This article is dedicated to the application of machine learning methods for phenotyping patients with HPV-associated cervical lesions using Toll-like receptor (TLR) testing data. One of the most relevant directions in the field of medical diagnostics and research is the search for effective methods of classifying pathological conditions based on biomarkers and clinical data. This study presents a new approach to phenotyping patients with HPV-associated cervical lesions using agglomerative clustering. The results of our research demonstrate the effectiveness of agglomerative clustering for phenotyping patients with HPV-associated cervical lesions. We identified several distinct clusters, each characterized by unique mRNA expression profiles of TLRs. These findings can contribute to a more accurate classification and diagnosis of patients, as well as facilitate an individualized approach to the treatment and management of HPV-associated cervical lesions. Thus, this work confirms the potential of agglomerative clustering in the field of medical research and diagnostics, especially when analyzing mRNA TLR expression data in the context of HPV-associated cervical lesions.