Research Article

The Recommending Agricultural Product Sales Promotion Mode in E-Commerce Using Reinforcement Learning with Contextual Multiarmed Bandit Algorithms

Table 6

The average performance of the last 7 days of the algorithm updated daily under 20 products.

AlgorithmIncomeScore

LinUCB9509.523.1692
Hybrid-LinUCB12104.625.3376
hLinUCB9515.722.7945
CoLin8502.621.6711