Skip to main content

Next best action prediction

Import libraries​

from ChannelAttributionPro import *

Set the password​

password="mypassword"

Load data​

Data = pd.read_csv("https://app.channelattribution.net/data/Data.csv")

print(Data)

Train the model​

Params=next_best_action_train(Data, var_path="path", var_conv="total_conversions", 
                              var_null="total_null", row_sep="", cha_sep=">", password=password)

print(Params)

Params is a data.frame containing all the conversion rates estimated using a combination of Markov models. Params can be stored and used to predict next best action given a path.

Predict next best action for a given path​

new_path="beta > alpha"
res=next_best_action(new_path,Params,sep=">",password=password)

print(res)

res contains the suggested next action, its conversion rate and the conversion rates for all the available actions.