Demand Prediction Of Private Cars With Parallel Multi-Task Learning Model Using Ridge Linear Classification In Comparison With K-Nearest Neighbour Algorithm For Better Accuracy

The research work intends to enhance the precision of forecasting private car demand by applying Ridge Linear classifier algorithm and comparing its accuracy with K-Nearest Neighbour (KNN) algorithm. Materials and methods: The UCI repository database were used to download the experimental data. The total number of iterations for each group is 10 iterations with total iterations of 20 calculated by performing an independent sample t-test with a Gpower of 0.8 along with alpha quality and beta qualities of 0.05 and 0.2 respectively also the Confidence Interval (CI) of 95%. The K-Nearest Neighbors method (KNN) and the Ridge Linear classifier models were used to predict the demand for private automobiles. Results: As compared to the Ridge Linear classifier model, which has more accuracy rate of 71.36 percent, when compared to the K-Nearest Neighbours method (KNN) classifier has a 70.90 percent. The significance value was analyzed using the independent sample t- Test analysis where p<0.05 and this work showed a significance value of p=0.19. There is no significant difference between the two groups. Conclusion: The accuracy of Ridge Linear Model is 71.36% is higher compared with K-Nearest Neighbors algorithm that has an accuracy rate of 70.90% in analysis of to predict the demand of private Cars with improved accuracy rate.