CLOUD SERVICE QoS PREDICTION ALGORITHM BASED ON TIME PERCEPTION AND USER'S INTEREST

In this study, a cloud service QoS algorithm based on the time perception and the user's interest (denoted by CSQI) is proposed to recommend the cloud service precisely to the users for meeting their demand. Different from the current QoS value prediction, CSQI processes the data in the earlier stage, removes the negative correlation data, and improves the predictive data ratio of the positive correlation by fully considering the multiple time-slice perceptual ranking. Moreover, the effect of the user's interest of the regional similarity in the period time slice on QoS is emphasized. In addition, the service curve of the regional similarity user in the multiple time-slice is fitted based on gray relation theory, and the first K users of the ranking synthesis similarity are weighted to avoid the loss of the QoS history data. Experimental data indicate that the CSQI algorithm has better precision than other algorithms.