Population projections are increasingly employed as tools for understanding and modeling the economic, social and environment futures of a limited area. Population growth projection is the mathematical calculation that depend on the future rate, it is dependent on three main component of population assumptions, which are fertility, mortality and net migration of people of a country. Machine learning has been found useful for projecting future values. For population growth to be projected the machine learning has been applied to construct the map between year and population growth. This is germane for population, planning, budgeting, education, commercial sector and the health system. The study investigates the growth rate of Nigerian population data, employing time series projecting machine methods and analyzed by Linear Regression, k-NN, Support vector machine, and decision tree classifiers. The underlining projection method is hinged on the most accurate machine learning algorithm technique that flag less error rate. The increase and decrease of some example dataset does not impact the behavior of the method is equally analyzed. The outcome of the results reveals that the linear regression gives less error rate than the rest classifiers that was used to project population growth of Nigeria.

Classifier Linear regression k-NN Decision tree and Support vector Population growth.