A hybrid collaborative filtering mechanism for product recommendation system

The collaborative model is the needed framework to find a good product in both user- and budget-friendly. These collaborative filtering models have provided product recommendations based on the ratings. Recently, neural networks such as Convolution neural models, recurrent neural networks, boosting models and optimization procedures were implemented for the recommendation system to find the accurate product rating. But, due to the vast amount of data, less accuracy score was reported for finding the book ratings. Considering these issues, the collaborative filtering model has been introduced in the recommendation system. However, this collaborative filtering is only effective for small data, and large data requires additional intelligent models that have maximized the error rate. So, the current research article has aimed to design a novel chimp-based Deep Neural Collaborative Filtering (CbDNCF) for the recommendation system. Initially, the dataset was filtered in a preprocessing layer of the novel CbDNCF. Consequently, the noiseless data is trained in the classification layer. Further, the feature extraction and highest rating prediction process were performed. Incorporating the Chimp functions in the deep neural classification layer has afforded the finest forecasting outcomes. Consequently, the designed model is tested using the Python programming language with book product datasets. Its robustness is measured by measuring the key metrics with other existing models. Hence, the planned approach has earned good prediction results of 97.7% accuracy and the lowest error rate of 0.02%, which is quite better than the associated models.