Furniture Style Clustering

Five experts were invited to collect vocabulary for styling from the internet, and a total of 108 styling terms were obtained. Combined with the images of furniture styles selected and processed in Section 4.1, the experts were invited again to score the 108 styles. For common styling terms, 1 point was added; for less common terms, the score was 0. The accumulated scores for the 108 different styling terms reflect the popularity or market performance of each style. By analyzing the score data, the K-means clustering method was used to effectively classify different styles. K-means is a clustering method based on distance metrics, aiming to minimize the squared distance from each sample point to its cluster center. The Elbow Method was used to analyze the relationship between the number of clusters and the total within-cluster sum of squares (Inertia). The clustering result is shown in the figure below, where it is evident that when the value of K is 5, it provides the optimal classification, i.e., five categories were chosen as the number of clusters.