Clarifying the correlation between live banding, perceived value, and consumer repurchase is to enhance the economic benefits of live banding better. In this paper, a K-nearest neighbor classification algorithm is proposed in the context of big data analysis technology, and the principle and distance criterion of the algorithm are explained. Then the KNN algorithm is optimized using the Gaussian kernel density function, and the optimization process of the algorithm is given. Finally, the optimized KNN algorithm is used to mine and analyze the indicators of Taobao live banding data, and the performance evaluation is also done for the algorithm. Regarding live-streaming with goods and sensory value, the ratings of A, B, C and D accounted for 38.41%, 36.73%, 34.54% and 35.4%, respectively. In terms of the association between live banding and consumer repurchase, the average value of the data is 33.46%, the maximum value is 43.65%, and the minimum value is 18.16%. Big data analysis shows a strong correlation between live banding, sensory value and consumer repurchase. Live banding enhances consumer sensory value, sensory value influences consumer repurchase, and consumer repurchase behavior promotes live banding optimization, which continuously improves the quality of the product and marketing atmosphere of live banding.