Search

Community detection becomes an important problem with the booming of social networks. As an excellent clustering algorithm, Mean-Shift can not be applied directly to community detection, since Mean-Shift can only handle data with coordinates, while the data in the community detection problem is mostly represented by a graph that can be treated as data with a distance matrix (or similarity matrix). Fortunately, a new clustering algorithm called Medoid-Shift is proposed. The Medoid-Shift algorithm preserves the benefits of Mean-Shift and can be applied to problems based on distance matrix, such as community detection. One drawback of the Medoid-Shift algorithm is that there may be no data points within the neighborhood region defined by a distance parameter. To deal with the community detection problem better, a new algorithm called Revised Medoid-Shift (RMS) in this work is thus proposed. During the process of finding the next medoid, the RMS algorithm is based on a neighborhood defined by KNN, while the original Medoid-Shift is based on a neighborhood defined by a distance parameter. Since the neighborhood defined by KNN is more stable than the one defined by the distance parameter in terms of the number of data points within the neighborhood, the RMS algorithm may converge more smoothly. In the RMS method, each of the data points is shifted towards a medoid within the neighborhood defined by KNN. After the iterative process of shifting, each of the data point converges into a cluster center, and the data points converging into the same center are grouped into the same cluster.

Additional file 1: Table S1. Effects of two levels (A and B) of chitinase-containing culture supernatants (CCS) of strain TC -1 on death rates of Spodoptera exigua larvae. Table S2. A broad-spectrum screening assay of primary chitinase with the highest chitinolytic activity in the culture supernatant of the strain TC-1 by isoelectric precipitation. Table S3. A narrow-spectrum screening assay of primary chitinase with the highest chitinolytic activity in the culture supernatant of the strain TC-1 by isoelectric precipitation. Table S4. Toxicity of the pH 6.7-precipitated crude chitinase from the culture supernatant of strain TC-1 against the second- and third-instar larvae of Spodoptera exigua1. Table S5. Factors and levels of response surface design for optimization of chitinase production conditions of strain TC-1. Table S6. Credibility analysis of the regression model. Figure S1. The standard curve of N-acetylglucosamine.