International Journal of Computer Applications (0975 – 8887) Volume 129 – No.10, November2015 39 Comparison of Swarm Intelligence Techniques for Improved Information Retrieval System Priyanka J. Howale Department of Information Technology Pimpri Chinchwad College of Engineering Pune, India Sanketa S. Pradhan Department of Information Technology Pimpri Chinchwad College of Engineering Pune, India Shraddha G. Lohar Department of Information Technology Pimpri Chinchwad College of Engineering Pune, India Mehul D.Redekar Department of Information Technology Pimpri Chinchwad College of Engineering Pune, India Anagha N. Chaudhari Department of Information Technology Pimpri Chinchwad College of Engineering Pune, India ABSTRACT Optimization is an important and critical step in the data mining process and it has a huge impact on the success of a data mining process. Selecting a set of feature which is optimal for a given task is a problem which plays an important role in a wide variety of context including pattern recognition, adaptive control and machine learning Clusters are formed of the reduced dataset using Swarm Intelligence Technique algorithms i.e. Particle Swarm Optimization(PSO),Ant Colony Optimization(ACO),Cluster Hypothesis is verified which is the intra cluster distance should be minimum and inter cluster distance should be maximum. Most relevant documents are stored i+n the clusters An Information Retrieval System is used for retrieval of data from the clusters. When user enters a query from a Graphical User Interface, using Information Retrieval algorithm the document is searched and retrieved from the clusters. It is then given as an output to the user Keywords Optimization, Swarm Intelligence Technique, Clusters. 1. INTRODUCTION For the complex data sets there is a problem in retrieval the necessary information from particular records. As the original datasets are multidimensional in nature, so for retrieving the particular information, datasets need to be multidimensionality reduced. Hence, for these there are different optimization techniques or algorithms , and with the help of those algorithms the datasets are first reduced and then that datasets are provided as an input to the algorithms i.e Particle Swarm Optimization, Ant Colony Optimization and Then clusters are obtain for information retrieval system[1]. Evolutionary algorithms technique:  Swam Intelligence(SI) Swarm Intelligence algorithms are -  Particle Swarm Optimization(PSO)  Ant Colony Optimization(ACO) Using all these algorithms and with the help of comparison between these algorithms there is a retrieval of information from the particular data sets, and as well as development of IR system also takes place. 2. MODULES Module 1(Study and Preparation of Dataset): Input Dataset is required for optimization of IR using Evolutionary Algorithm. Here Instead of applying already available dataset for System , preparing new dataset in this module. Module 2 (Comparison of Swarm Intelligence Technique algorithms) Comparison of Swarm intelligence techniques algorithms based on computation time, reduction of dataset using optimization. Algorithms:  Particle Swarm Optimization  Ant Colony optimization Module 3 ( Formation of Cluster ) Best Algorithm between these four Algorithms are selected and clusters are form for this technique. Best algorithm is identified on the basis of their performance and accuracy, efficiency, time complexity. Algorithms:  Particle Swarm Optimization  Ant Colony Optimization Module 4 ( Information Retrieval ) Clusters obtain for the Best algorithm are used for information retrieval from system .There are many applications of IR so any kind of information may retrieve from the system .information retrieval is based on the input given to that System(Input Dataset).