The demands placed by designers on workpiece performance and functionality are increasing rapidly. Important aspects of manufacturing’s contribution to the fulfillment of these demands are the conditions at the work piece edges. However, Burrs are often created on the workpiece edges in micro-machining. In many cases, time consuming and expensive deburring processes have to be applied in order to ensure the desired part functionality. Burrs make troubles on production lines in terms of deburring cost, quality of products and cutting tool wear. To prevent problems caused by burrs in micro-machining, prediction and control of burr size is desirable. Experimental studies show that burr formation in micro-milling is a highly complex process depending on a number of parameters such as material properties, tool geometry and cutting parameters. It is very difficult to establish the relationship between burr sizes and cutting conditions. A web-based micro-machining burr expert system for burr sizes prediction and control was developed using ASP.NET platform. Burrs types and sizes prediction and cutting conditions optimization for burr controlling which based on the reasoning method of BP neural networks are realized. Operation results show that the system is reliable. It provides a new technology for burrs modelling and controlling.