Personalized news recommendation is important for users to find interesting news from massive information. As a heated topic that has wide applications in the industry, it has been extensively studied over decades and has made huge progress. In this paper, we discuss the topic of news recommendation. We implement an attention based model and demonstrate the great performance-boosting modern deep learning techniques bring in. The following chapter will be structured as follow: in chapter 2, we will briefly introduce the scenario of news recommendation. In Chapter 3, we discuss the details of the implementation of our model. In Chapter 4, we introduce the dataset we use. Chapter 5 will be a recap and conclusion Figure.1.