An Efficient Way of Classifying and Clustering Documents Based on SMTP

Abstract

In text processing, the similarity measurement is the important process. It measures the similarities between the two documents. In this project we proposed the new similarity measurement. The computation of similarity measurement is based on the feature of two documents. Our proposed system contains three case to compute the similarity. The three cases are, both two documents contains features, only one document contains feature, there is no feature into the two documents. In first case, the similarity is increased when the differences of feature value is decreased between the two documents. Then the given differences are scaled. In second case, fixed value is given to the similarity. In third cases there is no contribution to the similarity. Finally our proposed measure method achieves the better performance compared than other measurement methods.

Authors and Affiliations

U. Umamaheswari, Mr. G. Shivaji Rao

Keywords

Related Articles

LTE Techniques in 4G Wireless System Using Protocol Stack

LTE, an acronym for Long-Term Evolution. 4G LTE is a standard for wireless broadband technology designed to support roaming Internet access via cell phones and handheld devices. Due to the requirements on high data rate...

Active Shape Model Based Pose Estimation Using Hausdorff Matching

The Oxford College Of Engineering, Bangalore, IndiaAbstract— paper presents a novel algorithm for automatic locating, extracting and recognizing human pose. The proposed algorithm consists of four stages; (i) training o...

Structural Health Monitoring on RCC Slab by Use of Piezoelectric Material

Piezoelectric lead zirconate titanate (PZT) is being gradually applied into practice as a new intelligent material for structural health monitoring. In order to study the damage detection properties of PZT on concrete s...

Cube Divisor Cordial Labeling For Some Graphs

Let G = {V (G), E (G)} be a simple graph and 􀢌 ∶ 􀢂(􀡳) → {􀫚, 􀫛,· · · , |􀢜|} be a bijection. For each edge 􀢛􀢜 , assign the label 1. If either [􀢌(􀢛)]􀫜 /􀢌 (􀢜) or [􀢌(􀢜)]􀫜 /􀢌 (􀢛) and the label 0 otherwise 􀢌 is called a cube d...

Experimental Study on Bricks Using Water Treatment Sludge

Sludge resulting from water treatment plants creates problems of disposal. The sludge generated in most of the treatment systems around the world is discharged into the nearest watercourse, which leads to accumulative r...

Download PDF file
  • EP ID EP19022
  • DOI -
  • Views 293
  • Downloads 11

How To Cite

U. Umamaheswari, Mr. G. Shivaji Rao (2014). An Efficient Way of Classifying and Clustering Documents Based on SMTP. International Journal for Research in Applied Science and Engineering Technology (IJRASET), 2(11), -. https://www.europub.co.uk/articles/-A-19022