Monitoring serum glycomics is one of the most important emerging approaches for diagnosis of various cancers, and the majority of previous studies were based on MALDI-MS or HPLC analysis. Considering the difference of these analytical methods employed for serum glycomics, it is necessary to compare the effectiveness of different analytical methods for monitoring the aberrant changes in serum glycomics. In this study, a strategy based on machine learning was firstly applied for comparing the analysis results of MALDI-MS and HPLC on the same serum glycomics of hepatocellular carcinoma (HCC) samples. The capability of these two analytical methods for identifying HCC is demonstrated by the classification results obtained from MALDI-MS and HPLC data. In addition, by comparing glycomics which were significantly correlated with HCC based on MALDI-MS and HPLC, some N-glycans which may be the potential biomarkers for HCC were identified, validating the capability of these two analytical methods for the differentiated identification in the analysis of glycomics. Meanwhile, it is noteworthy that various physiological and environmental factors may cause the aberrant changes in glycosylation, and all these interference factors may be minimized by analyzing the same sample sets of HCC. Overall, these results showed that MALDI-MS and HPLC are complementary in qualitative and quantitative analysis of serum glycomics.