The excessive amount of digital information has made it crucial to extract the relevant information. This hinders researchers in finding documents pertaining to their research. There exist various state-of-the-art techniques, such as co-citation, bibliographic coupling, and their recent extensions like citation proximity analysis and citation order analysis, that recommend the relevant documents against the posed query. Most of these approaches are statistical in nature and thus can further be extended by incorporating some semantics to enhance the results. In this paper, we present an extension of a co-citation-based technique to identify the most relevant documents to co-cited document(s).The proposed system explores in-text citation frequencies and in-text citation patterns of co-cited documents within the different logical sections of cited-by papers. Furthermore, we have evaluated the proposed approach with the co-citation approach and citation proximity analysis (CPA) approach on a dataset acquired from CiteSeer. The outcomes revealed that most of the time the proposed approach outperformed other state-of-the-art techniques. The average correlation of the proposed approach is increased by 68% as compared to the co-citation-based approach. In comparison with CPA approach, the average correlation of the proposed approach is increased by 39% with respect to gold-standard rankings.