Research On Characteristics Analysis And Evaluation Of Criminals Based On Classification

In recent years, there exists several new criminal features and trends, which seriously affect people’s normal lives. A large amount of criminal data has been accumulated by police offices and other criminal agencies for criminal investigation. The urgent work is how to find the important information hidden behind the datasets. The application of data mining in criminal analysis has been received more attention. By using classification algorithms to predict crime types, the relationship among crime characteristics can be mined to guide the police in tracing the source of crime, which can help the police to predict the criminal inclinations and fight against crime in time.Due to various reasons in the process of data collection, there are often a large number of missing values in actual criminal dataset, which seriously affects the classification accuracy. Prior to building the classification model, an effective data filling algorithm is necessary to fill the missing values of the initial dataset. Therefore, based on Grey Relational Analysis theory of GBWKNN filling algorithm and combined with mutual k-nearest neighbor thought of MKNNI filling algorithm, a novel data filling algorithm called GMKNN is proposed in order to improve the classification accuracy. The algorithm replaces the Euclidean distance formula used in MKNNI filling algorithm with the Grey relational grade formula to eliminate the effect of noise from the nearest neighbors and effectively deal with the discrete attributes.By comparing with several popular data filling algorithms based on a real criminal dataset with lots of missing values, then classify the final data. And the results of the experiment show that higher classification accuracy can be obtained by using GMKNN algorithm, which is up to 77.837%. The classification rules of credibility in this algorithm is higher, the analyze and evaluate of criminal characteristics is more accurate.