A method and system for automobile intelligent data analysis and decision-making

Abstract

The present invention proposes an automobile intelligent data analysis and decision-making method and system, which relates to the field of data analysis and decision-making technology, divides brand category data, performance category data and customer demand data, associates various data, obtains category expectation data based on the obtained associated integrated data, matches the obtained associated expected brand data and associated expected performance data with the current automobile inventory data, calculates the demand matching degree, and obtains the optimal automobile recommendation sequence; obtains customer recommendation feedback information, determines whether to update the data association, and then updates the optimal automobile recommendation sequence. The present invention realizes accurate recommendation and dynamic adjustment by deeply mining the association between historical data, real-time inventory information and customer demand. The system not only improves the accuracy and personalization of recommendations, but also enhances customer experience and market competitiveness.

Description

Technical Field

[0001] The present invention proposes an automobile intelligent data analysis and decision-making method and system, which relate to the technical field of data analysis and decision-making, and specifically to the technical field of automobile intelligent data analysis and decision-making. Background Art

[0002] With the continuous development of the automobile market and the increasing diversification of consumer needs, automobile recommendation systems have become an important tool in the field of automobile sales and services. However, traditional automobile recommendation systems often rely on static database information and simple matching algorithms, which makes it difficult to accurately capture and respond to customers' personalized needs and dynamically changing market environments. In the prior art, automobile recommendation systems usually build customer profiles by collecting and analyzing information such as customers' car purchase history, browsing behavior, and search records, and recommend automobile brands and models that customers may be interested in based on these profile information. However, these systems often ignore the real-time and diversity of automobile inventory data, as well as the important impact of customer feedback on recommendation results. In addition, traditional systems also have certain limitations in data processing and association analysis, making it difficult to fully explore and utilize the potential value in the data. Summary of the invention

[0003] The present invention provides an intelligent automobile data analysis and decision-making method and system to solve the problem that the existing technology ignores the real-time and diversity of automobile inventory data, as well as the important impact of customer feedback on recommendation results. In addition, the traditional system also has certain limitations in data processing and association analysis, and it is difficult to fully explore and utilize the potential value in the data.

[0004] The present invention proposes an automobile intelligent data analysis and decision-making method, the method comprising:

[0005] S1. Obtain historical automobile inventory data, divide it into brand category data and performance category data, preprocess it, obtain the preprocessed brand category data, performance category data and customer demand data, associate various data, and obtain associated integrated data;

[0006] S2, acquiring brand category expected data and performance category expected data of the target vehicle according to the associated integrated data, and acquiring associated expected brand data and associated expected performance data according to the brand category expected data and performance category expected data;

[0007] S3, matching the associated expected brand data and the associated expected performance data with the current car inventory data respectively, calculating the demand matching degree of each brand matching car and performance matching car, and obtaining the optimal car recommendation sequence according to the demand matching degree;

[0008] S4. Obtain customer recommendation feedback information, determine whether to update data association according to the customer recommendation feedback data, obtain a determination result, update the data association according to the determination result, and then update the optimal car recommendation sequence.

[0009] Further, the S1 includes:

[0010] Acquire historical automobile inventory data, classify the historical automobile inventory data according to automobile brand types, and obtain multiple brand type data; classify the historical automobile inventory data according to automobile performance types, and obtain multiple performance type data;

[0011] Number each brand type data to obtain the required number data for each car;

[0012] Number each performance category data to obtain performance number data of each car;

[0013] Obtain historical customer demand data, number each customer demand data, and obtain the customer number data of each customer;

[0014] Preprocessing the brand category data, performance category data and customer number data to obtain preprocessed brand category data, performance category data and customer demand data;

[0015] Obtaining automobile brand category data and performance category data determined by each customer demand data, and associating each of the customer demand data with the automobile brand category data and performance category data corresponding thereto to obtain a plurality of associated data;

[0016] The plurality of associated data are combined to obtain associated integrated data.

[0017] Further, the S2 includes:

[0018] Obtain the page views in the associated integrated data to perform expected sorting on the brand categories, and obtain the expected data sorting of the brand categories;

[0019] Obtain the page views in the associated integrated data to perform expected sorting on the performance categories, and obtain the expected data sorting of the performance categories;

[0020] Calculating the sum of the rankings of each associated data in the associated integrated data, and sorting the associated data according to the sum of the rankings to obtain an associated expected ranking;

[0021] The brand category expectation data and performance category requirement data in the associated expectation sorting are the associated expected brand data and the associated expected performance data.

[0022] Further, the S3 includes:

[0023] Match the associated desired brand data with the current car inventory data to obtain multiple brand matching cars for the customer;

[0024] Matching the associated expected performance data with the current car inventory data to obtain multiple performance-matched cars for the customer;

[0025] Calculate the demand matching degree of each brand matching car according to the associated expected brand data combined with multiple brand matching car data;

[0026] Calculate the demand matching degree of each performance matching vehicle according to the associated expected performance data combined with the data of multiple performance matching vehicles;

[0027] Sort the demand matching degree of brand-matching cars to obtain the demand brand sequence;

[0028] Sort the demand matching degree of performance-matching cars to obtain the demand performance sequence;

[0029] Obtain the car corresponding to the associated data with the smallest sum of the serial numbers of the required brand and required performance in the associated expectation sorting as the best recommended car;

[0030] The cars corresponding to the associated data with the second smallest to the Nth smallest serial numbers in the sum of the required brand sequence and the required performance sequence are obtained in sequence, and then the optimal car recommendation sequence is obtained, and cars are recommended to customers according to the optimal car recommendation sequence to obtain car recommendation data.

[0031] Further, the S4 includes:

[0032] Obtain customer recommendation feedback information, classify the recommendation feedback information according to preset abnormality categories, and obtain feedback abnormality classification information of multiple preset abnormality categories;

[0033] Compare the feedback abnormal classification information of each preset abnormal category with the preset abnormal threshold of the corresponding category, and compare the category abnormality results;

[0034] According to the type anomaly comparison result, the associated data of the target car data corresponding to the abnormal type is updated, and then the optimal car recommendation sequence is updated.

[0035] Furthermore, the system comprises:

[0036] The demand association module is used to obtain historical automobile inventory data, divide it into brand category data and performance category data, pre-process it, obtain the pre-processed brand category data, performance category data and customer demand data, associate various data, and obtain associated integrated data;

[0037] an expectation acquisition module, configured to acquire brand category expectation data and performance category expectation data of a target vehicle according to the associated integrated data, and acquire associated expected brand data and associated expected performance data according to the brand category expectation data and performance category expectation data;

[0038] An expectation matching module, used to match the associated expected brand data and the associated expected performance data with the current car inventory data respectively, calculate the demand matching degree of each brand matching car and performance matching car, and obtain the optimal car recommendation sequence according to the demand matching degree;

[0039] The association update module is used to obtain customer recommendation feedback information, determine whether to update the data association according to the customer recommendation feedback data, obtain a judgment result, update the data association according to the judgment result, and then update the optimal car recommendation sequence.

[0040] Furthermore, the demand association module includes:

[0041] An information acquisition module is used to acquire historical automobile inventory data, classify the historical automobile inventory data according to automobile brand types, and obtain multiple brand type data, and classify the historical automobile inventory data according to automobile performance types, and obtain multiple performance type data;

[0042] The numbering module is used to number each brand type data and obtain the required number data for each car;

[0043] Number each performance category data to obtain performance number data of each car;

[0044] Obtain historical customer demand data, number each customer demand data, and obtain the customer number data of each customer;

[0045] A preprocessing module, used for preprocessing the brand category data, performance category data and customer number data to obtain preprocessed brand category data, performance category data and customer demand data;

[0046] The association integration module is used to obtain the brand category data and performance category data of the automobile determined by each customer demand data, and associate each of the customer demand data with the corresponding brand category data and performance category data of the automobile to obtain a plurality of associated data;

[0047] The plurality of associated data are combined to obtain associated integrated data.

[0048] Furthermore, the expectation acquisition module includes:

[0049] An expected ranking module is used to obtain the page views in the associated integrated data to perform expected ranking on the brand categories, thereby obtaining the expected data ranking of the brand categories;

[0050] Obtain the page views in the associated integrated data to perform expected sorting on the performance categories, and obtain the expected data sorting of the performance categories;

[0051] Calculating the sum of the rankings of each associated data in the associated integrated data, and sorting the associated data according to the sum of the rankings to obtain an associated expected ranking;

[0052] The expected data acquisition module is used for the brand category expected data and performance category required data in the associated expected sorting, namely the associated expected brand data and the associated expected performance data.

[0053] Furthermore, the expected matching module includes:

[0054] A matching module, used to match the associated desired brand data with the current car inventory data to obtain multiple brand matching cars of the customer;

[0055] Matching the associated expected performance data with the current car inventory data to obtain multiple performance-matched cars for the customer;

[0056] A matching degree calculation module, used to calculate the demand matching degree of each brand matching car according to the associated expected brand data combined with multiple brand matching car data;

[0057] Calculate the demand matching degree of each performance matching vehicle according to the associated expected performance data combined with the data of multiple performance matching vehicles;

[0058] A matching sorting module is used to sort the demand matching degree of brand-matching cars and obtain the demand brand sequence;

[0059] Sort the demand matching degree of performance-matching cars to obtain the demand performance sequence;

[0060] The car recommendation module is used to obtain the car corresponding to the associated data with the smallest sum of the serial numbers of the required brand and required performance in the associated expected sorting, as the optimal recommended car;

[0061] The cars corresponding to the associated data with the second smallest to the Nth smallest serial numbers in the sum of the required brand sequence and the required performance sequence are obtained in sequence, and then the optimal car recommendation sequence is obtained, and cars are recommended to customers according to the optimal car recommendation sequence to obtain car recommendation data.

[0062] Furthermore, the association update module includes:

[0063] The abnormal feedback module is used to obtain customer recommendation feedback information, classify the recommendation feedback information according to preset abnormality categories, and obtain feedback abnormality classification information of multiple preset abnormality categories;

[0064] An anomaly comparison module, used to compare the feedback anomaly classification information of each preset anomaly category with the preset anomaly threshold of the corresponding category, and compare the category anomaly results;

[0065] The association updating module is used to update the association data of the target vehicle data corresponding to the abnormal category according to the category abnormality comparison result, and then update the optimal vehicle recommendation sequence.

[0066] Beneficial effects of the present invention: Through sophisticated data classification, preprocessing and association analysis, customer needs can be more accurately grasped, thereby improving the accuracy and pertinence of recommendations. Personalized recommendations based on customer needs can enhance customers' car-buying experience and increase customer satisfaction with brands and services. By analyzing customer needs and inventory data, market trends can be predicted, inventory structure can be optimized, and inventory backlogs and waste can be reduced. Through an intelligent recommendation system, it is possible to quickly respond to market changes and provide automotive products that better meet market demand, thereby enhancing the company's market competitiveness. The recommendation system based on big data can provide companies with rich data support and help them make more scientific and reasonable decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 A schematic diagram of an automobile intelligent data analysis and decision-making method;

[0068] Figure 2 The figure is a schematic diagram of an automobile intelligent data analysis and decision-making system. DETAILED DESCRIPTION

[0069] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0070] In one embodiment of the present invention, a method for analyzing and making decisions about automobile intelligent data is proposed by the present invention, and the method comprises:

[0071] S1. Obtain historical automobile inventory data, divide it into brand category data and performance category data, preprocess it, obtain the preprocessed brand category data, performance category data and customer demand data, associate various data, and obtain associated integrated data;

[0072] S2, acquiring brand category expected data and performance category expected data of the target vehicle according to the associated integrated data, and acquiring associated expected brand data and associated expected performance data according to the brand category expected data and performance category expected data;

[0073] S3, matching the associated expected brand data and the associated expected performance data with the current car inventory data respectively, calculating the demand matching degree of each brand matching car and performance matching car, and obtaining the optimal car recommendation sequence according to the demand matching degree;

[0074] S4. Obtain customer recommendation feedback information, determine whether to update data association according to the customer recommendation feedback data, obtain a determination result, update the data association according to the determination result, and then update the optimal car recommendation sequence.

[0075] The working principle of the above technical solution is as follows: collect historical automobile inventory data from a database or external data source, which includes automobile brands, models, performance parameters and inventory status. Classify the collected data according to brand categories and performance categories for subsequent analysis. Clean the data, remove noise and duplicates, handle missing values, and possibly perform normalization or standardization to ensure data consistency and comparability; obtain customer automobile purchase demand data through questionnaires, online behavior analysis or sales records. Associate the pre-processed brand category data, performance category data and customer demand data to find out the type of car that the customer is most likely to be interested in. Obtain the brand and performance data of each vehicle that the customer has browsed historically, and associate the brand with the performance; integrate the associated data into a comprehensive associated integrated data to provide a basis for subsequent recommendation analysis. Extract the customer's expected data on brand categories and performance categories from the associated integrated data. Match the extracted expected data with the current automobile inventory data, and calculate the matching degree of each car with the expected data, including brand matching degree and performance matching degree. Sort all candidate cars according to the calculated demand matching degree. Select the car with the highest matching degree as the head of the recommendation list to form the optimal car recommendation sequence. Obtain customer feedback data on the recommended cars, including whether to accept the recommendation, satisfaction, etc. Based on customer feedback, evaluate the accuracy and effectiveness of the current data association. If improvement is found, update the data association information to improve the accuracy of the recommendation. Recalculate the recommendation sequence using the updated data association to provide recommendations that better meet customer needs. Through query languages ​​such as SQL, connections can be established between multiple database tables to retrieve and integrate relevant information based on common fields (such as customer ID, model ID, etc.) or data warehouse technology. Through the ETL process, data from different data sources is extracted, transformed and loaded into a centralized storage environment. In this process, data association is usually performed as part of data conversion, and association is achieved by defining mapping relationships between data.

[0076] The technical effects of the above technical solutions are: through fine data classification, preprocessing and correlation analysis, customer needs can be grasped more accurately, thereby improving the accuracy and pertinence of recommendations. Personalized recommendations based on customer needs can enhance customers' car-buying experience and increase customer satisfaction with brands and services. By analyzing customer needs and inventory data, market trends can be predicted, inventory structure can be optimized, and inventory backlogs and waste can be reduced. Through an intelligent recommendation system, it is possible to respond quickly to market changes and provide automotive products that are more in line with market demand, thereby enhancing the company's market competitiveness. The recommendation system based on big data can provide companies with rich data support and help them make more scientific and reasonable decisions.

[0077] In one embodiment of the present invention, the S1 includes:

[0078] Acquire historical automobile inventory data, classify the historical automobile inventory data according to automobile brand types, and obtain multiple brand type data; classify the historical automobile inventory data according to automobile performance types, and obtain multiple performance type data;

[0079] The performance category data includes engine performance, driving performance, and handling performance;

[0080] Number each brand type data to obtain the required number data for each car;

[0081] Number each performance category data to obtain performance number data of each car;

[0082] Obtain historical customer demand data, number each customer demand data, and obtain the customer number data of each customer;

[0083] Preprocessing the brand category data, performance category data and customer number data to obtain preprocessed brand category data, performance category data and customer demand data;

[0084] Obtaining automobile brand category data and performance category data determined by each customer demand data, and associating each of the customer demand data with the automobile brand category data and performance category data corresponding thereto to obtain a plurality of associated data;

[0085] The associated data includes brand category data and performance category data of historically browsed vehicles of each customer demand data;

[0086] The plurality of associated data are combined to obtain associated integrated data.

[0087] The working principle of the above technical solution is as follows: extract the brand information of the car from the historical car inventory data, and classify the data according to different brands to form multiple brand category data sets. Extract the performance information of the car from the historical car inventory data, including engine performance, driving performance and handling performance, and classify the data according to these performance characteristics to form multiple performance category data sets. Assign a unique number (demand number) to each brand category data for subsequent data processing and association. Assign a unique number (performance number) to each performance category data for data tracking and matching. Extract customer information from historical customer demand data, and assign a unique customer number to each customer.

[0088] The brand category data, performance category data and customer demand data are cleaned, denoised, and missing values ​​are processed to ensure data consistency and accuracy. Based on historical customer demand data, the brand category and performance category that each customer pays attention to when browsing or purchasing a car are analyzed, and the customer demand data is associated with the corresponding brand category data and performance category data. This association is usually achieved through customer numbers, demand numbers and performance numbers. For each customer demand data, an associated data record containing the car brand categories and performance categories that the customer has browsed or paid attention to in the past is generated. All associated data records are combined to form associated integrated data. This integrated data fully reflects the needs and preferences of different customers for different brands and performance cars, providing an important basis for subsequent car recommendations and market analysis.

[0089] The technical effects of the above technical solution are: through data classification and numbering, historical automobile inventory data and customer demand data are made more orderly and easy to manage, and the efficiency of data processing is improved. Based on the customer's historical browsing and purchasing behavior, it is possible to more accurately grasp the customer's brand preferences and performance requirements, thereby providing more personalized car recommendations. The associated integrated data provides market analysts with a rich data source, helping them to understand market demand and trends more accurately, and providing strong support for enterprises to formulate market strategies and product planning. By meeting the personalized needs of customers, improving customer satisfaction and loyalty, and thus optimizing the overall customer experience. Based on the analysis of associated integrated data, enterprises can make more scientific and reasonable decisions, such as adjusting inventory structure, optimizing product lines, etc., to improve market competitiveness and profitability.

[0090] In one embodiment of the present invention, S2 includes:

[0091] Obtain the page views in the associated integrated data and perform the desired sorting (from large to small) on the brand categories to obtain the desired data sorting of the brand categories;

[0092] Obtain the page views in the associated integrated data and perform expected sorting (from large to small) on the performance categories to obtain the expected data sorting of the performance categories;

[0093] Calculating the sum of the rankings of each associated data in the associated integrated data, and sorting the associated data according to the sum of the rankings to obtain an associated expected ranking;

[0094] The brand category expectation data and performance category requirement data in the associated expectation sorting are the associated expected brand data and the associated expected performance data.

[0095] The working principle of the above technical solution is as follows: the pageview data corresponding to each brand category is extracted from the associated integrated data. According to the size of the pageviews, the brand categories are sorted in descending order, that is, the brand category with the most pageviews is ranked first, and the brand category with the least pageviews is ranked last. This sorting result is the brand category expected data sorting, which reflects the customer's overall interest and preference for different brand categories. The pageview data corresponding to each performance category is extracted from the associated integrated data. According to the size of the pageviews, the performance categories are sorted in descending order to generate the performance category expected data sorting. This sorting result reflects the customer's preference and expectation for different performance characteristics. For each associated data in the associated integrated data (that is, the association between each customer demand and the corresponding brand category and performance category), first determine the position (that is, ranking) of the corresponding brand category in the brand category expected data sorting, and the position of the performance category in the performance category expected data sorting. Add these two rankings to obtain the sum of the sorting of the associated data. According to the size of the sorting sum, all associated data are sorted in ascending order (because the smaller the sorting sum, the more the associated data meets the customer's overall expectations). This sorting result is the association expectation sorting, which comprehensively considers the expected sorting of brand categories and performance categories, and reflects the priority of each associated data in the overall expectation. According to the association expectation sorting, the brand category and performance category corresponding to each associated data can be directly extracted from the sorting result as the associated expected brand data and associated expected performance data.

[0096] The technical effect of the above technical solution is: by comprehensively considering the ranking of page views of brand categories and performance categories, as well as the sum of the ranking of related data, a comprehensive and objective evaluation standard is provided, which helps to more accurately identify the overall needs and expectations of customers. Based on the ranking of related expectations, companies can formulate automobile recommendation strategies more accurately, giving priority to recommending brands and performance combinations that meet the overall expectations of customers, thereby improving the success rate of recommendations and customer satisfaction. By deeply understanding customers' brand preferences and performance requirements, companies can better adjust their product structure and optimize market layout, thereby improving their market competitiveness and market share. The results of the related expectation ranking provide strong data support for the company's decision-making, enabling companies to make more scientific and reasonable decisions based on customer behavior data and expectation data. By providing recommendations and services that are more in line with customer expectations, companies can significantly enhance customers' shopping experience and satisfaction, thereby improving customer loyalty and brand reputation.

[0097] In one embodiment of the present invention, S3 includes:

[0098] Match the associated desired brand data with the current car inventory data to obtain multiple brand matching cars for the customer;

[0099] Matching the associated expected performance data with the current car inventory data to obtain multiple performance-matched cars for the customer;

[0100] Calculate the demand matching degree of each brand matching car according to the associated expected brand data combined with multiple brand matching car data;

[0101] Calculate the demand matching degree of each performance matching vehicle according to the associated expected performance data combined with the data of multiple performance matching vehicles;

[0102] Degree of matching of the requirements:

[0103] The calculation formula for the demand matching degree of the brand matching car is:

[0104]

[0105] Among them, P A is the demand matching degree of brand matching automobile, j is the total number of brand types associated with the expected brand data, (P g +P t ) i is the matching coefficient between the associated expected brand data of the i-th brand category and the brand category data of the multiple brand matching cars, P g is the associated expected brand data of brand category, P t For multiple brands matching automobile brand category data, when the associated expected brand data of the i-th brand category has the same brand category data in multiple brand matching automobile data, let (Pg +P t ) i is 2. When the associated expected brand data of the i-th brand category does not have the same brand category data in multiple brand matching car data, let (P g +P t ) i =1, P k The total number of brands in the current automobile inventory data;

[0106] The calculation formula of the performance matching degree of the automobile is:

[0107]

[0108] Among them, P B is the performance matching degree of the car, M is the total number of performance types associated with the expected performance data, (X g +X t ) N is the matching coefficient between the associated expected performance data of the Nth performance category and the performance category data of the plurality of performance matching vehicles, X g is the expected performance data associated with the performance category, X t is the performance category data of multiple performance matching vehicles. When the associated expected performance data of the Nth performance category has the same performance category data in multiple performance matching vehicle data, let (X g +X t ) N =2, when the associated expected performance data of the Nth performance category does not have the same performance category data in multiple performance matching vehicle data, let (X g +X t ) N =1, X k It is the total number of performance types in the current automobile inventory data.

[0109] Sort the demand matching degree of brand-matching cars to obtain the demand brand sequence;

[0110] Sort the demand matching degree of performance-matching cars to obtain the demand performance sequence;

[0111] Obtain the car corresponding to the associated data with the smallest sum of the serial numbers of the required brand and required performance in the associated expectation sorting as the best recommended car;

[0112] The cars corresponding to the associated data with the second smallest to the Nth smallest serial numbers in the sum of the required brand sequence and the required performance sequence are obtained in sequence, and then the optimal car recommendation sequence is obtained, and cars are recommended to customers according to the optimal car recommendation sequence to obtain car recommendation data.

[0113] The working principle of the above technical solution is as follows: match the associated expected brand data with the current car inventory data, find all cars that meet the customer's brand expectations, and form a brand matching car set. Match the associated expected performance data with the current car inventory data, find all cars that meet the customer's performance expectations, and form a performance matching car set. According to the given formula, calculate the demand matching degree of each brand matching car. The formula takes into account the expected strength of each brand category in the associated expected brand data (reflected by the matching coefficient) and the richness of the brand category in the inventory (adjusted by the total number of brand categories). Calculate the demand matching degree of each performance matching car, taking into account the expected strength of the performance category and the richness of the performance category in the inventory. Sort the demand matching degrees of brand matching cars in descending order to generate a demand brand sequence, which represents the ranking of cars with high to low brand matching degrees. Sort the demand matching degrees of performance matching cars in descending order to generate a demand performance sequence. In the associated expectation sorting, find the car corresponding to the associated data with the smallest sum of serial numbers in the demand brand sequence and the demand performance sequence as the best recommended car. This ensures that the recommended car is as close as possible to the customer's comprehensive expectations in terms of brand and performance. Select the cars corresponding to the associated data with the second smallest to the Nth smallest serial number in the required brand sequence and the required performance sequence in turn to form the optimal car recommendation sequence. The cars in this sequence are arranged in descending order of the degree of matching with the customer's needs. According to the optimal car recommendation sequence, recommend cars to customers and generate car recommendation data. This data may include detailed information about the car, reasons for recommendation, price, etc., so that customers can make purchase decisions.

[0114] The technical effect of the above technical solution is: by comprehensively considering the expectations and matching of brand and performance, it is possible to more accurately identify cars that meet customer needs and improve the accuracy and pertinence of recommendations. Personalized recommendations based on customer needs can significantly improve customers' car-buying experience and increase customer satisfaction with brands and services. By analyzing customer needs and inventory data, companies can more accurately predict market trends, optimize inventory structure, and reduce inventory backlogs and waste. By providing recommendations and services that better meet customer expectations, companies can enhance their market competitiveness, attract more potential customers and increase market share. The entire recommendation process is based on a large amount of data and complex calculations, which provides companies with rich data support and decision-making basis, and helps companies make more scientific and reasonable market strategies and product plans.

[0115] In one embodiment of the present invention, the S4 includes:

[0116] Obtain customer recommendation feedback information, classify the recommendation feedback information according to preset abnormality categories, and obtain feedback abnormality classification information of multiple preset abnormality categories;

[0117] Compare the feedback abnormal classification information of each preset abnormal category with the preset abnormal threshold of the corresponding category, and compare the category abnormality results;

[0118] According to the type anomaly comparison result, the associated data of the target car data corresponding to the abnormal type is updated, and then the optimal car recommendation sequence is updated.

[0119] When the feedback abnormality classification information of any preset abnormality type is greater than the preset abnormality threshold of the corresponding type, the associated data is updated by the feedback abnormality classification information of any preset abnormality type corresponding to the customer recommendation feedback information;

[0120] When the feedback abnormality classification information of any preset abnormality type is less than or equal to the preset abnormality threshold of the corresponding type, no update is performed.

[0121] The working principle of the above technical solution is: obtain customer feedback information on car recommendations, which may include satisfaction scores, specific comments, improvement suggestions, etc. Classify the feedback information according to preset abnormality categories (such as dissatisfied brand, dissatisfied performance, overpriced, etc.) to identify different types of abnormalities or problems. This step will generate feedback abnormality classification information of multiple preset abnormality categories. Compare the feedback abnormality classification information of each preset abnormality category with the preset abnormality threshold of the corresponding category. The abnormality threshold is set based on historical data, business rules or expert opinions, and is used to determine whether the degree of abnormality in the feedback reaches the standard that requires action. Based on the comparison result, a category abnormality comparison result is generated. If the feedback abnormality classification information of a preset abnormality category is greater than its corresponding preset abnormality threshold, it is considered that there is a significant abnormality or problem in the category; otherwise, it is considered that the category is within an acceptable range. When the feedback abnormality classification information of any preset abnormality category is greater than the preset abnormality threshold of the corresponding category, the system will update the associated data according to the abnormal information. The update operation may include adjusting the brand preference weight, performance expectation weight, price sensitivity, etc. to reflect the latest feedback from customers. The updated associated data will re-enter the calculation and generation process of the optimal car recommendation sequence. The system will re-evaluate and sort the car recommendation list based on the updated data and algorithm logic to more accurately reflect the customer's current needs and preferences. If the feedback abnormal classification information of a preset abnormal category is less than or equal to the preset abnormal threshold of the corresponding category, the system considers that the category does not need to be updated and keeps the current recommendation sequence unchanged.

[0122] The technical effects of the above technical solution are: by collecting and analyzing customer feedback information and dynamically adjusting the recommendation system according to the feedback results, the system can adapt to changes in customer needs more quickly and improve the accuracy and satisfaction of recommendations. By promptly responding to and processing customer feedback, the problems and dissatisfaction encountered by customers in the car purchase process are solved, and the customer's car purchase experience and brand loyalty are enhanced. The classification and exception handling of feedback information provide important data support for the optimization of the recommendation algorithm. By continuously learning and improving the algorithm, the intelligence level and recommendation effect of the recommendation system can be improved. Through accurate recommendations and timely feedback processing, the customer's satisfaction and acceptance of the recommendation results can be improved, thereby improving the overall customer satisfaction and brand image. Encouraging customers to provide feedback not only helps to improve the recommendation system, but also enhances the customer's sense of participation and belonging, and promotes a good interactive relationship between customers and enterprises.

[0123] In one embodiment of the present invention, the system comprises:

[0124] The demand association module is used to obtain historical automobile inventory data, divide it into brand category data and performance category data, pre-process it, obtain the pre-processed brand category data, performance category data and customer demand data, associate various data, and obtain associated integrated data;

[0125] an expectation acquisition module, configured to acquire brand category expectation data and performance category expectation data of a target vehicle according to the associated integrated data, and acquire associated expected brand data and associated expected performance data according to the brand category expectation data and performance category expectation data;

[0126] An expectation matching module, used to match the associated expected brand data and the associated expected performance data with the current car inventory data respectively, calculate the demand matching degree of each brand matching car and performance matching car, and obtain the optimal car recommendation sequence according to the demand matching degree;

[0127] The association update module is used to obtain customer recommendation feedback information, determine whether to update the data association according to the customer recommendation feedback data, obtain a judgment result, update the data association according to the judgment result, and then update the optimal car recommendation sequence.

[0128] The working principle of the above technical solution is as follows: collect historical automobile inventory data from a database or external data source, which includes automobile brands, models, performance parameters and inventory status. Classify the collected data according to brand categories and performance categories for subsequent analysis. Clean the data, remove noise and duplicates, handle missing values, and possibly perform normalization or standardization to ensure data consistency and comparability; obtain customer automobile purchase demand data through questionnaires, online behavior analysis or sales records. Associate the pre-processed brand category data, performance category data and customer demand data to find out the type of car that the customer is most likely to be interested in. Obtain the brand and performance data of each vehicle that the customer has browsed historically, and associate the brand with the performance; integrate the associated data into a comprehensive associated integrated data to provide a basis for subsequent recommendation analysis. Extract the customer's expected data on brand categories and performance categories from the associated integrated data. Match the extracted expected data with the current automobile inventory data, and calculate the matching degree of each car with the expected data, including brand matching degree and performance matching degree. Sort all candidate cars according to the calculated demand matching degree. The car with the highest matching degree is selected as the head of the recommendation list to form the optimal car recommendation sequence. Customer feedback data on the recommended cars is obtained, including whether the recommendation is accepted, satisfaction, etc. Based on customer feedback, the accuracy and effectiveness of the current data association are evaluated. If improvement is found, the data association information is updated to improve the accuracy of the recommendation. The recommendation sequence is recalculated using the updated data association to provide recommendations that better meet customer needs.

[0129] The technical effects of the above technical solutions are: through fine data classification, preprocessing and correlation analysis, customer needs can be grasped more accurately, thereby improving the accuracy and pertinence of recommendations. Personalized recommendations based on customer needs can enhance customers' car-buying experience and increase customer satisfaction with brands and services. By analyzing customer needs and inventory data, market trends can be predicted, inventory structure can be optimized, and inventory backlogs and waste can be reduced. Through an intelligent recommendation system, it is possible to respond quickly to market changes and provide automotive products that are more in line with market demand, thereby enhancing the company's market competitiveness. The recommendation system based on big data can provide companies with rich data support and help them make more scientific and reasonable decisions.

[0130] In one embodiment of the present invention, the demand association module includes:

[0131] An information acquisition module is used to acquire historical automobile inventory data, classify the historical automobile inventory data according to automobile brand types, and obtain multiple brand type data, and classify the historical automobile inventory data according to automobile performance types, and obtain multiple performance type data;

[0132] The performance category data includes engine performance, driving performance, and handling performance;

[0133] The numbering module is used to number each brand type data and obtain the required number data for each car;

[0134] Number each performance category data to obtain performance number data of each car;

[0135] Obtain historical customer demand data, number each customer demand data, and obtain the customer number data of each customer;

[0136] A preprocessing module, used for preprocessing the brand category data, performance category data and customer number data to obtain preprocessed brand category data, performance category data and customer demand data;

[0137] The association integration module is used to obtain the brand category data and performance category data of the automobile determined by each customer demand data, and associate each of the customer demand data with the corresponding brand category data and performance category data of the automobile to obtain a plurality of associated data;

[0138] The associated data includes brand category data and performance category data of historically browsed vehicles of each customer demand data;

[0139] The plurality of associated data are combined to obtain associated integrated data.

[0140] The working principle of the above technical solution is as follows: extract the brand information of the car from the historical car inventory data, and classify the data according to different brands to form multiple brand category data sets. Extract the performance information of the car from the historical car inventory data, including engine performance, driving performance and handling performance, and classify the data according to these performance characteristics to form multiple performance category data sets. Assign a unique number (demand number) to each brand category data for subsequent data processing and association. Assign a unique number (performance number) to each performance category data for data tracking and matching. Extract customer information from historical customer demand data, and assign a unique customer number to each customer.

[0141] The brand category data, performance category data and customer demand data are cleaned, denoised, and missing values ​​are processed to ensure data consistency and accuracy. Based on historical customer demand data, the brand category and performance category that each customer pays attention to when browsing or purchasing a car are analyzed, and the customer demand data is associated with the corresponding brand category data and performance category data. This association is usually achieved through customer numbers, demand numbers and performance numbers. For each customer demand data, an associated data record containing the car brand categories and performance categories that the customer has browsed or paid attention to in the past is generated. All associated data records are combined to form associated integrated data. This integrated data fully reflects the needs and preferences of different customers for different brands and performance cars, providing an important basis for subsequent car recommendations and market analysis.

[0142] The technical effects of the above technical solution are: through data classification and numbering, historical automobile inventory data and customer demand data are made more orderly and easy to manage, and the efficiency of data processing is improved. Based on the customer's historical browsing and purchasing behavior, it is possible to more accurately grasp the customer's brand preferences and performance requirements, thereby providing more personalized car recommendations. The associated integrated data provides market analysts with a rich data source, helping them to understand market demand and trends more accurately, and providing strong support for enterprises to formulate market strategies and product planning. By meeting the personalized needs of customers, improving customer satisfaction and loyalty, and thus optimizing the overall customer experience. Based on the analysis of associated integrated data, enterprises can make more scientific and reasonable decisions, such as adjusting inventory structure, optimizing product lines, etc., to improve market competitiveness and profitability.

[0143] In one embodiment of the present invention, the expectation acquisition module includes:

[0144] The expected sorting module is used to obtain the page views in the associated integrated data to perform expected sorting (from large to small) on the brand categories, and obtain the expected data sorting of the brand categories;

[0145] Obtain the page views in the associated integrated data and perform expected sorting (from large to small) on the performance categories to obtain the expected data sorting of the performance categories;

[0146] Calculating the sum of the rankings of each associated data in the associated integrated data, and sorting the associated data according to the sum of the rankings to obtain an associated expected ranking;

[0147] The expected data acquisition module is used for the brand category expected data and performance category required data in the associated expected sorting, namely the associated expected brand data and the associated expected performance data.

[0148] The working principle of the above technical solution is as follows: the pageview data corresponding to each brand category is extracted from the associated integrated data. According to the size of the pageviews, the brand categories are sorted in descending order, that is, the brand category with the most pageviews is ranked first, and the brand category with the least pageviews is ranked last. This sorting result is the brand category expected data sorting, which reflects the customer's overall interest and preference for different brand categories. The pageview data corresponding to each performance category is extracted from the associated integrated data. According to the size of the pageviews, the performance categories are sorted in descending order to generate the performance category expected data sorting. This sorting result reflects the customer's preference and expectation for different performance characteristics. For each associated data in the associated integrated data (that is, the association between each customer demand and the corresponding brand category and performance category), first determine the position (that is, ranking) of the corresponding brand category in the brand category expected data sorting, and the position of the performance category in the performance category expected data sorting. Add these two rankings to obtain the sum of the sorting of the associated data. According to the size of the sorting sum, all associated data are sorted in ascending order (because the smaller the sorting sum, the more the associated data meets the customer's overall expectations). This sorting result is the association expectation sorting, which comprehensively considers the expected sorting of brand categories and performance categories, and reflects the priority of each associated data in the overall expectation. According to the association expectation sorting, the brand category and performance category corresponding to each associated data can be directly extracted from the sorting result as the associated expected brand data and associated expected performance data.

[0149] The technical effect of the above technical solution is: by comprehensively considering the ranking of page views of brand categories and performance categories, as well as the sum of the ranking of related data, a comprehensive and objective evaluation standard is provided, which helps to more accurately identify the overall needs and expectations of customers. Based on the ranking of related expectations, companies can formulate automobile recommendation strategies more accurately, giving priority to recommending brands and performance combinations that meet the overall expectations of customers, thereby improving the success rate of recommendations and customer satisfaction. By deeply understanding customers' brand preferences and performance requirements, companies can better adjust their product structure and optimize market layout, thereby improving their market competitiveness and market share. The results of the related expectation ranking provide strong data support for the company's decision-making, enabling companies to make more scientific and reasonable decisions based on customer behavior data and expectation data. By providing recommendations and services that are more in line with customer expectations, companies can significantly enhance customers' shopping experience and satisfaction, thereby improving customer loyalty and brand reputation.

[0150] In one embodiment of the present invention, the expected matching module includes:

[0151] A matching module, used to match the associated desired brand data with the current car inventory data to obtain multiple brand matching cars of the customer;

[0152] Matching the associated expected performance data with the current car inventory data to obtain multiple performance-matched cars for the customer;

[0153] A matching degree calculation module, used to calculate the demand matching degree of each brand matching car according to the associated expected brand data combined with multiple brand matching car data;

[0154] Calculate the demand matching degree of each performance matching vehicle according to the associated expected performance data combined with the data of multiple performance matching vehicles;

[0155] Degree of matching of the requirements:

[0156] The calculation formula for the demand matching degree of the brand matching car is:

[0157]

[0158] Among them, P A is the demand matching degree of brand matching automobile, j is the total number of brand types associated with the expected brand data, (P g +P t ) i is the matching coefficient between the associated expected brand data of the i-th brand category and the brand category data of the multiple brand matching cars, P g is the associated expected brand data of brand category, P t For multiple brands matching automobile brand category data, when the associated expected brand data of the i-th brand category has the same brand category data in multiple brand matching automobile data, let (P g +P t ) i is 2. When the associated expected brand data of the i-th brand category does not have the same brand category data in multiple brand matching car data, let (P g +P t ) i =1, P k The total number of brands in the current automobile inventory data;

[0159] The calculation formula of the performance matching degree of the automobile is:

[0160]

[0161] Among them, P B is the performance matching degree of the car, M is the total number of performance types associated with the expected performance data, (X g +X t ) N is the matching coefficient between the associated expected performance data of the Nth performance category and the performance category data of the plurality of performance matching vehicles, X gis the expected performance data associated with the performance category, X t is the performance category data of multiple performance matching vehicles. When the associated expected performance data of the Nth performance category has the same performance category data in multiple performance matching vehicle data, let (X g +X t ) N =2, when the associated expected performance data of the Nth performance category does not have the same performance category data in multiple performance matching vehicle data, let (X g +X t ) N =1, X k It is the total number of performance types in the current automobile inventory data.

[0162] A matching sorting module is used to sort the demand matching degree of brand-matching cars and obtain the demand brand sequence;

[0163] Sort the demand matching degree of performance-matching cars to obtain the demand performance sequence;

[0164] The car recommendation module is used to obtain the car corresponding to the associated data with the smallest sum of the serial numbers of the required brand and required performance in the associated expected sorting, as the optimal recommended car;

[0165] The cars corresponding to the associated data with the second smallest to the Nth smallest serial numbers in the sum of the required brand sequence and the required performance sequence are obtained in sequence, and then the optimal car recommendation sequence is obtained, and cars are recommended to customers according to the optimal car recommendation sequence to obtain car recommendation data.

[0166] The working principle of the above technical solution is as follows: match the associated expected brand data with the current car inventory data, find all cars that meet the customer's brand expectations, and form a brand matching car set. Match the associated expected performance data with the current car inventory data, find all cars that meet the customer's performance expectations, and form a performance matching car set. According to the given formula, calculate the demand matching degree of each brand matching car. The formula takes into account the expected strength of each brand category in the associated expected brand data (reflected by the matching coefficient) and the richness of the brand category in the inventory (adjusted by the total number of brand categories). Calculate the demand matching degree of each performance matching car, taking into account the expected strength of the performance category and the richness of the performance category in the inventory. Sort the demand matching degrees of brand matching cars in descending order to generate a demand brand sequence, which represents the ranking of cars with high to low brand matching degrees. Sort the demand matching degrees of performance matching cars in descending order to generate a demand performance sequence. In the associated expectation sorting, find the car corresponding to the associated data with the smallest sum of serial numbers in the demand brand sequence and the demand performance sequence as the best recommended car. This ensures that the recommended car is as close as possible to the customer's comprehensive expectations in terms of brand and performance. Select the cars corresponding to the associated data with the second smallest to the Nth smallest serial number in the required brand sequence and the required performance sequence in turn to form the optimal car recommendation sequence. The cars in this sequence are arranged in descending order of the degree of matching with the customer's needs. According to the optimal car recommendation sequence, recommend cars to customers and generate car recommendation data. This data may include detailed information about the car, reasons for recommendation, price, etc., so that customers can make purchase decisions.

[0167] The technical effect of the above technical solution is: by comprehensively considering the expectations and matching of brand and performance, it is possible to more accurately identify cars that meet customer needs and improve the accuracy and pertinence of recommendations. Personalized recommendations based on customer needs can significantly improve customers' car-buying experience and increase customer satisfaction with brands and services. By analyzing customer needs and inventory data, companies can more accurately predict market trends, optimize inventory structure, and reduce inventory backlogs and waste. By providing recommendations and services that better meet customer expectations, companies can enhance their market competitiveness, attract more potential customers and increase market share. The entire recommendation process is based on a large amount of data and complex calculations, which provides companies with rich data support and decision-making basis, and helps companies make more scientific and reasonable market strategies and product plans.

[0168] In one embodiment of the present invention, the association update module includes:

[0169] The abnormal feedback module is used to obtain customer recommendation feedback information, classify the recommendation feedback information according to preset abnormality categories, and obtain feedback abnormality classification information of multiple preset abnormality categories; the feedback abnormality categories include brand dissatisfaction, performance dissatisfaction, car dissatisfaction, etc.

[0170] An anomaly comparison module, used to compare the feedback anomaly classification information of each preset anomaly category with the preset anomaly threshold of the corresponding category, and compare the category anomaly results;

[0171] The association updating module is used to update the association data of the target vehicle data corresponding to the abnormal category according to the category abnormality comparison result, and then update the optimal vehicle recommendation sequence.

[0172] When the feedback abnormality classification information of any preset abnormality type is greater than the preset abnormality threshold of the corresponding type, the associated data is updated by the feedback abnormality classification information of any preset abnormality type corresponding to the customer recommendation feedback information;

[0173] When the feedback abnormality classification information of any preset abnormality type is less than or equal to the preset abnormality threshold of the corresponding type, no update is performed.

[0174] The working principle of the above technical solution is: obtain customer feedback information on car recommendations, which may include satisfaction scores, specific comments, improvement suggestions, etc. Classify the feedback information according to preset abnormality categories (such as dissatisfied brand, dissatisfied performance, overpriced, etc.) to identify different types of abnormalities or problems. This step will generate feedback abnormality classification information of multiple preset abnormality categories. Compare the feedback abnormality classification information of each preset abnormality category with the preset abnormality threshold of the corresponding category. The abnormality threshold is set based on historical data, business rules or expert opinions, and is used to determine whether the degree of abnormality in the feedback reaches the standard that requires action. Based on the comparison result, a category abnormality comparison result is generated. If the feedback abnormality classification information of a preset abnormality category is greater than its corresponding preset abnormality threshold, it is considered that there is a significant abnormality or problem in the category; otherwise, it is considered that the category is within an acceptable range. When the feedback abnormality classification information of any preset abnormality category is greater than the preset abnormality threshold of the corresponding category, the system will update the associated data according to the abnormal information. The update operation may include adjusting the brand preference weight, performance expectation weight, price sensitivity, etc. to reflect the latest feedback from customers. The updated associated data will re-enter the calculation and generation process of the optimal car recommendation sequence. The system will re-evaluate and sort the car recommendation list based on the updated data and algorithm logic to more accurately reflect the customer's current needs and preferences. If the feedback abnormal classification information of a preset abnormal category is less than or equal to the preset abnormal threshold of the corresponding category, the system considers that the category does not need to be updated and keeps the current recommendation sequence unchanged.

[0175] The technical effects of the above technical solution are: by collecting and analyzing customer feedback information and dynamically adjusting the recommendation system according to the feedback results, the system can adapt to changes in customer needs more quickly and improve the accuracy and satisfaction of recommendations. By promptly responding to and processing customer feedback, the problems and dissatisfaction encountered by customers in the car purchase process are solved, and the customer's car purchase experience and brand loyalty are enhanced. The classification and exception handling of feedback information provide important data support for the optimization of the recommendation algorithm. By continuously learning and improving the algorithm, the intelligence level and recommendation effect of the recommendation system can be improved. Through accurate recommendations and timely feedback processing, the customer's satisfaction and acceptance of the recommendation results can be improved, thereby improving the overall customer satisfaction and brand image. Encouraging customers to provide feedback not only helps to improve the recommendation system, but also enhances the customer's sense of participation and belonging, and promotes a good interactive relationship between customers and enterprises.

[0176] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for automobile intelligent data analysis and decision-making, characterized in that: The method comprises: S1. Obtain historical automobile inventory data, divide it into brand category data and performance category data, preprocess it, obtain the preprocessed brand category data, performance category data and customer demand data, associate various data, and obtain associated integrated data; Wherein, the S1 includes: Acquire historical automobile inventory data, classify the historical automobile inventory data according to automobile brand types, and obtain multiple brand type data; classify the historical automobile inventory data according to automobile performance types, and obtain multiple performance type data; Number each brand type data to obtain the required number data for each car; Number each performance category data to obtain performance number data of each car; Obtain historical customer demand data, number each customer demand data, and obtain the customer number data of each customer; Preprocessing the brand category data, performance category data and customer number data to obtain preprocessed brand category data, performance category data and customer demand data; Obtaining automobile brand category data and performance category data determined by each customer demand data, and associating each of the customer demand data with the automobile brand category data and performance category data corresponding thereto to obtain a plurality of associated data; Combining the plurality of associated data to obtain associated integrated data; S2, acquiring brand category expected data and performance category expected data of the target vehicle according to the associated integrated data, and acquiring associated expected brand data and associated expected performance data according to the brand category expected data and performance category expected data; Wherein, the S2 includes: Obtain the page views in the associated integrated data to perform expected sorting on the brand categories, and obtain the expected data sorting of the brand categories; Obtain the page views in the associated integrated data to perform expected sorting on the performance categories, and obtain the expected data sorting of the performance categories; Calculating the sum of the rankings of each associated data in the associated integrated data, and sorting the associated data according to the sum of the rankings to obtain an associated expected ranking; The brand category expectation data and performance category requirement data in the associated expectation sorting are the associated expected brand data and the associated expected performance data; S3, matching the associated expected brand data and the associated expected performance data with the current car inventory data respectively, calculating the demand matching degree of each brand matching car and performance matching car, and obtaining the optimal car recommendation sequence according to the demand matching degree; Wherein, the S3 includes: Match the associated desired brand data with the current car inventory data to obtain multiple brand matching cars for the customer; Matching the associated expected performance data with the current car inventory data to obtain multiple performance-matched cars for the customer; Calculate the demand matching degree of each brand matching car according to the associated expected brand data combined with multiple brand matching car data; Calculate the demand matching degree of each performance matching vehicle according to the associated expected performance data combined with the data of multiple performance matching vehicles; Sort the demand matching degree of brand-matching cars to obtain the demand brand sequence; Sort the demand matching degree of performance-matching cars to obtain the demand performance sequence; Obtain the car corresponding to the associated data with the smallest sum of the serial numbers of the required brand and required performance in the associated expectation sorting as the best recommended car; Sequentially obtain the cars corresponding to the associated data with the second smallest to the Nth smallest serial numbers in the sum of the required brand sequence and the required performance sequence, and then obtain the optimal car recommendation sequence, and recommend cars to customers according to the optimal car recommendation sequence to obtain car recommendation data; S4. Obtain customer recommendation feedback information, determine whether to update data association according to the customer recommendation feedback data, obtain a determination result, update the data association according to the determination result, and then update the optimal car recommendation sequence.

2. The method for automobile intelligent data analysis and decision-making according to claim 1, characterized in that: The S4 includes: Obtain customer recommendation feedback information, classify the recommendation feedback information according to preset abnormality categories, and obtain feedback abnormality classification information of multiple preset abnormality categories; Compare the feedback abnormal classification information of each preset abnormal category with the preset abnormal threshold of the corresponding category, and compare the category abnormality results; According to the type anomaly comparison result, the associated data of the target car data corresponding to the abnormal type is updated, and then the optimal car recommendation sequence is updated.

3. An automobile intelligent data analysis and decision-making system, characterized in that: The system comprises: The demand association module is used to obtain historical automobile inventory data, divide it into brand category data and performance category data, pre-process it, obtain the pre-processed brand category data, performance category data and customer demand data, associate various data, and obtain associated integrated data; Wherein, the demand association module includes: An information acquisition module is used to acquire historical automobile inventory data, classify the historical automobile inventory data according to automobile brand types, and obtain multiple brand type data, and classify the historical automobile inventory data according to automobile performance types, and obtain multiple performance type data; The numbering module is used to number each brand type data and obtain the required number data for each car; Number each performance category data to obtain performance number data of each car; Obtain historical customer demand data, number each customer demand data, and obtain the customer number data of each customer; A preprocessing module, used for preprocessing the brand category data, performance category data and customer number data to obtain preprocessed brand category data, performance category data and customer demand data; The association integration module is used to obtain the brand category data and performance category data of the automobile determined by each customer demand data, and associate each of the customer demand data with the corresponding brand category data and performance category data of the automobile to obtain a plurality of associated data; Combining the plurality of associated data to obtain associated integrated data; an expectation acquisition module, configured to acquire brand category expectation data and performance category expectation data of a target vehicle according to the associated integrated data, and acquire associated expected brand data and associated expected performance data according to the brand category expectation data and performance category expectation data; Wherein, the expectation acquisition module includes: An expected ranking module is used to obtain the page views in the associated integrated data to perform expected ranking on the brand categories, thereby obtaining the expected data ranking of the brand categories; Obtain the page views in the associated integrated data to perform expected sorting on the performance categories, and obtain the expected data sorting of the performance categories; Calculating the sum of the rankings of each associated data in the associated integrated data, and sorting the associated data according to the sum of the rankings to obtain an associated expected ranking; An expected data acquisition module, used for the brand category expected data and performance category required data in the associated expected sorting, namely the associated expected brand data and the associated expected performance data; An expectation matching module, used to match the associated expected brand data and the associated expected performance data with the current car inventory data respectively, calculate the demand matching degree of each brand matching car and performance matching car, and obtain the optimal car recommendation sequence according to the demand matching degree; Wherein, the expected matching module includes: A matching module, used to match the associated desired brand data with the current car inventory data to obtain multiple brand matching cars of the customer; Matching the associated expected performance data with the current car inventory data to obtain multiple performance-matched cars for the customer; A matching degree calculation module, used to calculate the demand matching degree of each brand matching car according to the associated expected brand data combined with multiple brand matching car data; Calculate the demand matching degree of each performance matching vehicle according to the associated expected performance data combined with the data of multiple performance matching vehicles; A matching sorting module is used to sort the demand matching degree of brand-matching cars and obtain the demand brand sequence; Sort the demand matching degree of performance-matching cars to obtain the demand performance sequence; The car recommendation module is used to obtain the car corresponding to the associated data with the smallest sum of the serial numbers of the required brand and required performance in the associated expected sorting, as the optimal recommended car; Sequentially obtain the cars corresponding to the associated data with the second smallest to the Nth smallest serial numbers in the sum of the required brand sequence and the required performance sequence, and then obtain the optimal car recommendation sequence, and recommend cars to customers according to the optimal car recommendation sequence to obtain car recommendation data; The association update module is used to obtain customer recommendation feedback information, determine whether to update the data association according to the customer recommendation feedback data, obtain a judgment result, update the data association according to the judgment result, and then update the optimal car recommendation sequence.

4. The automobile intelligent data analysis and decision-making system according to claim 3 is characterized in that: The association update module comprises: The abnormal feedback module is used to obtain customer recommendation feedback information, classify the recommendation feedback information according to preset abnormality categories, and obtain feedback abnormality classification information of multiple preset abnormality categories; An anomaly comparison module, used to compare the feedback anomaly classification information of each preset anomaly category with the preset anomaly threshold of the corresponding category, and compare the category anomaly results; The association updating module is used to update the association data of the target vehicle data corresponding to the abnormal category according to the category abnormality comparison result, and then update the optimal vehicle recommendation sequence.

Images