Ramamohanarao Kotagiri

A probabilistic approach to detecting network scans

This paper presents a probabilistic approach for detecting network scans in real-time. Unlike previous approaches, our model takes into consideration both the number of destinations or ports accessed by a source, as well as how unusual these accesses are. We demonstrate the effectiveness of our approach in terms of accuracy and throughput, based on an analysis of the unusual sources that were found in real-life packet trace files.

Exploring constraints to efficiently mine emerging patterns from large high-dimensional datasets

Emerging patterns (EPs) were proposed recently to capture changes or di erences betw een datasets: an EP is a multivariate feature whose support increases sharply from a background dataset to a target dataset, and the support ratio is called its gro wth rate. Interesting long EPs often have low support; mining such EPs from high-dimensional datasets is a great challenge due to the combinatorial explosion of the number of candidates. We propose a Constraint-based EP Miner, ConsEPMiner, that utilizes tw o types of constraints for e ectively pruning the search space: External constrain tsare user-giv enminimums on support, growth rate, and growth-rate improvement to con ne the resulting EP set. Inheren t constrain ts | same subset support, top growth rate, and same origin | are deriv ed from the propertiesof EPs and datasets, and are solely for pruning the search space and saving computation. ConsEPMiner can eÆciently mine all EPs at low support on large highdimensional datasets, with low minimums on growth rate and growth-rate improvement. In comparison, the widely known Apriori-like approach is ine ective on high-dimensional data. While ConsEPMiner adopts several ideas from DenseMiner [4], a recent constrain t-based association rule miner, its main new contributions are the introduction of inherent constrain ts and the ways to use them together with externalconstrain ts for eÆcient EP mining from dense datasets. Experiments on dense data show that, at low support, ConsEPMiner outperforms the Apriori-like approach by orders of magnitude and is more than twice as fast as the DenseMiner approach.

Blood Vessel Segmentation from Color Retinal Images using Unsupervised Texture Classification

Automated blood vessel segmentation is an important issue for assessing retinal abnormalities and diagnoses of many diseases. The segmentation of vessels is complicated by huge variations in local contrast, particularly in case of the minor vessels. In this paper, we propose a new method of texture based vessel segmentation to overcome this problem. We use Gaussian and L*a*b* perceptually uniform color spaces with original RGB for texture feature extraction on retinal images. A bank of Gabor energy filters are used to analyze the texture features from which a feature vector is constructed for each pixel. The fuzzy C-means (FCM) clustering algorithm is used to classify the feature vectors into vessel or non-vessel based on the texture properties. From the FCM clustering output we attain the final output segmented image after a post processing step. We compare our method with hand-labeled ground truth segmentation of five images and achieve 84.37% sensitivity and 99.61% specificity.

Data Mining: How Research Meets Practical Development?

Abstract.At the 2001 IEEE International Conference on Data Mining in San Jose, California, on November 29 to December 2, 2001, there was a panel discussion on how data mining research meets practical development. One of the motivations for organizing the panel discussion was to provide useful advice for industrial people to explore their directions in data mining development. Based on the panel discussion, this paper presents the views and arguments from the panel members, the Conference Chair and the Program Committee Co-Chairs. These people as a group have both academic and industrial experiences in different data mining related areas such as databases, machine learning, and neural networks. We will answer questions such as (1) how far data mining is from practical development, (2) how data mining research differs from practical development, and (3) what are the most promising areas in data mining for practical development.

Approximate clustering in very large relational data

Different extensions of fuzzy c‐means (FCM) clustering have been developed to approximate FCM clustering in very large (unloadable) image (eFFCM) and object vector (geFFCM) data. Both extensions share three phases: (1) progressive sampling of the VL data, terminated when a sample passes a statistical goodness of fit test; (2) clustering with (literal or exact) FCM; and (3) noniterative extension of the literal clusters to the remainder of the data set. This article presents a comparable method for the remaining case of interest, namely, clustering in VL relational data. We will propose and discuss each of the four phases of eNERF and our algorithm for this last case: (1) finding distinguished features that monitor progressive sampling, (2) progressively sampling a square N × N relation matrix RN until an n × n sample relation Rn passes a statistical test, (3) clustering Rn with literal non‐Euclidean relational fuzzy c‐means, and (4) extending the clusters in Rn to the remainder of the relational data. The extension phase in this third case is not as straightforward as it was in the image and object data cases, but our numerical examples suggest that eNERF has the same approximation qualities that eFFCM and geFFCM do.

A Sensor Web Middleware with Stateful Services for Heterogeneous Sensor Networks

As sensor networks become more pervasive there emerges a need for interfacing applications to perform common operations and transformations on sensor data. Web Services provide an interoperable and platform independent solution to these needs. A key challenge of using Web Services in this context is how to support ongoing sensor queries that persist over an extended period of time. In this paper we introduce Web Service Resource Framework (WSRF) mechanisms into the core services implementation of the NICTA Open Sensor Web Architecture (NOSA). NOSA is a suite of middleware services for sensor network applications which are built upon the OpenGIS Consortiums Sensor Web Enablement standard. WSRF expands the functionality of our services to handle simultaneous observational queries to heterogeneous Sensor Networks. It facilitates the adoption of a multi-user, multi-threaded service environment. Using components from the Globus Middleware platform, NOSA takes a major step forward to achieving the vision of a Sensor Grid.

Moving shape dynamics: A signal processing perspective

This paper provides a new perspective on human motion analysis, namely regarding human motions in video as general discrete time signals. While this seems an intuitive idea, research on human motion analysis has attracted little attention from the signal processing community. Sophisticated signal processing techniques create important opportunities for new solutions to the problem of human motion analysis. This paper investigates how the deformations of human silhouettes (or shapes) during articulated motion can be used as discriminating features to implicitly capture motion dynamics. In particular, we demonstrate the applicability of two widely used signal transform methods, namely the discrete Fourier transform (DFT) and discrete wavelet transform (DWT), for characterization and recognition of human motion sequences. Experimental results show the effectiveness of the proposed method on two state-of-the-art data sets.

On compressing weighted time-evolving graphs

Existing graph compression techniquesmostly focus on static graphs. However for many practical graphs such as social networks the edge weights frequently change over time. This phenomenon raises the question of how to compress dynamic graphs while maintaining most of their intrinsic structural patterns at each time snapshot. In this paper we show that the encoding cost of a dynamic graph is proportional to the heterogeneity of a three dimensional tensor that represents the dynamic graph. We propose an effective algorithm that compresses a dynamic graph by reducing the heterogeneity of its tensor representation, and at the same time also maintains a maximum lossy compression error at any time stamp of the dynamic graph. The bounded compression error benefits compressed graphs in that they retain good approximations of the original edge weights, and hence properties of the original graph (such as shortest paths) are well preserved. To the best of our knowledge, this is the first work that compresses weighted dynamic graphs with bounded lossy compression error at any time snapshot of the graph.

Continuous Detour Queries in Spatial Networks

We study the problem of finding the shortest route between two locations that includes a stopover of a given type. An example scenario of this problem is given as follows: “On the way to Bobs place, Alice searches for a nearby take-away Italian restaurant to buy a pizza.” Assuming that Alice is interested in minimizing the total trip distance, this scenario can be modeled as a query where the current Alices location (start) and Bobs place (destination) function as query points. Based on these two query points, we find the minimum detour object (MDO), i.e., a stopover that minimizes the sum of the distances: 1) from the start to the stopover, and 2) from the stopover to the destination. In a realistic location-based application environment, a user can be indecisive about committing to a particular detour option. The user may wish to browse multiple (k) MDOs before making a decision. Furthermore, when a user moves, the kMDO results at one location may become obsolete. We propose a method for continuous detour query (CDQ) processing based on incremental construction of a shortest path tree. We conducted experimental studies to compare the performance of our proposed method against two methods derived from existing k-nearest neighbor querying techniques using real road-network data sets. Experimental results show that our proposed method significantly outperforms the two competitive techniques.

Distributed training of multiclass conic-segmentation support vector machines on communication constrained networks

We present a distributed algorithm for training multiclass conic-segmentation support vector machines (CS-SVMs) on communication-constrained networks. The proposed algorithm takes advantage of the sparsity of the CS-SVM to minimise the communication overhead between nodes during training to obtain classifiers at each node which closely approximate the optimal (centralised) classifier. The proposed algorithm is also suited for wireless sensor networks where inter-node communication is limited by power restrictions and bandwidth. We demonstrate our algorithm by applying it to two datasets, one simulated and one benchmark dataset, to show that the global decision functions found by the nodes closely approximate the optimal decision function found by a centralised algorithm possessing all training data in one batch.