To develop a recommendation system that suggests products, movies, or songs to users based on their previous preferences, you will need to use collaborative filtering, content-based filtering, or a combination of both.

Collaborative Filtering: This method makes recommendations based on the preferences of similar users. You can use techniques such as user-based or item-based collaborative filtering to make recommendations.

Content-based Filtering: This method makes recommendations based on the characteristics of the items that a user has previously liked. You can use techniques such as natural language processing and machine learning to extract features from the items and make recommendations based on those features.

Hybrid: You can also combine both Collaborative Filtering and Content-based Filtering for better recommendation results.

To implement these methods, you can use pre-trained models or libraries such as Tensorflow, scikit-learn, Surprise, etc. Additionally, you'll need to have a good understanding of the specific domain or topic that your recommendation system will be operating in, as well as a large dataset of user preferences and item information.

It's important to note that creating a recommendation system is a challenging task, and it's recommended to have a good understanding of machine learning and data science. Additionally, it's important to evaluate and test the performance of your recommendation system using metrics such as precision, recall, and F1-score.