|
Detection of Unsolicited E-mails
and Summarization by Keyword Extraction
-- Shanmugasundaram Hariharan
Electronic mail (E-mail) serves as a popular mode of communication in every day life. E-mail offers several advantages
like speed of delivery, cheaper cost, acknowledgment report, transparent service and distributed environment.
Managing these e-mails requires huge attention as spammers try to induce large amount of spam or unsolicited mails. This
paper focuses on detecting these unwanted mails, called `unsolicited mails', received by the user. The mail messages are
parsed through a filter that would identify the spam immediately, thereby generating an alert. These mails are then
clustered effectively. The results obtained were promising and provide a platform for further improvements to build a
domain-independent personalizer system. Also, a detailed mechanism to summarize the mail contents is proposed.
© 2010 IUP. All Rights Reserved.
An Analysis of Some
Prime Generating Sieves
-- Alok Chakrabarty and Bipul Syam Purkayastha
Prime number generation is vital to prime factorization and primality testing, and is also used for generating
random numbers. Prime number generation gives a better understanding of the fascinating nature of prime numbers
which helps to generate large primes which are used in public key cryptosystems for e-security. Further, prime
number generation involves heavy number crunching, thus it is also used as a benchmark for comparing the hardware performance
and the capabilities of compilers. Prime number generation programs are among the choicest programs for
demonstrating programming basics to beginners. The present paper thus discusses some commonly used techniques of
generating prime numbers employing the sieve theory. It begins with the famous Sieve of Eratosthenes (SoE) and discusses
some of its efficient extensions. The paper also provides an overview of the Pritchard's wheel sieve technique. Finally
we provide a comparative complexity analysis of the SoE and its quoted extensions.
© 2010 IUP. All Rights Reserved.
ANN Model for Coconut Yield Prediction Using
Optimal Discriminant Plane Method at Bay Islands
-- M Balakrishnan and K Meena
The main focus of this study is to investigate a distributed neural network used to forecast production in Andaman
and Nicobar Islands using weather parameters. The data relating to coconut yield from Central Agricultural Research
Institute (CARI) have been collected for the period 1980 to 2006. The data such as average yearly rainfall, average mean
temperature, relative humidity, wind speed, evaporation and sunshine hours of relevant period (1980 to 2006) have also been
obtained. A multilayer perceptron with backpropagation and optimal discriminant plane method algorithm has been used.
The network is trained using 17 patterns each of nine inputs. In this study, to convert nine inputs into two inputs, the
nine dimensional vectors are mapped into a two-dimensional space by using a transformation. The transformed
two-dimensional vector does not represent any individual feature, instead, it is a combination of nine features with no dimensional quantity.
© 2010 IUP. All Rights Reserved.
Similarity Measures
for Real World Data Mining
-- Nagalakshmi H S and Suhasini M
In real world, data is described not in terms of crisp/singular numeric data but also in terms of multiple data wherein
the data values of each entity are of varied sizes. For an effective data mining implementation, the databases which
contain such multiple values of data have to be analyzed and aggregated to derive relevant meaningful data. This aspect
of analyzing such data is one of the most important and fundamental basis of cluster analysis. The crux of cluster analysis
lies in the design of similarity measures which aim at capturing the degree of similarities between the entities. In this paper,
the authors present two such degrees of similarity measures for multiple valued data types and deal with the clustering of
such multiple valued data considering real life examples. The dataset considered is of 50 individuals and their areas of study
and the areas of expertise in relevance to their fields of study. The degree of similarity perceived between entities or
between the data and the query, denotes the relevance/alikeness in the given data.
© 2010 IUP. All Rights Reserved.
Mizoram Butterflies Data Storage Retrieval
Using Client-Server Technology
-- Brindha Senthilkumar
Primary data are collected from the school of life sciences of Mizoram University. The present project deals with
taxonomical data organization of Mizoram butterfly species. The project is developed based on client-server technology. Visual
basic 2008 with .NET Framework 3.5 is used as front-end (Client) and SQL server 2000 is used as back-end (server). This is
the pioneer work on developing database for Mizoram butterflies.In this paper, addition and deletion of butterfly
species' taxonomical data are accomplished. Uploading and downloading of butterfly images and search mechanisms according
to butterfly family category are designed. Displaying of data is achieved through DataGridView component.
The Mizoram_butterfly database backup and restore mechanisms are implemented through VB.NET. Input data are
validated before adding into the database. The whole package is setup and deployed using VB.NET in-built tool and distributed to
the client terminals of Biotechnology and Zoology Department of Mizoram University.
© 2010 IUP. All Rights Reserved.
Research Note
The Knight's Reach Puzzle
-- Pinaki Chakraborty,
P C Saxena and
C P Katti
© 2010 IUP. All Rights Reserved.
|
