Findability Platform Predict Plus

Discrete Target Variable Data - Binary

FP Predict+ identifies if the target variable data is binary and models the data for binary classification. You can choose one of the three modes of modeling: 1. Model for Least Frequent Target variable Value (LF) 2. Model for Most Frequent Target variable Value (MF) 3. Model for Both Values of Target Variable (BV) In case of LF and MF modes, you get decile wise performance analytics and in case of BV mode, you get confusion matrix with precision, recall and AUC-RoC values for model evaluation

Discrete Target Variable Data - Multiclass

FP Predict+ identifies if the target variable data is discrete with multiple values and models the data for multi-class classification. You can model for data with high number of classes easily. You get confusion matrix with precision, recall and AUC-RoC values for model evaluation

Continuous Target Variable - Estimation

FP Predict+ identifies if the target variable data is continuous and models the data for multi-model estimation. It provides you multiple models from the data and selects best model for each record from Prediction file. It provides you ME, RMSE, MAE, MPE, MAPE and MASE for model evaluation

Time Series Forecast

FP Predict+ can handle time series data and perform modeling and forecast. Your data can be at a time interval of daily, weekly, monthly or yearly. You can provide data with multiple group-ids and FP Predict+ handles each group independently. It provides you ME, RMSE, MAE, MPE, MAPE and MASE for model evaluation