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.
[1]
Xiaoli Z. Fern,et al.
Active Learning of Constraints for Semi-Supervised Clustering
,
2014,
IEEE Transactions on Knowledge and Data Engineering.
[2]
Raymond J. Mooney,et al.
A Mutually Beneficial Integration of Data Mining and Information Extraction
,
2000,
AAAI/IAAI.
[3]
Markus Junker,et al.
Learning for Text Categorization and Information Extraction with ILP
,
1999,
Learning Language in Logic.
[4]
Derek Greene,et al.
Constraint Selection by Committee: An Ensemble Approach to Identifying Informative Constraints for Semi-supervised Clustering
,
2007,
ECML.
[5]
Fakhri Karray,et al.
An Efficient Concept-Based Mining Model for Enhancing Text Clustering
,
2010,
IEEE Transactions on Knowledge and Data Engineering.
[6]
Raymond J. Mooney,et al.
Integrating constraints and metric learning in semi-supervised clustering
,
2004,
ICML.
[7]
S. S. Ravi,et al.
A SAT-based Framework for Efficient Constrained Clustering
,
2010,
SDM.
[8]
Thomas Hofmann,et al.
The Cluster-Abstraction Model: Unsupervised Learning of Topic Hierarchies from Text Data
,
1999,
IJCAI.
[9]
Ian Davidson,et al.
Constrained Clustering: Advances in Algorithms, Theory, and Applications
,
2008
.
[10]
Ido Dagan,et al.
Knowledge Discovery in Textual Databases (KDT)
,
1995,
KDD.
[11]
Fakhri Karray,et al.
Enhancing Text Clustering Using Concept-based Mining Model
,
2006,
Sixth International Conference on Data Mining (ICDM'06).
[12]
Rong Jin,et al.
Active query selection for semi-supervised clustering
,
2008,
2008 19th International Conference on Pattern Recognition.
[13]
George Karypis,et al.
A Comparison of Document Clustering Techniques
,
2000
.