Null-values imputation using different modification random forest algorithm

<p><span lang="EN-US">Today, the world lives in era of information and data. Therefore, it has become vital to collect keep them a database perform set processes obtain essential details. The null value problem will appear through these processes, which significantly influences behaviour such as analysis prediction gives inaccurate outcomes. In this concern, authors decide utilise random forest technique by modifying calculate values from datasets got University California Irvine (UCL) machine learning repository. scenario consists connectionist bench, phishing websites, breast cancer, ionosphere, COVID-19. modified algorithm is based on three matters number values. samples chosen are founded proposed less redundancy bootstrap. Each tree distinctive features depending hybrid selection. final effect considered ranked voting for classification. This found that executed more suitable accuracy results than traditional relied four parameters sufficient imputing value, grown 9.5%, 6.5%, 5.25% one, two same row datasets, respectively.</span></p>