Informatics Publishing Limited

Amit Sinhal ¹, Bhupendra Verma ²

Affiliations
1 Dept. of Computer Sc. & Engg. Technocrats Institute of Technology, Bhopal, IN
2 Dept. of Computer Sc. & Engg. Technocrats Institute of Technology, (Excellence) Bhopal, IN

Abstract

Accurately estimating the code size, cost, effort and schedule is the most important challenge facing by the software companies and researchers these days. Their key implications for the management of code development as a result of each the overestimates and underestimates have a direct effect to impose destruction on software companies. Heap of models projects over the years by varying researchers for ending effort estimations. Additionally a number of the studies for early stage effort estimations recommend the importance of early estimations. New paradigms supply alternatives to estimate the code development effort, especially the Machine Intelligence (MI) that exploits mechanisms of interaction between humans and processes domain data with the intention of building Intelligent Systems (IS). Among IS, Artificial Neural Network and Fuzzy logic are the most well-liked soft computing techniques for code development effort estimation. The aim of this study is to investigate soft computing techniques within the existing models and to produce a novel design of a soft computing based effort estimation model.

Keywords

Effort Estimation, Soft Computing, Neural Network, Fuzzy Logic, Neuro-Fuzzy, COCOMO, MMRE, PRED.

Full Text

     

Neetesh Gupta ¹, Niket Bhargava ², Md. Ilyas Khan ¹, Shiv Kumar ³, Bhupendra Verma ⁴

Affiliations
1 Department of Information Technology, Technocrat Institute of Technology, Bhopal (M.P.), IN
2 Department of Computer Science, Bansal Institute of Technology, Bhopal (M.P.), IN
3 Department of Information Technology, Technocrat Institute of Technology-Bhopal (M.P.), IN
4 Department of Computer Science, Technocrat Institute of Technology, Bhopal (M.P.), IN

Abstract

For the purpose of content-based image retrieval (CBIR) An up-to-date comparison of state-of-the-art low-level color and texture feature extraction approach is presented in this paper. The CBIR problem is identified by us because there is a need to search the huge databases having images efficiently and effectively. in this paper we suggest a color and texture feature extraction algorithms. Special attention is given for CBIR is the similarity measurement using correlation coefficient with distinct distance matrices properties. A New approach for image retrieval technique is proposed to improve retrieval performance, and reduce the extraction search times. Matching is performed between the test image and the object image and quality of matching is measured in terms of grading.

Keywords

Similarity, Measurement, CBIR, Low Level Feature Extraction, Correlation Coefficient.

Full Text

     

Ankita Jain ¹, Illyas Khan ², Bhupendra Verma ³

Affiliations
1 Technocrats Institute of Technology, Bhopal, India, Masters in Technology in Information Technology, IN
2 Department of IT Truba Institute of Technology, Bhopal, IN
3 Technocrats Institute of Technology, Bhopal, IN

Abstract

The Semantic Web is an extension of the current web in which information is given well-defined meaning, better enabling computers and people to work in cooperation. Web mining is the application of data mining technologies to automatically interact and discover information from web documents, which can be in structured or unstructured form. We present an enterprise framework regarding semantic web mining, which can be used to not only improve the quality of web mining results but also enhances the functions and services and the interoperability of educational information systems and standards in the educational field. For online education system we propose an Ontology-based approach to share online data and retrieve all relevant data about students and their courses. Thus semantic web ontology help build better web mining analysis in educational institute and web mining in-turns helps contract basis more powerful ontology.

Keywords

Semantic Web Mining, Ontology, RDF, XML.

Full Text

     

Manisha S. Patil ¹, Sini Shibu ¹, Bhupendra Verma ²

Affiliations
1 Department of CSE, Technocrats Institute of Technology, Bhopal (M.P.), IN
2 Computer Science & Engineering Department, Technocrats Institute of Technology, University of RGPV, Bhopal (M.P.), IN

Abstract

In the recent years more research towards Indian handwritten character recognition system. Some problems are faced on the classification and recognition. India is the multilingual country, Marathi which is used in Maharashtra state. It is herein proposed the isolated handwritten Marathi numerals recognition system which uses multiple feature extraction methods. Ring data feature extraction method which not till used as per my knowledge for Marathi handwritten numerals recognition. Ring data features are specially, rotation invariant features are those features, which remain constant for rotated and unrotated images. Fourier Descriptor describes the shape in terms of its spacial frequency content. Regular geometrical moments are used for image transformation. The classifier, namely, Minimum distance classifier is used for classification and recognition purpose. This classifier is trained with 16 and 30 dimensional Fourier Descriptors (FD) and Ring data of training samples this recognition system is experimented with a 500 database samples.

Keywords

Fourier Descriptors, Marathi Handwritten Numerals, Minimum Distance Classifier, Regular Geometrical Moments, Ring Data Features.

Full Text

     

Ashish Dehariya ¹, Neetesh Gupta ¹, Bhupendra Verma ²

Affiliations
1 IT Deptt., Technocrats Institute of Technology, Bhopal, IN
2 Technocrats Institute of Technology, Bhopal, IN

Abstract

In its previous part, contribution reviews over the application of artificial neural networks to medical diagnosis and characterizes its advantages and problems in the context of the medical background. Then paradigm of neural networks is introduced and the main problems of medical data base and the basic approaches for training and testing a network by medical data are described. Additionally, the problem of interfacing the network and its result is given and the optimal Back propagation algorithm approach with its flow diagram is presented. On one hand growing neural network and on the other rules based systems works for Medical diagnosis purposes. Diagnostic capabilities of artificial neural network using optimal back propagation based medical diagnosis system is less error prone than fuzzified system of medical diagnosis so it is more advantageous than fuzzification based approach. Proposed diagnosis system has the capabilities to take intelligent decision based on given symptoms of diseases takes as a input and outcomes shows the bacterial infection level so this concept can be used for the purpose of developing Efficient medical diagnosis system and definitely will help to overcome the problems of lack of Experts in Medical fields, wrong diagnosis decisions due to human experts stress also it would help to enhance healthy population amount if such a system will reach to common peoples and use by them.

Keywords

Artificial Feed Forward Neural Networks, Back Propagation, Fuzzification, Medical Diagnosis, Optimal Back Propagation.

Full Text