Many real applications can be modeled using bipartite graphs, such as users vs. files in a P2P system, traders vs. stocks in a financial trading system, conferences vs. authors in a scientific publication network, and so on. We introduce two operations on bipartite graphs: 1) identifying similar nodes (Neighborhood formation), and 2) finding abnormal nodes (Anomaly detection). And we propose algorithms to compute the neighborhood for each node using random walk with restarts and graph partitioning; we also propose algorithms to identify abnormal nodes, using neighborhood information. We evaluate the quality of neighborhoods based on semantics of the datasets, and we also measure the performance of the anomaly detection algorithm with manually injected anomalies. Both effectiveness and efficiency of the methods are confirmed by experiments on several real datasets. 1

Many real applications can be modeled using bipartite graphs, such as users vs. files in a P2P system, traders vs. stocks in a financial trading system, conferences vs. authors in a scientific publication network, and so on. We introduce two operations on bipartite graphs: 1) identifying similar nodes (relevance search), and 2) finding nodes connecting irrelevant nodes (anomaly detection). And we propose algorithms to compute the relevance score for each node using random walk with restarts and graph partitioning; we also propose algorithms to identify anomalies, using relevance scores. We evaluate the quality of relevance search based on semantics of the datasets, and we also measure the performance of the anomaly detection algorithm with manually injected anomalies. Both effectiveness and efficiency of the methods are confirmed by experiments on several real datasets. 1.

Abstract. The development of technology generates huge amounts of non-textual information, such as images. An efficient image annotation and retrieval system is highly desired. Clustering algorithms make it possible to represent visual features of images with finite symbols. Based on this, many statistical models, which analyze correspondence between visual features and words and discover hidden semantics, have been published. These models improve the annotation and retrieval of large image databases. However, image data usually have a large number of dimensions. Traditional clustering algorithms assign equal weights to these dimensions, and become confounded in the process of dealing with these dimensions. In this paper, we propose weighted feature selection algorithm as a solution to this problem. For a given cluster, we determine relevant features based on histogram analysis and assign greater weight to relevant features as compared to less relevant features. We have implemented various different models to link visual tokens with keywords based on the clustering results of K-means algorithm with weighted feature selection and without feature selection, and evaluated performance using precision, recall and correspondence accuracy using benchmark dataset. The results show that weighted feature selection is better than traditional ones for automatic image annotation and retrieval.

the input data is a collection of images that are annotated with a given keyword, such as “car”. The problem is to attribute the annotation to specific parts of the images. There exists plenty of suitable input data readily

Content-based image retrieval (CBIR) systems can be used also for other purposes than online access to unannotated image databases. In particular, when a CBIR system is equipped with an automatic image segmentation subsystem, keyword annotations given on image level can be focused on specific image segments. In this paper, we show that our PicSOM CBIR system is able to reveal semantic knowledge not only from keyword annotations but also from recorded online use of the system. This automatically extracted high abstraction level visual information can then be used to further improve the accuracy of the system and to categorize the objects of the database with semantic concepts. This process, we claim, then helps to bridge the semantic gap between low-level visual features available for computers and the high-level semantic terms used by the humans. The results of the experiments described in this paper support that view. 1.