Estimation of Data Prediction using WSM Method

Existing Traffic Flow Forecasting Methods Mainly predicts shallow traffic Use models, and many are real still not satisfactory for worldly applications. This situation prompts a rethinking of Traffic flow is massive deep architectural models basically a prediction problem Traffic data. In this paper, useful missing data for ensemble filtering We propose an estimation algorithm. A Other similar users (items) to user (item) If so, our approach accounts for missing data Predicting and informing users, products or How to recover missing data using both Determines what to predict. Alternative: TUNNEL, SOIL, INDOOR, WATER and RANDOM FOREST. Evaluation option: MAE, SD of MAE, MAPE, MAPE of SD from the result it is seen that RANDOM FOREST and is got the first rank whereas is the WATER got is having the lowest rank. The value of the dataset for Range of Data Prediction in WSM (Weighted sum model) Method shows that it results in RANDOM FOREST and top ranking.