Research on Recommendation System Using Big Data Based on Deep Learning
Man Li, Allam Maalla, Sanhua Liu · 2021 IEEE 2nd International Conference on Information Technology, Big Data and Artificial Intelligence (ICIBA) · 2021
According to the recommendation algorithm based on deep learning, this article will verify the validity of the recommendation model through Item2vec and DeepFM models. To analyze the relationship between historical user behavior data and the users of the product are interested in, the Spark SQL scripts can be run according to statistical analysis methods, and generate the training set and test set, which is needed by train the model, and build predictive models, and get the collection of products that each user is interested in and improve the prediction effect through model optimization. Experimental verification and analysis show that the recommendation system can handle massive amounts of user and product data. It can also extract features from these data for training automatically, which is conducive to improving the quality and efficiency of users' shopping decisions. we first complete feature extraction based on the processed data set. At present, the most used feature extraction method is feature extraction based on artificial experience and statistical methods. However, because the actual application scenarios of the recommendation system are often more complicated, we used Spark SQL scripts to extract features in experiments. First, extract the experimental data from the distributed database, and use the workflow scheduling script to generate the data set required for model training and testing.