由于Relief算法比较简单，运行效率高，并且结果也比较令人满意，因此得到广泛应用，但是其局限性在于只能处理两类别数据，因此1994年Kononeill对其进行了扩展，得到了ReliefF作算法，可以处理多类别问题。该算法用于处理目标属性为连续值的回归问题。-Because Relief algorithm is relatively simple, high efficiency, and the results are more satisfactory, so widely used, but its limitation is that only two types of data can be processed, so Kononeill in 1994 to expand it, get ReliefF algorithm , Can handle multiple categories of problems. The algorithm is used to deal with the regression problem that the target attribute is a continuous value.