Authors: D. Yuvaraj; Mavaluru Dinesh; M. Sivaram; S. Nageswari
Addresses: Department Computer of Science, Cihan University-Duhok, Kurdistan Region, Iraq ' Department of Information Technology, College of Computing and Informatics, Saudi Electronic University, Kingdom of Saudi Arabia ' Department of Computer Networking, Lebanese French University (LFU), Erbil, Iraq ' Department of Computer Science, Bharath Niketan Engineering College, Theni, India
Abstract: Some data is historical such as the judicial or medical which the factor of time is very important. Temporal databases can deal with these data because it provides a systematic way of dealing with historical data. The temporal data mining deals with these types of data, that has time stamping and influences by the factor of time after mining. The relevance feedback technique is included in the individual's revisitation habits and memory strength. We have additionally executed and assessed the performance by utilising a trace driven methodology dependent on the online real behaviour dataset. We also presented the large level datasets introducing a profile that encode the user subjective notation of similarity in domain. These profiles can be gained ceaselessly from connection with client. We further show how the client profile might be embedded in a system that utilisation relevance feedback mechanism. At long last, we compute the scalability of our system by utilising various datasets from the various domains.
Keywords: temporal data; relevance feedback; skyline pyramid; scalable framework; data mining; behaviour dataset; scientific strategies; uniform resource locator; time-stamp data; pattern; web revisitation.
World Review of Science, Technology and Sustainable Development, 2022 Vol.18 No.1, pp.20 - 30
Received: 09 May 2019
Accepted: 30 Dec 2019
Published online: 29 Oct 2021 *