This paper presents an efficient algorithm for personalized graph summarization (PGS), named IPGS, which aims to provide a faster and more effective way to summarize large graph datasets, including biosensors. The algorithm uses weighted locality-sensitive hashing (LSH) to identify similar nodes for merging, which enhances the efficiency of the summarization process. IPGS is designed to produce a summary graph with a similar compression ratio to the state-of-the-art PGS algorithm but with significantly faster execution times. The algorithm maintains a correction set to ensure lossless summarization, allowing for the reconstruction of the original graph if needed. The paper evaluates the performance of IPGS on eight real-world datasets, including the Bio–Mouse–Gene dataset, which contains 43.1 K nodes and 14.5 M edges. The results show that IPGS achieves a compression ratio comparable to PGS but with faster execution times. Additionally, IPGS outperforms the state-of-the-art algorithm LDME in terms of execution time while achieving a similar compression ratio. The algorithm is also effective in preserving the influence of target nodes in the summary graph, making it suitable for personalized data mining and visualization tasks. The study highlights the importance of graph summarization in the analysis of large-scale datasets, particularly in the context of biosensors and healthcare. The proposed algorithm demonstrates the effectiveness of personalized graph summarization in capturing key patterns and trends within the data, enabling more efficient data mining and visualization. The results show that IPGS is a scalable and efficient solution for summarizing large graph datasets, making it a valuable tool for researchers and practitioners in the field of data mining and graph analysis.This paper presents an efficient algorithm for personalized graph summarization (PGS), named IPGS, which aims to provide a faster and more effective way to summarize large graph datasets, including biosensors. The algorithm uses weighted locality-sensitive hashing (LSH) to identify similar nodes for merging, which enhances the efficiency of the summarization process. IPGS is designed to produce a summary graph with a similar compression ratio to the state-of-the-art PGS algorithm but with significantly faster execution times. The algorithm maintains a correction set to ensure lossless summarization, allowing for the reconstruction of the original graph if needed. The paper evaluates the performance of IPGS on eight real-world datasets, including the Bio–Mouse–Gene dataset, which contains 43.1 K nodes and 14.5 M edges. The results show that IPGS achieves a compression ratio comparable to PGS but with faster execution times. Additionally, IPGS outperforms the state-of-the-art algorithm LDME in terms of execution time while achieving a similar compression ratio. The algorithm is also effective in preserving the influence of target nodes in the summary graph, making it suitable for personalized data mining and visualization tasks. The study highlights the importance of graph summarization in the analysis of large-scale datasets, particularly in the context of biosensors and healthcare. The proposed algorithm demonstrates the effectiveness of personalized graph summarization in capturing key patterns and trends within the data, enabling more efficient data mining and visualization. The results show that IPGS is a scalable and efficient solution for summarizing large graph datasets, making it a valuable tool for researchers and practitioners in the field of data mining and graph analysis.