V. Jawahar Senthil Kumar

Evaluating the Performance of Similarity Measures Used in Document Clustering and Information Retrieval

This paper presents the results of an experimental study of some similarity measures used for both Information Retrieval and Document Clustering. Our results indicate that the cosine similarity measure is superior than the other measures such as Jaccard measure, Euclidean measure that we tested. Cosine Similarity measure is particularly better for text documents. Previously these measures are compared with the conventional text datasets but the proposed system collects the datasets with the help of API and it returns the collection of XML pages. These XML pages are parsed and filtered to get the web document datasets. In this paper, we compare and analyze the effectiveness of these measures for these web document datasets.

Shallow NLP techniques for noun phrase extraction

The field of Information Retrieval plays an important role in searching on the Internet. Most of the information retrieval systems are limited to the query processing based on keywords. In information retrieval system the matching of the query against a set of text record is the core of the system. Retrieval of the relevant natural language text document is of more challenge. Todays most search engines are based on keyword based (bag of words) techniques, which results in some disadvantages. For text retrieval key phrases can help to narrow the search results or rank retrieved documents. We exploit shallow NLP techniques to support a range of NL queries and snippets over an existing keyword-based search. This paper describes a simple system for choosing noun phrases from a document as key phrases. The noun phrase extractor is made up of three modules: tokenization; part-of-speech tagging; noun phrase identification using Chunking. A preliminary evaluation was conducted to test this technique with the standard IR benchmark collections such as classic test collections and then with the web snippets collection from the search engines results. The experimental results have been encouraging.

Decision-based neighborhood-referred unsymmetrical trimmed variants filter for the removal of high-density salt-and-pepper noise in images and videos

A fixed

Firefly Algorithm Approach for Localization in Wireless Sensor Networks

3\times 3

Cross layered adaptive rate optimised error control coding for WSN

3×3 decision-based algorithm is proposed for the enhancement of images, and videos that are heavily corrupted by salt-and-pepper noise are proposed. The algorithm uses unsymmetrical trimmed variants for the noise removal. The corrupted pixel is replaced based on the number of non-noisy pixel in the current processing window. The proposed algorithm was applied on various grayscale, and videos that gave excellent peak signal-to-noise ratio, high image enhancement factor, low mean square error, and very good SSIM with excellent edge preservation even at high noise densities. If all the pixels of the current processing window are noisy, then instead of unsymmetrical midpoint, global trimmed mean of the image is replaced as output. The proposed algorithm shows excellent noise suppression capability, when compared to standard and existing filters in terms of both qualitative and quantitative measures at highly noisy environment.

Energy efficient, obstacle avoidance path planning trajectory for localization in wireless sensor network

Abstract In a smart and decision making environment the location information of the sensors and devices under monitoring and control, is very much important, otherwise the sensed data becomes meaningless. This paper proposes three intelligent algorithms namely differential evolution localization algorithm, firefly localization algorithm, and a hybrid firefly differential evolution localization algorithm for wireless sensor networks localization problem. The proposed algorithms are range based and distributed localization algorithms. The algorithms are studied, analyzed and compared with respect to time complexity, convergence and accuracy of the estimated location information.

Performance investigation of multi-relay cooperative diversity networks☆

Development of sensor technology has led to low power, low cost and small sized distributed wireless sensor networks (WSN). The self organizable and distributed characteristics of wireless sensor networks had made it for monitoring and control applications at home and other environment. In most of these applications location information plays a crucial role in increasing the performance and reliability of the network. This paper makes an attempt in analyzing and implementation of a novel nature based algorithm known as firefly localization algorithm. This is a distributed algorithm which uses range based trilateration method for distance measurement required for estimating the location of the sensor node. The algorithm is simple to implement and has better convergence and accuracy.

Unsymmetrical Trimmed Midpoint as Detector for Salt and Pepper Noise Removal

This work aims a cascaded decision based filter for the removal of high density in image and video is proposed. The proposed scheme operates in two levels. First level is a decision based median filter and the later employs a decision based unsymmetrical trimmed variants. In the first stage the corrupted pixels are found and replaced by the median of the current processing window. The second level again inspects for the occurrence of outliers and eliminates them using unsymmetrical trimmed variants depending upon the content of the current processing window. The Main idea of the proposed algorithm is to formulate an algorithm that eliminates the outliers for high density outlier noise. The proposed algorithm is found to exhibit excellent noise elimination capability for noise densities as high as 90%. The proposed scheme shows better results quantitatively and qualitatively in image and video than the standard and other cascaded algorithms.