This paper summarizes 13 indoor space design indicators that affect perception and behavior from three aspects: physical elements, other elements, and spatial attributes. Then, through the K-means algorithm and FP-growth correlation rule algorithm, we obtain the user’s behavioral habit data and carry out the simple correlation analysis and curve estimation regression analysis of the user’s perception and behavior based on the indexes in turn. On this basis, it is proposed that the indoor space stay time factor of green art users and spatial perception have 10 highly significant correlation indexes with 0.01 boundaries (two-tailed). Under the quadratic function model of the time factor R2=0.3425, the width in the spatial scale and the significance of the time factor have a P-value of 0.0065. The peak value of indoor natural light intake of the neighborhoods adopting the green design is 2,667.3% higher than that of traditional neighborhoods on cloudy days. The cost is 2667.3951 more, but the green interior design that is based on perception and behavioral association is highly effective.