Design and Application of Student Employment Recommendation Platform using Levy K-Means Algorithm in Cloud Technology

Many students face challenges when trying to establish connections with employers. Its aim to serve as a bridge, helping them navigate the job market more effectively. However, the existing models are trapping to the local optima and it takes huge time for finding the global pest pint. To address these issues and overcomes in this paper by using the improved Levy Flight (LF)-based K-means clustering algorithm for grouping the student parameters and job varies based on the existing employs’ details. Initially the Word-2vec is utilized to extracting the meaning full text information’s from the dataset. Then the Levy Flight K-means algorithm (LK) was used to clustering data separately then label it. Then finally the SimRank model was used to generate the final recommendation list. The proposed model is evaluated and compare with the existing model and various metrices such as Root Mean Square Error (RMSE), Mean Square Error (MSE) and Estimated Time (ET) are used. The proposed LK method obtains RMSE of 0.9568 and 0.8657 for MovieLens dataset and Epinions dataset compared to Improved K-means clustering and SimRank (IK-means S).