Dawei Zhou

A Local Algorithm for Structure-Preserving Graph Cut

Nowadays, large-scale graph data is being generated in a variety of real-world applications, from social networks to co-authorship networks, from protein-protein interaction networks to road traffic networks. Many existing works on graph mining focus on the vertices and edges, with the first-order Markov chain as the underlying model. They fail to explore the high-order network structures, which are of key importance in many high impact domains. For example, in bank customer personally identifiable information (PII) networks, the star structures often correspond to a set of synthetic identities; in financial transaction networks, the loop structures may indicate the existence of money laundering. In this paper, we focus on mining user-specified high-order network structures and aim to find a structure-rich subgraph which does not break many such structures by separating the subgraph from the rest. A key challenge associated with finding a structure-rich subgraph is the prohibitive computational cost. To address this problem, inspired by the family of local graph clustering algorithms for efficiently identifying a low-conductance cut without exploring the entire graph, we propose to generalize the key idea to model high-order network structures. In particular, we start with a generic definition of high-order conductance, and define the high-order diffusion core, which is based on a high-order random walk induced by user-specified high-order network structure. Then we propose a novel High-Order Structure-Preserving LOcal Cut (HOSPLOC) algorithm, which runs in polylogarithmic time with respect to the number of edges in the graph. It starts with a seed vertex and iteratively explores its neighborhood until a subgraph with a small high-order conductance is found. Furthermore, we analyze its performance in terms of both effectiveness and efficiency. The experimental results on both synthetic graphs and real graphs demonstrate the effectiveness and efficiency of our proposed HOSPLOC algorithm.

REACH 2 -Mote: A Range-Extending Passive Wake-Up Wireless Sensor Node

A wireless sensor network that employs passive radio wake-up of the sensor nodes can reduce the energy cost for unnecessary idle listening and communication overhead, extending the network lifetime. A passive wake-up radio is powered by the electromagnetic waves transmitted by a wake-up transmitter rather than a battery on the sensor node. However, this method of powering the wake-up radio results in a short wake-up range, which limits the performance of a passive wake-up radio sensor network. In this article, we describe our design of a passive wake-up radio sensor node—REACH2-Mote—using a high-efficiency, energy-harvesting module and a very low power wake-up circuit to achieve an extended wake-up range. We implemented REACH2-Mote in hardware and performed field tests to characterize its performance. The experimental results show that REACH2-Mote can achieve a wake-up range of 44 feet. We also modeled REACH2-Mote and evaluated its performance through simulations, comparing its performance to that of another passive wake-up radio approach, an active wake-up radio approach, and a conventional duty cycling approach. The simulation results show that REACH2-Mote can significantly extend the network lifetime while achieving high packet delivery rate and low latency.

Bi-Level Rare Temporal Pattern Detection

Nowadays, temporal data is generated at an unprecedentedspeed from a variety of applications, such as wearable devices, sensor networks, wireless networks, etc. In contrast to suchlarge amount of temporal data, it is usually the case that onlya small portion of them contains information of interest. Forexample, for the ECG signals collected by wearable devices, most of them collected from healthy people are normal, andonly a small number of them collected from people with certain heart diseases are abnormal. Furthermore, even forthe abnormal temporal sequences, the abnormal patterns mayonly be present in a few time segments and are similar amongthemselves, forming a rare category of temporal patterns. Forexample, the ECG signal collected from an individual with acertain heart disease may be normal in most time segments, and abnormal in only a few time segments, exhibiting similarpatterns. What is even more challenging is that such raretemporal patterns are often non-separable from the normalones. Existing works on outlier detection for temporal datafocus on detecting either the abnormal sequences as a whole, orthe abnormal time segments directly, ignoring the relationshipbetween abnormal sequences and abnormal time segments.Moreover, the abnormal patterns are typically treated asisolated outliers instead of a rare category with self-similarity. In this paper, for the first time, we propose a bi-level(sequence-level/ segment-level) model for rare temporal patterndetection. It is based on an optimization frameworkthat fully exploits the bi-level structure in the data, i.e., therelationship between abnormal sequences and abnormal timesegments. Furthermore, it uses sequence-specific simple hiddenMarkov models to obtain segment-level labels, and leverages the similarity among abnormal time segments to estimate the model parameters. To solve the optimization framework, we propose the unsupervised algorithm BIRAD, and also thesemi-supervised version BIRAD-K which learns from a single labeled example. Experimental results on both synthetic andreal data sets demonstrate the performance of the proposedalgorithms from multiple aspects, outperforming state-of-the-arttechniques on both temporal outlier detection and rarecategory analysis.

Jointly Modeling Label and Feature Heterogeneity in Medical Informatics

Multiple types of heterogeneity including label heterogeneity and feature heterogeneity often co-exist in many real-world data mining applications, such as diabetes treatment classification, gene functionality prediction, and brain image analysis. To effectively leverage such heterogeneity, in this article, we propose a novel graph-based model for Learning with both Label and Feature heterogeneity, namely L2F. It models the label correlation by requiring that any two label-specific classifiers behave similarly on the same views if the associated labels are similar, and imposes the view consistency by requiring that view-based classifiers generate similar predictions on the same examples. The objective function for L2F is jointly convex. To solve the optimization problem, we propose an iterative algorithm, which is guaranteed to converge to the global optimum. One appealing feature of L2F is that it is capable of handling data with missing views and labels. Furthermore, we analyze its generalization performance based on Rademacher complexity, which sheds light on the benefits of jointly modeling the label and feature heterogeneity. Experimental results on various biomedical datasets show the effectiveness of the proposed approach.

Discovering rare categories from graph streams

Nowadays, massive graph streams are produced from various real-world applications, such as financial fraud detection, sensor networks, wireless networks. In contrast to the high volume of data, it is usually the case that only a small percentage of nodes within the time-evolving graphs might be of interest to people. Rare category detection (RCD) is an important topic in data mining, focusing on identifying the initial examples from the rare classes in imbalanced data sets. However, most existing techniques for RCD are designed for static data sets, thus not suitable for time-evolving data. In this paper, we introduce a novel setting of RCD on time-evolving graphs. To address this problem, we propose two incremental algorithms, SIRD and BIRD, which are constructed upon existing density-based techniques for RCD. These algorithms exploit the time-evolving nature of the data by dynamically updating the detection models enabling a “time-flexible” RCD. Moreover, to deal with the cases where the exact priors of the minority classes are not available, we further propose a modified version named BIRD-LI based on BIRD. Besides, we also identify a critical task in RCD named query distribution, which targets to allocate the limited budget among multiple time steps, such that the initial examples from the rare classes are detected as early as possible with the minimum labeling cost. The proposed incremental RCD algorithms and various query distribution strategies are evaluated empirically on both synthetic and real data sets.