An Iterative PSO for Web worth Optimization through random velocity

Efficiently updating the discovered association rules thus becomes a crucial issue. In this paper, we have considered the problem of web data mining. These days the Internet has been well known as an enormous information storehouse comprising of a mixture of information sorts and a lot of inconspicuous educational learning, which can be found through an extensive variety of information mining or machine learning standards. All these sorts of strategies are focused around shrewd registering methodologies, or thereabouts called computational discernment, which are broadly utilized as a part of the exploration of database, information mining, machine learning, and data recovery. In our approach we first prepare the dataset. The dataset is considered from Google Trends. Google Trends is the part of Google. In this we have considered the data of the colleges from the 6 months. The six months data is considered for each website. The transaction ratings based on the apriori algorithm. It is based on 30 % minimum support. Based on the apriori algorithm we have obtained the associated ranking. The ranking is then applied to the particle swarm optimization. Then we apply random velocity PSO for predicting the future trends. The major difference between random weight PSO and typical PSO is that the velocities and positions of the particles are defined in terms of the changes of probabilities and the particles are formed by integers in {0, 1} with the limitation set and it will be randomized in each iteration. The archived results by this algorithm are efficient and optimal in comparison to the trends.