Research On Code Similarity Detection Technology Based On Local Sensitivity Hash

With the development of the society,the software system is widely used in all aspects of daily life.All kinds of software systems are likely to face the need of system reconstruction and intellectual property protection.However,in recent years,the efficiency of detection technology are not improved significantly.In this paper,we apply the local sensitive hash to the nearest neighbor search in the key information matrix of the Gao Weiyuan code,so that we can get the similar result.The source code will be pre processed,the source code will be converted into q-gram identification sequence.Through the similar Jplag code block detection technology based on the classification and use of local sensitive hash algorithms classification took place in different barrel set q-gram signature sequence code,in the barrel of the data will be re organized to form the prefix tree data structure.In order to locate similar code in the collection of different code,need to hash the same query code,similar to a.block of code prefix tree probability is greater than the given threshold,location search algorithm uses the prefix tree of similar code to accurately.In the process of using local sensitive hash to classify the similar code blocks into the same bucket,it can effectively reduce the time cost of invalid search and improve the processing efficiency.Based on the research of a large of the relevant literature,the code similarity detection method based on locality-sensitive hash is fully studied and researched.The experimental results show that the method proposed in this paper has better detection efficiency than the original method.The experimental results show that the proposed method improves the processing efficiency by about 10%compared with the original method.