The paper addresses the challenge of data mining and visualization in large sensory datasets, particularly in the context of biosensors. It introduces a personalized graph summarization (PGS) algorithm named IPGS, which aims to enhance the efficiency of summarizing large graph datasets while maintaining a similar compression ratio to state-of-the-art methods. The proposed algorithm uses weighted locality-sensitive hashing (LSH) to improve the efficiency of node identification and merging, making it suitable for handling high-dimensional data. The effectiveness of IPGS is demonstrated through experiments on eight publicly available datasets, including the Bio–Mouse–Gene dataset, which has over 43,000 nodes and 14.5 million edges. The results show that IPGS achieves better execution times and comparable compression ratios to existing PGS algorithms, while providing a more personalized and tailored summary that highlights connections relevant to specific target nodes. The paper also includes a detailed study on the Bio–Mouse–Gene dataset to illustrate the practical benefits of graph summarization in biosensor research.The paper addresses the challenge of data mining and visualization in large sensory datasets, particularly in the context of biosensors. It introduces a personalized graph summarization (PGS) algorithm named IPGS, which aims to enhance the efficiency of summarizing large graph datasets while maintaining a similar compression ratio to state-of-the-art methods. The proposed algorithm uses weighted locality-sensitive hashing (LSH) to improve the efficiency of node identification and merging, making it suitable for handling high-dimensional data. The effectiveness of IPGS is demonstrated through experiments on eight publicly available datasets, including the Bio–Mouse–Gene dataset, which has over 43,000 nodes and 14.5 million edges. The results show that IPGS achieves better execution times and comparable compression ratios to existing PGS algorithms, while providing a more personalized and tailored summary that highlights connections relevant to specific target nodes. The paper also includes a detailed study on the Bio–Mouse–Gene dataset to illustrate the practical benefits of graph summarization in biosensor research.