Abstract: The Fermi/LAT (Large Area Telescope) source catalog contains approximately 4000 blazars, of which over 40% are classified as Blazar Candidates of Uncertain type (BCUs). Determining the precise classification of these BCUs is crucial for advancing our understanding of blazar properties. In this study, we analyzed a sample of Fermi blazars by calculating their effective spectral indices between radio (R), optical (O), X-ray (X), and \gamma-ray (\gamma) bands, including \alpha_\rmRO, \alpha_\rmRX, \alpha_\rmR\gamma, \alpha_\rmOX, \alpha_\rmO\gamma, and \alpha_\rmX\gamma. From this analysis, we obtained a sample of 3387 blazars with at least two valid spectral indices, consisting of 828 flat-spectrum radio quasars (FSRQs), 1382 BL Lacertae objects (BL Lacs), and 1177 BCUs. Using the effective spectral indices as feature parameters, we performed unsupervised classification on the 1177 BCUs via the Gustafson-Kessel (GK) fuzzy clustering algorithm. The results show that: among the 1177 BCUs, 904 were classified as FSRQs and 273 as BL Lacs. For sources with inconsistent classifications between our study and previous literature, most were identified as FSRQs in our work but as BL Lacs in previous studies. From the perspective of multi-wavelength spectral indices, low-synchrotron-peaked BL Lacs (LBLs) exhibit similar radiation properties to FSRQs across multiple bands.