International Journal of Computer Science and Technology Vol. 5 Issue Spl 2
  S.No. Research Topic Paper ID

Fault Detection and Diagnosis of Multivariable Nonlinear Process Using PCA Method

Ajit Kumar Satapathy, Ashoke Sutradhar, Rajesh Kanungo


This paper describes the application of Principal Component Analysis (PCA) for fault detection and diagnosis (FDD) for a process in real time. Principal Component Analysis (PCA) is one of the classical multivariate statistical procedures within the class of linear methods. Fault Detection and Diagnosis (FDD) scheme proposed in this paper is formed by one PCA model per behavior type, i.e., a PCA model for normal operation condition and one PCA model for each of the possible situations. The method of fault detection and diagnosis is based on the definition of a threshold minimum. The PCA model outputs are compared with their threshold Minimum, with and without the faults. One that does not exceed its threshold level gives us an indication of fault. The technique has been applied in real time on two interactive tank system and four types of faults could be detected very efficiently by this method.


Service-Oriented Architecture (SOA) Based Web Application Testing

Arabinda Tripathy, Priyabrata Nayak


Service-Oriented Architecture (SOA) is a flexible set of design principles used during the phases of systems development and integration in computing. A system based on a SOA will package functionality as a suite of interoperable services that can be used within multiple, separate systems from several business domains. SOA also generally provides a way for consumers of services, such as web-based applications, to be aware of available SOAbased services. XML is commonly used for interfacing with SOA services, though this is not required. This paper will provide a comprehensive guidance on best practices for testing SOA Solutions. This document includes a review of the following topics that will need to be addressed to ensure a successful SOA implementation:
1. How do we validate services for functional requirements?
2. How do we manage testing the complexity of testing – 100’s of services?
3. How can end-to-end testing be included in the automated release process?
4. How does access to multiple versions of Web services affect the end-to-end testing process?
5. Multi-user matching of request/responses
6. How can BPEL testing done
7. The key to SOA quality is to involve business analysts


Name Entity Recognition for Odia Language Using Support Vector Machine

Bishwa Ranjan Das, Srikanta Patnaik


This paper talks about a new approach to recognize named entities for Odia languages. This paper reports about the development of a NER system for Odia using Support Vector Machine (SVM). Named Entity Recognition (NER) aims to classify each word of a document into predefined target named entity classes. Starting with named entity annotated corpora and a set of features; it is required to build a base-line NER System. Then some language specific rules are added to the system to recognize some specific NE classes. Also it has been added some gazetteers and context patterns to the system to increase the performance. As identification of rules and context patterns requires language knowledge. To prepare rules and identify context patterns for Odia Language. The system is able to recognize two classes of NEs with 80% f-value, 81% precision value, 80% recall value respectively.


Recommender Systems Survey Using Data Partitioning Clustering

Rajashree Sukla, Pratyush Rn. Mohapatra, Biswadarsi Biswal


The Recommender systems have developed in parallel with the web for distinguish applications. They were initially based on demographic, content based and collaborative filtering .currently these systems are incorporating social information. In the future they will use implicit, local and personal information from the internet of things. This article provides an overview of recommender system as well as collaborative filtering methods and algorithms that how we can collect information for our day to day life using the clustering concept of data mining. It also explains their evolution, provides an original classification for these systems, identifies area of future implementation like collecting vital information about the user’s use of things. And also develop certain areas of future implementation and develops certain area selected for past, present or future importance. Here in this paper the overall recommender system are enlighten and how they are effectively used for the day to day life and commercial life are used.


Performance Improvement of Association Rule Mining Algorithms Through ASRMOLE Approach

Biswaranjan Nayak, Srinivas Prasad


This paper deals with the effective utilization of association rule mining algorithms in large databases used for especially business organizations where the amount of transactions and items plays a crucial role for decision making. Frequent item-set generation and the creation of strong association rules from the frequent item-set patterns are the two basic steps in association rule mining. We have taken suitable illustration of market basket data for generating different item-set frequent patterns and association rule generation through this frequent pattern by the help of Apriori Algorithm and taken the same illustration for FP-Growth association rule mining and a FP-Growth Tree has been constructed for frequent itemset generation and from that strong association rules have been created. For performance study of Apriori and FP-Tree algorithms, experiments have been performed. The customer purchase behaviour i.e. seen in the food outlet environments is mimicked in these transactions. By using the synthetic data generation process, the observations has been plotted in the graphs by taking minimum support count with respect to execution time. From the graphs it has that as the minimum support values decrease, the execution times algorithms increase exponentially which is happened due to decrease in the minimum support threshold values make the number of item-sets in the output to be exponentially increased. It has been established from the graphs that the performance of FP-Growth is better than Apriori algorithm for all problem sizes with factor 2 for high minimum support values to very low level support magnitude.


Implications of Data Mining in Retail Sector of India: From the Recent Growth Perspective

Diptimayee Mishra, SatyaRanjan Mohaptra, Biswadarsi Biswal


Since past decade data mining have got a rich focus due to its significance in decision making and it has become an essential component in various industries. The field of data mining have been prospered and posed into new areas such as manufacturing, insurance, medicine, retail etc. Hence, this paper reviews the various trends of data mining and its relative applications from past to present and discusses how effectively can be used for targeting profitable customers in campaigns.


Comparative Study of Use of RFID Tag to Acquire Data for Safe Mobility of Blind Man

Ghanashyam Sahoo


Blind people need to become as independent as possible in their daily life in order to guarantee a safe mobility. Mobile technology along with RFID can be used to locate persons or objects, can be used to realize navigation systems in an intelligent environment. Such systems open new opportunities to improve the speed, easiness, and safety mobility of the visually impaired person. Using these technologies together with a Text To Speech systems and a mobile based database a cost effective, easy-to-use orientation and navigation system can be developed. Ito provide a comfortable movement in a defined location, by acquiring the location information through different ways of data gathering methods using RFID. But the system is only for selected areas because the dynamic data acquisition requires a vast memory, intelligent system. In this paper a comparative study of acquiring location based information using RFID for the safe mobility of visually impaired person.


Speech Coding Using Code Excited Linear Prediction

Hemanta Kumar Palo, Kailash Rout


The main problem with the speech coding system is the optimum utilization of channel bandwidth. Due to this the speech signal is coded by using as few bits as possible to get low bit-rate speech coders. As the bit rate of the coder goes low, the intelligibility, SNR and overall quality of the speech signal decreases. Hence a comparative analysis is done of two different types of speech coders in this paper for understanding the utility of these coders in various applications so as to reduce the bandwidth and by different speech coding techniques and by reducing the number of bits without any appreciable compromise on the quality of speech. Hindi language has different number of stops than English , hence the performance of the coders must be checked on different languages. The main objective of this paper is to develop speech coders capable of producing high quality speech at low data rates. The focus of this paper is the development and testing of voice coding systems which cater for the above needs.


Short-Term Scheduling of Hydro Thermal System Using Particle Swarm Optimization

M.Madhuri, Rupali Mohanty, Surya Prasad Mishra, Sthitaprajna Rath Krupajal


In recent years various heuristic optimization methods have been developed. This paper presents an efficient and reliable Particle Swarm Optimization (PSO) based solution tosolve short term scheduling of hydro thermal system. The solution approaches based on PSO technique is implemented and demonstrated to solve the hydro thermal scheduling problem with quadratic thermal cost function. PSO algorithm is compare with some well-known heuristic method and results conformed the performance of solving nonlinear optimization problems. PSO algorithms are also capable of finding very nearly global solutions within a reasonable time.


PAPR Reduction of OFDM Signal Using SLM Based Phase Sequence

Madhusmita Panda, Sukant Behera


HIGH Peak-to-Average Power Ratio (PAPR) is a well known drawback of Orthogonal Frequency-Division Multiplexing (OFDM) systems. Selected Mapping (SLM) is a technique used to reduce the peak-to-average power ratio (PAPR) in Orthogonal Frequency-Division Multiplexing (OFDM) systems. In this reported work, rows of normalised Riemann matrices are selected as phase sequence vectors for the SLM technique. MATLAB simulations show PAPR reduction of around 2.3 dB using the proposed method compared with methods reported in the literature.


A Survey Report on Applications of Big Data

ManmathNath Das, Mausumi Parhi, Sharmila Subudhi, Priyabrata Nayak


Now-a-days data is becoming more extreme and well organized, so new data warehousing and data mining technologies have been introduced for the ease of users. Cloud computing plays an important role as it gives organizations the ability to analyze evolutionary data often in a more economically way as it offers computing resources on demand. The introduction of Cloud Computing has helped a lot in the handling of the increasing amount of unstructured data as it is capable of storing data having an unknown or dynamic structure. Also, it delivers the ability to scale more easily and efficiently in a horizontal way, by adding relatively cheap commodity nodes as per its infrastructure. The same applies to big data due to its capabilities to store and handle huge amounts of data in an economically way by effectively making use of parallel and distributed computing.


Mining the Factors Affecting the Failure of Students Pursuing Secondary Education in Rural Areas

P.Sunil Kumar, Ashok Kumar Panda, Satyaranjan Mohapatra


Government of Odisha has taken number of initiatives to motivate the people living in rural areas to promote theirchildren studying in high schools to go for their higher education.Although the statistics show that the overall percentage of success has improved immensely in the last decade, the problem of failures still persists in rural areas. The number of failures is mainly due to three major factorsnamely sentiments, teaching-learning environment and social environment. This paper is based on a survey work which proposes to apply association rule mining measures like support confidence and other interesting measures on these major factors of school failures to understand the problem in a better way and to have a proper planning for the academicians.


A New Approach for Resource Allocation in Cloud Computing

Pandaba Pradhan, Swarnamayee Pani


Today Cloud computing has emerged as a new technology that has got huge potentials in enterprises and markets. Clouds can make it possible to access applications and associated data from anywhere. Companies are able to rent resources from cloud for storage and other computational purposes so that their infrastructure cost can be reduced significantly. Further they can make use of company-wide access to applications, based on pay-as-you-go model. Hence there is no need for getting licenses for individual products. However one of the major pitfalls in cloud computing is related to optimizing the resources being allocated. Because of the uniqueness of the model, resource allocation is performed with the objective of minimizing the costs associated with it. The other challenges of resource allocation are meeting customer demands and application requirements. In this paper, one new resource allocation algorithm is proposed to satisfy the customers demand


An Empirical Novel Architectural Framework for Cloud Computing

Pragnyashree Tripathy, Dr. Srinivas Prasad


Cloud computing has emerged as a new computing paradigm that impacts several different research fields, including software testing. Testing cloud applications has its own peculiarities that demand for novel testing methods and tools. On the other hand, cloud computing also facilitates and provides opportunities for the development of more effective and scalable software testing techniques. Testing of cloud applications has a number of features that make it different from conventional software testing. Mutation testing is a fault-based testing technique that has the advantage of easy automation, and can check the thoroughness of testing performed. This paper provides Novel architectural framework for cloud computing with embedded mutation engine that may lead to better performance.


Smart Techniques of Test Coverage of Object Oriented Program

Aratibala Sahu, Bibhuprasad Sahu, Prasant Kumar, Rajesh Kumar Subudhi


The benefit of coverage analysis is a structural testing technique that helps to eliminate the space or gap in a test suite and find out the position to stop the testing. This is a implementation of new techniques to coverage about the variables with program slicing. Providing the power according to their importance, so that the user can focus on the importance variables to generate the highest test coverage. The present available method to compute the basic coverage based on the program structure matrix. This paper report is bigger than that obtained by the old measure, because the coverage about a variable takes only the related code into the account.


The Improved Potential of Neural Networks for Image Processing in Medical Domains

Pratyush Rn. Mohapatra, Satyaranjan Mohapatra, Rajashree Sukla


The principal method of obtaining physical information about the biological human body is called medical imaging. It is accomplished by creation of specialized images of human body or its parts for clinical purposes. Over the past twenty to thirty years clinical applications are habitually utilizing medical imaging in different forms and helping in better disease diagnostic and treatment. In last decade or so the usage of Neural Networks in applications of Medical Imaging opened new doors for researchers, stirring them to excel in this domain. This paper is the summarized overview of research and development held in recent past highlighting the role of Neural Networks in advancement of Medical Imaging.


Innovative Low Cost Open Source Business Intelligence Systems For MID Size Companies

Priyabrata Nayak, Arabinda Tripathy, Manmathnath Das


BI is one of those buzz phrases that sound super- cool but are often misunderstood. Very small to very large companies use BI to improve decision making, cut costs and identify new business opportunities. BI converts data into useful information and through human analysis to knowledge. Knowledge is power but spending a giant wad of money on fancy BI software won’t do any good. SAP, Oracle, IBM, Microsoft are some of the big names in BI with big price tags. These proprietary vendors sell BI systems as fragmented products, buying all the pieces may prove to be overly expensive. On a feature by feature comparison open source BI tools can’t beat these leading closed source offering. The purpose of this paper is to encourage organizations to develop innovative open source BI solutions and to review the intangible benefits related to in house BI development in industry.


Technological Framework for Management Functions: An ERP Perspective

Debabrata Dash


Now a days, scientific management of organisation requires appropriate applying of information and communication technology tools such as Enterprise Resource Planning (ERP) systems. Therefore, identifying and studying these systems, especially form management perspective, are highly essential. One of the dimensions of this study is to review the relationship between ERP systems and business intelligence, which should be done in the form of studying the requirements for ERP systems with the purpose of supporting management decisions. In this paper, the fundamental expectations of managers of various levels of ERP systems for supporting their Decisions are surveyed. Based on the analysis of management specifications and the decision context of managers, some hypotheses about the relationship between ERP systems and management decisions are studied. In addition, the significance of the requirements for decision support (business intelligence) is investigated via statistical analysis tools. In the rest of the paper, in order to cover the identified requirements for management decisions support and building business intelligence for ERP systems, some Business Intelligence (BI) solutions are recommended. To present a practical framework for applying the solutions, the best order of solution implementation for meeting the requirements is searched by means of Multi Alternative Decision Making (MADM) methods. Further, by using Anthrop Shannon Method, the weight of criteria (requirements), and by applying the Technique Ordered Preference by Similarity to the Ideal Solution (TOPSIS) model, the priorities for implementing the most appropriate solutions are determined.


A Novel Noise Reduction Method For OCR System

Rajesh Kumar Subudhi, Bibhuprasad Sahu, Pratyush Rn. Mohapatra


Document images may be contaminated with noise during transmission, scanning or conversion to digital form. We can categorize noises by identifying their features and can search for similar patterns in a document image to choose appropriate methods for their removal. The majority of binarization techniques are complex and are compounded from filters and existing operations. However, the few simple thresholding methods available cannot be applied to many binarization problems. In this paper, we proposed a local binarization method based on a simple, novel thresholding method with dynamic and flexible windows. The transition method for image binarization is based on the concept of t-transition pixels, a generalization of edge pixels and t-transition sets. We introduce a novel unsupervised thresholding for unimodal histograms to estimate the transition sets. We also present dilation and incidence transition operator to store fine transition set.


Security Aspect of Mobile-Cloud Computing

Rajkumar Saha, Rosy Mishra, Mausumi Parhi, Padmini Patra


Latest advances in mobile communication networks and increasing penetration of smart phones are transforming the mobile Internet and empowering end users with rich mobile experience. However, the limited computing and information/energy storage capabilities of mobile devices are hampering their abilities to support increasingly sophisticated applications demanded by users. The emerging cloud computing technology offers a natural solution to extend the limited capabilities of mobile devices. The resulting new paradigm of mobile cloud computing is being embraced by researchers and practitioners as an exciting new way to extend the capabilities of mobile devices and mobile Platforms, which has the potential for profound impacts on the business environment and people’s daily life. Empowered by mobile cloud computing, previously infeasible mobile applications are finding their ways into mobile devices. As we’re adopting cloud computing, we’re more aware of the security concerns it raises than we were of issues created by other large-scale technologies we adopted in the past. This is a wonderful thing! But security has yet not been achieved. While there’s still plenty of room for cloud providers to improve, many aspects of cloud security must be the responsibility of the consumer. In this paper, we have analysed some security issues that are critical to the security of mobile-cloud computing based project.


Optimization of Total Harmonic Distortion in Multilevel Inverter Using Genetic Algorithm

Rupali Mohanty, Surya Prasad Mishra, M. Madhuri, Sthitaprajna Rath


So far several methods have been presented to optimize the total harmonic distortion in multilevel inverter & maintain the fundamental component at the desired value. This paper focus on elimination of harmonics in a cascaded Hbridge 11 level inverter. The heuristic technique is implemented to find out the minimum THD in cascade multilevel inverter. Using the mathematical theory of resultants, all solutions to this equivalent problem can be found. Theoretical results are verified by simulation and experiments for an 11-level H-bridge inverter. Results show that the proposed method effectively eliminate a great number of specific harmonics, and the output voltage is resulted in low total harmonic distortion.


A Neuro-Fuzzy Hybrid Network for Stock Future Index Forecasting

S. C. Nayak, B. B. Misra, H. S. Behera


In this paper we considered a hybrid network consisting of an Artificial Neural Network (ANN) and fuzzy logic, termed as Fuzzy Neural Network (FNN). A computationally less complex single layer artificial neural network has been considered as the basic neural architecture. The input data is converted to the fuzzy membership functions. The synaptic weight vectors as well as bias values are optimized by a global search optimization technique, Genetic Algorithm (GA). The model has been employed to forecast the one-day-ahead and one-week-ahead closing prices of some fast growing stock markets across the globe. The efficiency of the FNN model has been compared to an ANN trained with an error back propagation based on gradient descent method. The performances of the models are calculated in terms of Average Percentage of Errors (APE). Form simulation studies, it is observed that the hybrid network FNN gives competitive superior results, hence could be adapted as a promising forecasting tool for prediction of stock future indices.


Design of High Speed Ripple Carry Adder Using Mixed Logic Style

Sauvagya Ranjan Sahoo, Dr. Kamala Kanta Mahapatra


This paper presents a circuit design approach to perform very fast carry computation in cascaded full adder. The proposed approach consists of insertion of transmission gate adder between full adders designed by Feedthrough Logic (FTL) approach. A 16-bit ripple carry adder is designed by this proposed approach. Analysis shows that the insertion of transmission-gate full adder between FTL full adders does not violate the timing constraint and perform correct operation. Then a comparison analysis has been carried out by simulating the proposed adder architecture in 0.18 μm technology. The simulation results shows that the proposed adder improves the speed by a factor of 1.157 as compared existing FTL Ripple carry adder.


Designing A Hybrid Page Ranking Algorithm For Semantic Web Search Engine

Anisur Rehman, Sharmila Subudhi, Manmathnath Das, Yasobanta Rout


Now-a-days people rely on to search their required information on the web. In such a scenario it is up to the service provider to provide proper, relevant and quality information to the internet where user can submit their query and find out the result. But it is a challenge for service provider to provide proper, relevant and quality information to the internet user by using the web page contents and hyperlink between the web pages. The nextgeneration Web architecture, represented by the Semantic Web, provides the layered architecture possibly allowing overcoming this limitation. This paper deals with a hybrid approach of page ranking algorithm including the searching of spam pages which, if removed, will increase the speed of query execution.


Data Retrieval in Peer to Peer Network Using Text Mining

Smruti Smaraki Sarangi, Gyanendra Singha


This paper describes about the concept of data retrieval in a peer to peer system that is Gnutella software using Text Mining. Here we have used the ideal term-based pruning algorithm which is the modification of old pruning algorithm. This algorithm is used in text mining which retrieves data in Gnutella Peer to Peer network. It performs unsupervised learning on query results to discover information about the collection and uses this information on future queries to prune the results using some concept of text mining like precision, recall and F-measure concept. This algorithm is also used to prune the index data. The ideal term-based pruning algorithm is implemented in Gnutella System to retrieve the data and trace the result with the help of network simulator which is purely based on mathematical calculation. The result shows the differences comparing to the theoretical results which we used to obtain the data from the pruned index and the original index.


Load Shedding Strategy Using Fuzzy Logic

Srikant Kumar Dash, Sunil Kumar Mahapatro, Sudhansu Bhushan Pati, Prasanta Kumar Jena


As a perspective to ensure the voltage and fre-quency electrical network stability, the load shedding consti¬tutes a desirable action to maintain the network service quality and to control its vulnerability. In this paper, we propose a new intelligent load shedding strategy applying fuzzy control algo¬rithms. This strategy is based on the estimate, in real time, of the load quantity to shed. Calculation algorithms containing fuzzy controllers generate command vectors ensuring the load shed¬ding of a pre calculated proportion loads in order to reestablish the power balance and to lead the electrical network to a new stable condition.


Model of an Emotionally Cooperative System

Subhrajyoti Sahoo


People use devices in every emotional state.The nature of activity differs in different emotional state, for example we have a specific multimedia file for a particular feeling.By monitoring such behavior we can make the device capable of inferring the emotion of the user. And then we can design particular response. Presently the systems that can recognize emotional pattern of the user require costly equipment. This can hinder thepropagation of the devices. The solution presented in this paper is cost effective and can be integrated to any mainstream computing hardware.This paper presses on the need to integrating the technologies that are already innovated. We are dealing with small data using a tool innovated for big data and a methodology invented for biometrics. Because users are most honest with their devices, designing such an emotion sensing device requires respect for privacy of the user.


E-Learning 3.0: Agent Supervised Collaborative Learning in Web 3.0

Subrat Prasad Pattanayak, Sujata Dash


Today E-Learning has become a widely accepted paradigm for professional development and are expanding in academic setting, driving interest in lifelong learning. It has been evolving along with the changing technologies becoming more mature, faster and newer. The advent of the latest generation of World Wide Web has also impacted E-Learning. Hence with the transformation of the Web2.0 to Web3.0, E-learning has also got a new identity that is E-Learning 3.0.The basic technology on which this 3.0 concept is founded on is Artificial Intelligence specifically intelligent Multi Agents. It supports learning by interacting with author and learner in different manner and collaborates and coordinates the flow of content and knowledge in an advanced environment. So the role of intelligent multi agents will be very vital to give sufficient cooperation to both author and learner in E-Learning 3.0. This paper addresses the evolution of Web 3.0 and E-Learning 3.0, with a special reference to Intelligent Multi Agents for its possible potential to influence the new generation of Online Learning.


Prediction of Financial Time Series Databases Using an Efficient Hybrid Model

S. Chakravarty, P. Mohapatra


The increasing importance of time series prediction in business, science and engineering has drawn the attention of researchers, engineers and scientists. In this paper, the focus is mainly on the prediction of the most volatile financial time series data i.e. electricity price. The prediction of electricity price is necessary for both power producers as well as power consumers to maximize their benefits and utilities respectively. In this study an efficient Fuzzy Neural Hybrid Model (FNHM) is proposed to predict electricity prices for one hour in advance. The proposed hybrid model is consisting of a trigonometric functional link artificial neural network (FLANN) and fuzzy logic system. The FNHM uses a functional link neural network to the consequent part of the fuzzy rules. The consequent part of the model is a non-linear combination of input variables. The most accepted back propagation (BP) learning algorithm is used to train the parameters of the model. Further, a stochastic derivative free evolutionary genetic algorithm (GA) is used to further optimize the parameters to achieve more prediction accuracy and overcome the drawback of BP algorithm. The prices of Poland electricity market are taken as experimental data. The Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) are used to find out the performance of the proposed model. The MAPE and RMSE are found to be less when compared to other neural model like FLANN.


Breast Cancer Diagnosis and Prognosis in India: A Comparative Study Based on Data Mining Techniques

Sushree Swagatika, Smruti Ranjan Barik


Breast Cancer Diagnosis and Prognosis are two medical applications that pose a great challenge to the researchers. The use of machine learning and data mining techniques has revolutionized the whole process of breast cancer Diagnosis and Prognosis. Breast Cancer Diagnosis distinguishes benign from malignant breast lumps whereas Breast Cancer Prognosis predicts when Breast Cancer is likely to recur in patients that have had their cancers excised. Thus, these two problems are mainly in the scope of the classification problems. This paper compares different techniques applied in diagnosis and prognosis of breast cancer.


A Comparative Study on Detection Methods of Wood-Borers

Sushree Swagatika, Smruti Ranjan Barik


As per the State of Forest Report 2003, the total forest cover in the country is 67.83 mha, which constitutes 20.64% of its geographical area. This paper begins by looking at the problem, invasive species, which are a threat to India which are the reason for the development of the different detection methods. This paper concludes with a comparison of detection methods for detecting wood-borers.


An Approach for Designing Odia Spell Checker

Yasobanta Rout, Prabhat Ku.Santi, Sharmila Subudhi, Bibhuprasad Sahu


The aim of this paper is to provide an unified approach for designing and implementing the Odia spell checker and the same approach can be extend to any Indian language spell-checker & corrector. Odia language is rich in inflective, derivative and compound words. The most important factor affecting the spelling errors is the presence of large number of phonetics similar characters and modifiers, used to modify consonants and conjuncts. Implementing a spellchecker for highly inflected Indian language like Odia is a most challenging research task that we have carried out in this research project. We present a new spell-checking system for detecting the misspelled words and to find out the most appropriate suggestive words using the minimum edit-distance algorithm. The spell-checking task can be split into two parts, i.e. detection and actual correction of the spelling errors. All the misspell words are marked and allowed for correction. The basic framework has been designed and implemented in Java programming language and can be integrated any java based applications.