| With the rapid expansion of e-commerce website, the types of goods, provided customers with by web-based portal, are also increasingly diverse. On the one hand, It is to meet the diversified needs of the customer, on the other hand customers spend more time to browse the commodity than ever.Finally they lost in a large number of commodities and can not find the most satisfied goods. Therefore, in order to enhance the customer shopping experience and thereby to increase sales, the e-commerce portal need a guide as a professional guide in Shop to help customers to choose the goods with satisfied by them. In order to make the online shopping experience of the customer more natural, this paper presents research on the subject of intelligent shopping guider robot system based on natural language processing. The system allows the shopping guide robot and customers to exchange with Exchange with daily life term, the system initiatively grab the customer's interests and other characteristics to recommend customers for the goods satisfaction by them.Smart shopping guide robot system is essentially a "Smart QA+recommended" system. The workflow of the system: to match the semantic template of pre-stored in the system with ther input and then give customers the corresponding answer in the output template library; by grabing the interests of customers in the dialogue, to predict customer's purchase intention for recommending customers for the goods. So, the main work of this paper include:to manually construct semantic template Knowledge; to design recommended product algorithm.The innovation of this paper is:firstly, the system can give customers a response through a simple segmentation and matching and semantic templates; secondly, the system can also achieve the function of recommending goods for the customers,through automatly building,and updating the Knowledge of individual users to predict customer's purchase intention, in this paper, the intelligent shopping guide robot system has passed the3653G mobile phone store terms quiz tests and has proved that it is feasibie for the areas of expertise. |
