S. Suguna ¹, V. Sundaravadivelu ², B. Gomathi ¹

Affiliations
1 Department of Computer Science, Sri Meenakshi Govt. Arts College for Women (A), Madurai – 625002, Tamil Nadu, IN
2 PG and Research Department of Computer Science, Thiru A. Govindasamy Govt. Arts College, Tindivanam – 604002, Tamil Nadu, IN

Abstract

Background/Objectives: In this paper we have analyzed various issues with clustering and text mining. The collected documents are preprocessed and grouped using our proposed new algorithm based on position method. We proved our proposed color based constraint clustering algorithm out performs than K-Means and SOM algorithms in terms of time and reliability factors. Methods/Statistical Analysis: We identified the problem after analyzing the existing works with the help of articles from reputed journal papers and national and International level conferences. We proposed the new methodology for document grouping process, and color based constraint clustering process. Clustering can be considered as the most important semi-supervised learning problem which deals with finding a structure in a collection of unlabelled data. In this work the collected documents are preprocessed by stop word removal and stemming process and then the words are grouped according to their similarity using color code constraints. Performances of SOM and Kmeans, and color based constraint algorithms are analyzed for any kind of text document collections. Findings: In this work our proposed color based constraint (CBC) algorithm, SOM and K-Means algorithms performances are compared against time based frequency and reliability of retrieved documents. Here, the time needed to process the number of documents is analyzed. Reliability of retrieved documents can be made by using the number documents and the frequency measurement. We proved our proposed color based constraint clustering algorithm out performs than K-Means, and SOM algorithms in terms of time and reliability. Application/Improvements: Our work is useful for efficient information retrieval process. In future this work can be extended to maximize the grouping of words with minimum latency and one can also extend this work to develop an algorithm for maximize the grouping(clustering) of words in a document with color based constraints to increase the clustering performance for efficient text mining.

Keywords

Color Based Constraint, Clustering, Information Retrieval, Semi_Supervised Clustering Technique, Text Mining

Full Text

   

S. Suguna ¹, C. Ranjith Kumar ², D. Sheela Jeyarani ¹

Affiliations
1 Department of Computer Science, Sri Meenakshi Government Arts College, Madurai - 625002, Tamil Nadu, IN
2 Bharathiar University, Coimbatore - 641046, Tamil Nadu, IN

Abstract

Our paper comprises of almost maximum used state of the art techniques in video retrieval as a brief summary. As there is developments in all fields, media becomes more popular and so people begun to search videos to know the world happenings visually. This paper gives a review of existed combinational methods, algorithms and techniques. The image retrieval was obtained from multiple lower ranking methods. The survey of many methods leads to an introduction of combination of a best method. The main focus of this survey is focused on video retrieval techniques to identify the underlying factors that affect the performance of the combinational method and to judge the effectiveness to provide efficient semantic video retrieval. Finally we descript a broad discussion about the advantages and drawbacks that have been in state of the art. Due to these reasons semantic video retrieval became a challenging issue in various industries.

Keywords

Combinational Methods, State of the art, Techniques, Video Retrieval

Full Text

   

H. Salome Hema Chitra ¹, S. Suguna ¹, S. Naganandini Sujatha ²

Affiliations
1 Department of Computer Science, Sri Meenakshi Government Arts College for Women(A), Madurai – 625002, TamilNadu, IN
2 Department of Master of Computer Science, KLN College of Engineering, Sivagangaii – 630612, Tamil Nadu, IN

Abstract

Background/Objectives: In this paper we have analyzed various image analysis techniques and their issues in seed technology. The crucial role of these techniques is identification and classification of seeds, grading of seeds, quality determination of seeds in seed science and food processing sectors. We provide a complete survey of image analysis techniques and propose a processing module for seed identification and classification. Methods/Statistical analysis: We identified the problem of manual identification and classification of seed in agricultural sector. We provide five important processing steps as a proposed methodology for identification and classification of seed. By using this technology, we able to discriminate the defected seed from normal seed. Our proposed methodology includes noise removal, edge detection, segmentation and classification for seed identification. These methods provide a significant way for seed identification. Findings: In this work, we implemented a five processing module for seed identification. Seed image has taken for the basis of image acquisition and then a seed image is pre-processed by noise removal and image enhancement. Enhanced image goes through the processes of edge detection and segmentation. From the segmented image we extract features like colour, shape and texture for normal and defected seed which may help to identify the seed by image analysis techniques. Application/Improvements: Our work is useful to know about various issues in seed identification in agricultural sector. This work provides efficient methods to overcome the manual identification of seeds. We can also extend this work with an efficient identification and classification methodology in image analysis techniques in agricultural sector

Keywords

Classification, Edge Detection, Feature Extraction, Noise Removal, Segmentation

Full Text