Intelligent Car Recommendation System Based On Tag Expansio

In recent years,the development of computer technology has changed our life.As modern people enter a fast-paced life,the widespread use of cars has become a symbol of society.How to choose a car that meets the user’s own conditions in the massive car information and the era of big data has become an important issue.However,in the traditional car purchase recommendation model,there are problems such as lack of personalized recommendation for users,low recommendation accuracy,ignoring the influence of objective factors on car purchase decision,data sparsity and cold start in the car recommendation process.According to the current needs of people for car purchases,the relevant problems in the car recommendation model are analyzed,a car recommendation algorithm based on hybrid label expansion is designed,and an intelligent car recommendation system is built to accurately recommend cars for users.This research has broad application prospects.The main research work of the dissertation includes the following three aspects:(1)By combining the web crawler technology to obtain various mainstream car information from related car websites.For the obtained information,text data processing technology is used to clean and integrate the data,and a standard and normative car data set is obtained.(2)By studying the co-occurrence similarity and semantic similarity between label,the characteristics of car label in two dimensions are analyzed,and the method of mixing the maximum value of label similarity is adopted to design a car label extension algorithm based on mixed label similarity.According to the similar attributes to the label of cars and the same brand,the car label data is classified and expanded according to the car brand,and finally the label set expansion of all cars is completed.At the same time,aiming at objective factors such as car safety issues that are often overlooked in car purchase behavior and the influence of different geographic regions on car models,the safety weighting factor of the car brand and the geographical location weighting factor of the car model are designed to affect the car recommendation results.The weighted optimization process is carried out to ensure that the subjective needs of users for car purchase are unified with the actual objective factors,and finally a car recommendation algorithm based on mixed label expansion is obtained.By designing comparative experiments and analyzing multiple algorithm evaluation indicators,it is verified that the algorithm proposed in this paper has good feasibility and effectiveness in the field of car recommendation.(3)Combined with the car recommendation algorithm based on hybrid label extension,an intelligent car recommendation system is designed and implemented.The technologies used in the system include B/S framework,Django framework,and SQL database technology.System functions include user login,registration,car information query and search,car recommendation,user ratings and comments on cars,and background management.Through the demand analysis of the system,each module of the system is compactly designed,the coupling degree of the modules is reduced,the processing flow is optimized and the designed car recommendation algorithm based on the hybrid label extension is applied in the system.The above research shows that the car recommendation algorithm based on hybrid label expansion proposed in this paper has a good effect when applied to the car recommendation system.Compared with the traditional recommendation algorithm,this algorithm expands the original car label,effectively reduces the data sparsity and cold start problems in the recommendation system,and the algorithm guarantees the user’s subjective quality in the car recommendation process.Under the premise of car purchase intention,the final recommendation result is optimized by combining the objective factors in the car purchase behavior,making the car recommendation result more accurate and effective,and meeting the user’s car purchase needs.