art multiple clustering solutions.

Pruning High-Similarity Clusters to Optimize Data Diversity when Building Ensemble Classifiers

Diversity is a key component for building a successful ensemble classifier. One approach to diversifying the base classifiers in an ensemble classifier is to diversify the data they are trained on. While sampling approaches such as bagging have been used for this task in the past, we argue that since they maintain the global distribution, they do not create diversity. Instead, we make a principled argument for the use of [Formula: see text]-means clustering to create diversity. Expanding on previous work, we observe that when creating multiple clusterings with multiple [Formula: see text] values, there is a risk of different clusterings discovering the same clusters, which would in turn train the same base classifiers. This would bias the ensemble voting process. We propose a new approach that uses the Jaccard Index to detect and remove similar clusters before training the base classifiers, not only saving computation time, but also reducing classification error by removing repeated votes. We empirically demonstrate the effectiveness of the proposed approach compared to the state of the art on 19 UCI benchmark datasets.