Multi-dimensional Vector Recommendation Model Based On Call List’s Data

In the current case investigation,call list’s analysis plays an increasingly important role.The call list’s data contains a large amount of hidden information,and by mining these hidden information,it can provide effective clues for case investigation.Due to the importance of call list’s data in case investigation,methods for analyzing call list’s data are also endless.At present,most of the research on call list’s analysis stays at the stage of social network analysis.Although the analysis of call list’s data based on social network can achieve certain analysis results,with the arrival of the era of big data,the effect of analysis is more and more unsatisfactory.In the past few years,machine learning has made a big splash in data analysis,providing a new solution to many classic problems.Based on the above analysis,the multi-dimensional vector recommendation model based on call list’s data proposed in this thesis comprehensively considers various methods based on social network analysis,and measures the importance of personnel in the call list’s from multiple factors in the call list’s data.First of all,the dialogue data of this thesis is processed,and each object in the call list’s data is represented by a vector by characterizing the importance index of the affected object.Secondly,taking the above vector as input,fully consider the degree of influence of each factor on the object,and construct a mathematical model to describe the influence of the object.Finally,the object’s influence model is validated with real call list’s data,and the model is continuously modified by verification until the model achieves the best effect.In this thesis,the experimental data is verified by the data set of a certain activity case in a certain place,and the object influence model is verified by the call list’s data of known key people.The experiment uses a blank control to test,and compares the final model recommendation results with the actual case results to see the accuracy of the recommendation results.The multi-dimensional vector recommendation model proposed in this thesis comprehensively considers the influence degree of many factors on the importance of personnel,and the accuracy of recommendation results is very high.Compared with the shortcomings of traditional call list’s analysis method,such as low efficiency and high cost,the multidimensional vector recommendation model based on call list’s data proposed in this thesis improves the efficiency of traditional recommendation methods and improves work efficiency and cost.