We propose a method for discovering the dependency relationships between the topics of documents shared in social networks using the latent social interactions, attempting to answer the question: given a seemingly new topic, from where does this topic evolve? In particular, we seek to discover the pair-wise probabilistic dependency in topics of documents which associate social actors from a latent social network, where these documents are being shared. By viewing the evolution of topics as a Markov chain, we estimate a Markov transition matrix of topics by leveraging social interactions and topic semantics. Metastable states in a Markov chain are applied to the clustering of topics. Applied to the CiteSeer dataset, a collection of documents in academia, we show the trends of research topics, how research topics are related and which are stable. We also show how certain social actors, authors, impact these topics and propose new ways for evaluating author impact.

Motif discovery from bio sequences, a challenging task both experimentally and computationally, has been a topic of immense study in recent years. In this paper, we formulate the motif discovery problem as a multiple-instance problem and employ a multiple-instance learning method, the MILES method, to identify motif from biological sequences. Each sequence is mapped into a feature space defined by instances in training sequences with a novel instance-bag similarity measure. We employ I-norm SVM to select important features and construct classifiers simultaneously. These high-ranked features correspond to discovered motifs. We apply this method to discover transcriptional factor binding sites in promoters, a typical motif finding problem in biology, and show that the method is at least comparable to existing methods

We present a novel spectral clustering method that enables users to incorporate prior knowledge of the size of clusters into the clustering process. The cost function, which is named size regularized cut (SRcut), is defined as the sum of the inter-cluster similarity and a regularization term measuring the relative size of two clusters. Finding a partition of the data set to minimize SRcut is proved to be NP-complete. An approximation algorithm is proposed to solve a relaxed version of the optimization problem as an eigenvalue problem. Evaluations over different data sets demonstrate that the method is not sensitive to outliers and performs better than normalized cut.

Wireless ad-hoc sensor networks are being developed to carry out tasks such as target detection and tracking, environment monitoring, and data collection across the area of deployment. We explore the problem of using sensor networks to detect and track continuous objects, such as wild fire and bio-chemical material. The continuous objects are different from traditional one or many individual targets in that they are continuously distributed across a region and usually occupy a large area. These continuous objects tend to diffuse, increase in size, change in shape, or even split into multiple relatively smaller continuous objects. The fusion and dissemination of local boundary information becomes a very challenging problem. In the paper, we propose a dynamic cluster-based structure to track the movement of boundaries and facilitate the fusion and dissemination of boundary information in a sensor network.

Wireless ad-hoc sensor networks are being developed to collect data across the area of deployment. To stamp the collected data and facilitate communication protocols, it is necessary to identify the location of each sensor. Most existing localization algorithms make use of trilateration or multilateration based on range measurements obtained from RSSI, TOA, TDOA and AoA. In the paper, we first study some situations that most existing sensor localization methods fail to perform well. An example of such situations is when the topology of a sensor network is anisotropic. We propose a distributed sensor localization method with estimation-comparison-correction paradigm to address these conditions. In detail, multidimensional scaling (MDS) technique is applied to recover a series of local maps for adjacent sensors in two (or three) dimensional space. The maps along the route from an anchor to another anchor are stitched together to estimate the sensors' physical locations. Then, the estimated anchor locations are compared with the anchors' physical locations for correction, and estimated sensor locations are corrected at the same time to approximate their physical locations. By iterative estimation, comparison, and correction, the method reduces sensor localization errors caused by anisotropic network topology and complex terrain, which were seldom addressed in previous research.

Recently, a family of methods have been proposed for large scale ad hoc sensor networks that estimate the approximate positions of sensor nodes in the network. In the paper, we first investigate some situations that most existing sensor positioning methods fail to perform well. An example of such situations is that the topology of a sensor network is anisotropic. Then, we study the multidimensional scaling (MDS) technique, and a centralized sensor positioning algorithm is proposed based on MDS. We also propose a distributed sensor positioning algorithm, which is in iteration paradigm, to address these situations. The MDS technique is applied to recover a series of local maps for adjacent sensors, and these maps are then stitched together to form a portion of globe map. At last, the estimated anchor positions are compared with the anchors' physical position for accurate positioning.

This paper presents a novel domain-independent text segmentation method, which identifies the boundaries of topic changes in long text documents and/or text streams. The method consists of three components: As a preprocessing step, we eliminate the document-dependent stop words as well as the generic stop words before the sentence similarity is computed. This step assists in the discrimination of the sentence semantic information. Then the cohesion information of sentences in a document or a text stream is captured with a sentence-distance matrix with each entry corresponding to the similarity between a sentence pair. The distance matrix can be represented with a gray-scale image. Thus, a text segmentation problem is converted into an image segmentation problem. We apply the anisotropic diffusion technique to the image representation of the distance matrix to enhance the semantic cohesion of sentence topical groups as well as sharpen topical boundaries. At last, the dynamic programming technique is adapted to find the optimal topical boundaries and provide a zoom-in and zoom-out mechanism for topics access by segmenting text in variable numbers of sentence topical groups. Our approach involves no domain-specific training, and it can be applied to texts in a variety of domains. The experimental results show that our approach is effective in text segmentation and outperforms several state-of-the-art methods.

There is enormous amount of multilingual documents from various sources and possibly from different countries describing a single event or a set of related events. It is desirable to construct text mining methods that can compare and highlight similarities and differences of those multilingual documents. We discuss our ongoing research that seeks to model a pair of multilingual documents as a weighted bipartite graph with the edge weights computed by means of machine translation. We use spectral method to identify dense subgraphs of the weighted bipartite graph which can be considered as corresponding to sentences that correlate well in textual contents. We illustrate our approach using English and German texts.