Intelligent vehicle type screening method for vehicle renting platform

Abstract

The invention relates to an intelligent vehicle model screening method of a vehicle renting platform, and relates to the field of vehicle model screening of the platform, and the method comprises the steps: obtaining a vehicle renting demand and identity information inputted by a user in response to a vehicle renting signal, and screening a required vehicle model from a vehicle model database based on the vehicle renting demand; determining a preliminary screening number based on the required vehicle type; when the preliminary screening number is smaller than the rapid browsing critical number, defining the required vehicle model as a final screening vehicle model and outputting the final screening vehicle model; when the preliminary screening number is larger than the rapid browsing critical number, type selection association information is searched from a vehicle type association database based on the identity information; searching a conversion demand from a conversion database according to the type selection associated information; screening a converted vehicle model from a vehicle model database based on the conversion demand; determining a comprehensive screening vehicle type based on the required vehicle type and the converted vehicle type; and taking the comprehensively screened vehicle model as a final screened vehicle model, and outputting the final screened vehicle model on a vehicle model display interface. The method has the effects of reasonably screening the vehicle types required by the user and improving the screening accuracy.

Description

Technical Field

[0001] This invention relates to the field of vehicle selection on a platform, and in particular to an intelligent vehicle selection method for a car rental platform. Background Technology

[0002] In the digital transformation of the car rental industry, efficient matching of vehicle resources has become a core element for platforms to improve user experience and operational efficiency. As car rental users' needs become increasingly diversified (e.g., different scenarios such as business travel, family tourism, and short-distance commuting) and the platform's vehicle database continues to expand (covering various types and price ranges of vehicles including sedans, SUVs, MPVs, and new energy vehicles), the traditional "passive search + fixed classification" vehicle recommendation model can no longer meet users' demands for "precise, personalized, and efficient" matching. Therefore, intelligent vehicle selection mechanisms, as the core bridge connecting massive vehicle resources with dynamic user needs, are becoming key bottlenecks restricting the upgrading of car rental platform service quality and the improvement of conversion efficiency, particularly in terms of the accuracy of their selection logic, the real-time response, the flexibility of scenario adaptation, and the full utilization of data.

[0003] The core of the intelligent vehicle selection technology in car rental platforms revolves around four major stages: "demand analysis, feature matching, strategy optimization, and result presentation." It covers key technical modules such as user intent recognition, vehicle feature modeling, multi-dimensional algorithm ranking, and dynamic interactive feedback.

[0004] Regarding the aforementioned technologies, most car rental platforms currently rely on users actively inputting complete requirements (such as specifying the type of car, rental budget, pick-up and drop-off times, etc.). If users do not have clear initial requirements or their requirements are vague (such as only mentioning "suitable for family trips" without specifying the number of people or budget), the platform can often only output a generalized list of car models, causing users to spend a lot of time browsing and filtering. Summary of the Invention

[0005] To avoid the problem of an overly broad selection range due to insufficient or no user demand, this invention provides an intelligent vehicle selection method for car rental platforms.

[0006] A smart vehicle selection method for a car rental platform includes: Step 1: In response to the car rental signal, obtain the user's input car rental request and identity information, including historical car rental data; Step 2: Based on the car rental requirements, filter out the corresponding car models from the preset car model database; Step 3: Determine the initial screening quantity based on the required vehicle models; Step 4: When the number of vehicles selected in the initial screening is less than the preset threshold for quick browsing, define the required vehicle models as the final selected vehicle models and output them. Step 5: When the number of initial screenings exceeds the threshold number for quick browsing, the corresponding model selection information is retrieved from the preset model association database based on the identity information; Step 6: Find the corresponding conversion requirements from the preset conversion database using the selection association information; Step 7: Based on the conversion requirements, select the corresponding conversion models from the vehicle model database; Step 8: Determine the comprehensive screening models based on the required vehicle type and the converted vehicle type; Step 9: The comprehensively selected vehicle models are used as the final selected vehicle models and output on the preset vehicle model display interface. The vehicle model display interface displays the basic vehicle information corresponding to the comprehensively selected vehicle models.

[0007] By adopting the above technical solution, potential filtering needs can be mined based on user identity information, further narrowing down the range of vehicle models to be filtered. This effectively avoids the problem of too many filtering results due to concise user requests, thus shortening the time users spend selecting a car and improving the accuracy of vehicle model filtering.

[0008] Optionally, a method for updating the final selected vehicle models may also be included, which includes: Step 90: After the final selected vehicle model is output, in response to the vehicle viewing signal, the corresponding vehicle data is obtained and output to the preset vehicle data interface. The vehicle data interface displays detailed vehicle data, vehicle pictures from various angles, and vehicle rental plans. Step 91: When the user is in the vehicle data interface, obtain the corresponding vehicle data and vehicle viewing duration; Step 92: When the viewing time of the vehicle exceeds the preset viewing time, the vehicle data is defined as interest vehicle data; Step 93: Obtain the vehicle model selection based on the interest vehicle data; Step 94: Based on the interest-based vehicle selection, update the final selected vehicle models and output them to the vehicle model display interface.

[0009] By adopting the above technical solution, after outputting the final filtered models, the system identifies models of interest by combining user vehicle viewing behavior and duration, and then updates the filtering results based on the models of interest. This not only optimizes the range of models to be filtered based on users' real-time preferences, but also further shortens the time users spend selecting a car and improves the accuracy of model filtering.

[0010] Optionally, the method also includes a method for filtering vehicle models with a viewing duration of 0, the method comprising: Step 910: Obtain the model selection dwell time on the model display interface based on the model display interface; Step 911: When the selection dwell time exceeds the preset selection time or a preset refresh signal is received, a non-intersection filter model is obtained based on the required model and the converted model; Step 912: Output the non-intersection filtered models as the final filtered models to the model display interface; Step 913: Upon receiving the vehicle viewing signal, stop accumulating the selection dwell time and do not update the final filtered vehicle model.

[0011] By adopting the above technical solution, the updated results of non-intersecting filtered models can be retrieved based on the selection dwell time on the model display interface or the user refresh signal. At the same time, when the user triggers the vehicle viewing signal, the pause dwell time is accumulated without changing the filtering results, which effectively avoids the problem of users being dissatisfied with the existing filtering results but having no new models to choose from, and improves the adaptability of model filtering.

[0012] Optionally, the method for obtaining the selection dwell time includes: Step 9101: If the vehicle viewing time exceeds the preset basic loading time of the vehicle data interface, output the preset basic vehicle information to the vehicle data interface. Step 9102: Upon receiving the preset detailed data signal, output the preset vehicle detailed data to the vehicle data interface after a preset detailed loading time; Step 9103: Obtain the total loading time based on the basic loading time and the detailed loading time; Step 9104: If the vehicle viewing time exceeds the total loading time, stop accumulating the selection dwell time; Step 9105: If the vehicle viewing time does not exceed the total loading time, then the selection dwell time continues to accumulate.

[0013] By adopting the above technical solution, the system determines whether the user wants to view the selected car model based on the loading time and the user's operation, thus avoiding the situation where the system misjudges the car model of interest due to accidental touch, and improving the reliability of the system and the accuracy of car model selection.

[0014] Optionally, the method for obtaining the vehicle viewing duration includes: Step 91050: If the user is in the vehicle data interface, the vehicle viewing time will be continuously accumulated; Step 91051: If the user is in the vehicle display interface, stop accumulating the vehicle viewing time and output it; Step 91052: When the user's vehicle viewing time on the vehicle data interface exceeds the total loading time, the vehicle viewing time is continuously accumulated based on the previous vehicle viewing time. Step 91053: If the user's vehicle viewing time on the vehicle data interface has not exceeded the total loading time, then the previous vehicle viewing time is output.

[0015] By adopting the above technical solution, the system can achieve intelligent accumulation and interruption control of vehicle viewing time by accurately tracking interface switching behavior. When a user re-enters the vehicle data interface and the viewing time exceeds the total loading time, the system automatically continues the historical time accumulation; if the time does not exceed the limit, the previous record is used as valid data. This not only avoids record interruption caused by accidental operation, but also improves the accuracy of the filtering criteria.

[0016] Optionally, the method also includes an optimization method for the final selected vehicle models, the method comprising: Step 910510: Upon receiving the vehicle data viewed by the user, obtain the vehicle type corresponding to the vehicle data; Step 910511: Sort the vehicle types according to the vehicle viewing duration corresponding to the vehicle type to obtain the vehicle model viewing duration sort. Step 910512: Based on the vehicle model duration sorting, sort the vehicle types on the vehicle model display interface and output them.

[0017] By adopting the above technical solution, when users view vehicle data, the system can retrieve the corresponding vehicle type and sort them based on the viewing duration. The vehicles are then displayed on the model showcase interface according to this sorting. Prioritizing models that users are more interested in reduces the time users spend searching for their desired models and effectively improves the efficiency of model selection.

[0018] Optionally, the method also includes optimizing the final filtered vehicle models based on the vehicle viewing duration, the method comprising: Step 9105110: If the vehicle viewing time exceeds the preset viewing time threshold, obtain the advantage data of the vehicle type; Step 9105111: Based on the advantage data, find the vehicle type in the corresponding vehicle type that has the advantage data and whose other preset key data are all better than the vehicle type, and then output it to the vehicle model display interface; Step 9105112: If the number of vehicle types exceeding the viewing time threshold exceeds 1, then the advantage data of the corresponding vehicle types are integrated to obtain comprehensive advantage data. Step 9105113: Update the final selected vehicle models based on the comprehensive advantage data and output the updated models.

[0019] By adopting the above technical solution, we can extract the advantageous data of car models that users have viewed for a longer period of time and push better car models to them. When multiple car models meet the criteria, we can optimize the filtering results by integrating comprehensive advantageous data and expand the high-quality options by associating advantageous data, thereby improving the accuracy of car model filtering.

[0020] Optionally, the method also includes a method for filtering car models based on the historical car rental data, the method comprising: Step 80: Obtain the similarity between the user's input of the car rental request and the historical car rental data; Step 81: If the similarity of the demand is greater than the preset similarity threshold, then based on the user's input car rental demand and the historical car rental data, obtain similar filter models as the final filter models and output them to the model display interface. Step 82: If the similarity of the demand is less than the similarity threshold, then the corresponding car model is selected from the car model database based on the car rental demand input by the user. Step 83: Based on the required vehicle type, filter the vehicle types in the historical car rental data to obtain vehicles with no history as the final filtered vehicle types and output them to the vehicle display interface.

[0021] By adopting the above technical solution, differentiated filtering is performed based on the similarity between users' car rental needs and historical car rental data. Historical data is fully utilized to improve the relevance of the filtering. When there are significant differences in needs, new options are proactively recommended. This approach balances the continuity of user preferences with the possibility of exploration, thereby improving the accuracy of vehicle selection.

[0022] Optionally, the method for filtering the vehicle types in the historical car rental data according to the required vehicle type includes: Step 830: Based on the user's historical car rental data, find the corresponding historical car rental patterns; Step 831: If the historical car rental pattern exists, then based on the car rental demand and the historical car rental pattern, find the corresponding historical car model in the car model database and output it as the final filtered car model to the car model display interface; Step 832: If the historical car rental pattern does not exist, then based on the car rental demand and the identity information, the corresponding demanded car model is selected from the car model database as the final selected car model and output to the car model display interface.

[0023] By adopting the above technical solution and combining historical car rental data to mine rental patterns, when a clear pattern exists, matching historical car models are selected based on current rental needs; if no obvious pattern exists, corresponding car models are selected based on rental needs and user identity information. The accuracy of the selection is improved by utilizing historical patterns, and the relevance of the selection is ensured by using basic information when there is no pattern, thus further optimizing the matching of car rental models.

[0024] Optionally, it also includes a method for selecting the corresponding desired vehicle model from the vehicle model database based on the historical car rental data when the historical car rental pattern does not exist. This method includes: Step 8320: If the historical car rental pattern does not exist, then obtain the user's repeated car rental data based on the historical car rental data; Step 8321: If the duplicate car rental data exists, then the final selected car models are filtered from the car model database according to the car rental demand and the duplicate car rental data and output to the car model display interface; Step 8322: If the duplicate car rental data does not exist, then the car models without historical records are obtained by filtering based on the required car model and the historical car rental data and output as the final filtered car models to the car model display interface.

[0025] By adopting the above technical solution, when users do not have a clear historical car rental pattern, it is possible to further mine duplicate car rental information in historical data. This not only captures potential user preferences through duplicate data, but also provides new options when there are no duplicate records, taking into account both the accuracy and diversity of the screening, and further optimizing the screening effect of car rental models.

[0026] In summary, the present invention has at least one of the following beneficial technical effects: By uncovering users' hidden car rental needs through user information, and further filtering car models based on these hidden needs, the problem of excessively broad selection ranges due to limited or no user demand is avoided, thus reducing user selection time and improving the accuracy of car model filtering. By automatically updating the filtered car models based on user viewing time and operational behavior across different interfaces, the problem of incorrectly filtering car models and misjudging user interests is avoided, improving the accuracy of user car rental information collection and car model filtering. Furthermore, by comparing user needs with historical data, the problem of outputting historical rental data even when current rental needs differ from past rental needs is avoided, further improving the accuracy of car model filtering. Attached Figure Description Figure 1 This is a flowchart of an intelligent vehicle selection method for a car rental platform according to an embodiment of this application; Figure 2This is a block diagram of the vehicle display interface in the embodiments of this application; Figure 3 This is a flowchart of the method for updating the final selected vehicle models in the embodiments of this application; Figure 4 This is a block diagram of the vehicle data interface in the embodiments of this application; Figure 5 This is a flowchart of the vehicle model filtering method with a vehicle viewing duration of 0 in the embodiments of this application; Figure 6 This is a flowchart of the optimization method for the final selection of vehicle models in the embodiments of this application; Figure 7 This is a flowchart of the method for optimizing the final vehicle model selection based on vehicle viewing time in an embodiment of this application; Figure 8 This is a flowchart illustrating the method for filtering car models based on historical car rental data in this application embodiment. Detailed Implementation

[0027] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0028] This invention discloses an intelligent vehicle selection method for a car rental platform. (Refer to...) Figure 1 A smart vehicle selection method for a car rental platform includes: Step 1: In response to the car rental signal, obtain the user's car rental request and identity information.

[0029] The car rental signal refers to the user's entry into the car rental platform and the completion of their rental request. The response is indicated by the car rental platform's software icon; clicking this icon signifies a response.

[0030] Car rental needs refer to the user's desired car model. This is obtained by the system sending the user a form with available car models created by staff; the user fills out the form to obtain the desired information. Identity information refers to the user's personal information, including age, location, experience, and rental history. This identity information is related to car model selection. For example, if the information indicates a 22-year-old from a wealthy family who is familiar with niche supercar brands, they might rent a high-performance supercar rather than a family sedan or SUV. This is obtained by the system sending the user a pre-set personal information form after receiving a rental request; the user fills out the form to obtain the desired information.

[0031] Step 2: Based on your car rental needs, select the corresponding car model from the pre-set car model database.

[0032] The desired car model refers to the type of car a user might be looking for. The car model database stores a mapping between rental requests and desired car models. Experts in this field analyze the possible car models corresponding to the rental request based on their experience and available online information, then input this data into the database. For example, a married man might choose a family sedan or SUV, a business owner might choose a business sedan, and a young, independent woman might choose a car with an attractive appearance or car cover. When a user request is received, it is analyzed, and the corresponding car model that meets the user's needs is retrieved from the database and output.

[0033] Step 3: Determine the initial screening quantity based on the required vehicle models.

[0034] The initial screening quantity refers to the number of car models selected based on rental needs. This is determined by the system calculating the number of available car models from the requested models based on pre-entered information from professionals in the field.

[0035] Step 4: When the number of vehicles initially selected is less than the preset threshold for quick browsing, define the required vehicle models as the final selected vehicle models and output them.

[0036] The quick browse threshold refers to the number of vehicle models that can be browsed within a pre-set time. This threshold is calculated by inputting the fastest browsing time for images found online by staff. The final filtered vehicle models refer to the models ultimately determined and output based on user requirements.

[0037] If the number of vehicles selected in the initial screening is less than the preset threshold for quick browsing, it means that the range of vehicles selected meets the user's needs. At this point, the corresponding vehicles will be output as the final selected vehicles.

[0038] Step 5: When the number of initial screenings exceeds the threshold for quick browsing, the corresponding model selection information is retrieved from the preset model association database based on the identity information.

[0039] Model selection association information refers to hidden needs related to vehicle selection found within identity information. The vehicle model association database stores the mapping relationship between identity information and model selection association information. This information is generated by professionals in the field who analyze the identity information based on their experience and relevant online resources, then input this information into the database. For example, if the identity information includes gender (female), age (22), moderate spending power, ethnicity (Miao), and hobbies (bungee jumping, skydiving), then the model selection association information would be: 22 years old, moderate spending power, and a love for extreme sports.

[0040] If the number of initial screenings exceeds the threshold for quick browsing, it indicates that the screening scope is too broad and inconvenient for users to select a car. Therefore, we analyze the user's hidden needs based on their identity information.

[0041] Step 6: Find the corresponding conversion requirements from the preset conversion database by selecting the relevant information.

[0042] Conversion requirements refer to car rental needs analyzed based on user identity information. The conversion database stores the mapping relationship between selection-related information and car rental needs, pre-stored by professionals in this field based on the correlation between different car model characteristics and needs. For example, a 22-year-old with a moderate spending power who enjoys extreme sports would have a conversion requirement of a high-performance sports car with a moderate price, satisfying driving experience; a 32-year-old married man with a high spending power and a family-oriented lifestyle would have a conversion requirement of a car that provides comfortable and safe rides for his family. Upon receiving selection-related information, the database analyzes the hidden needs within the user's identity information.

[0043] Step 7: Based on the conversion requirements, select the corresponding conversion models from the vehicle database.

[0044] The "converted vehicle model" refers to the vehicle models selected based on user identity information. The selection method here involves the system obtaining conversion requests and then searching the vehicle model database for the corresponding model.

[0045] Step 8: Determine the comprehensive selection of vehicle models based on demand and conversion models.

[0046] The comprehensive vehicle selection refers to a narrower range of vehicles obtained after secondary screening. This is determined by identifying the intersection of the desired and target vehicle types, and using this intersection as the comprehensive selection criteria.

[0047] Step 9: Use the comprehensively filtered models as the final models and output them on the preset model display interface.

[0048] like Figure 2 The vehicle display interface shows the basic vehicle information corresponding to the comprehensively filtered models.

[0049] Reference Figure 3 It also includes a method for updating the final selected vehicle models, which includes: Step 90: After the final vehicle model is output, in response to the vehicle viewing signal, the corresponding vehicle data is obtained and output to the preset vehicle data interface.

[0050] Viewing vehicle signals refers to the user selecting a car model and wanting to view the signal for that specific vehicle. The response method is the vehicle icon corresponding to the final filtered car model on the car model display interface; clicking the vehicle icon triggers the response.

[0051] Vehicle data refers to the data of one vehicle in the final selection of models. Vehicle data includes vehicle performance, vehicle space, vehicle functions, and appearance data.

[0052] like Figure 4 The vehicle data interface displays detailed vehicle information, images of the vehicle from various angles, and vehicle rental plans. The data is retrieved by the system when a user clicks on a vehicle image, and the system searches for the corresponding vehicle data from a pool of vehicle models pre-entered by personnel in this field. For example, if a user clicks on an image of a BMW 330i, the system will retrieve and output the vehicle data for the BMW 330i.

[0053] Step 91: When the user is in the vehicle data interface, obtain the corresponding vehicle data and vehicle viewing duration.

[0054] Vehicle viewing duration refers to the time a user views the vehicle.

[0055] When a user is on the vehicle data interface, it means that the user is viewing the vehicle and may be interested in it. At this time, the system records the vehicle data and accumulates the viewing time.

[0056] Step 92: When the viewing time of a vehicle exceeds the preset viewing time, define the vehicle data as interest vehicle data.

[0057] Viewing duration refers to the minimum time required to view the vehicle's data. The viewing duration here is based on the shortest viewing time found online by staff.

[0058] Interest vehicle data refers to data related to vehicles that a user may be interested in.

[0059] When the viewing time for a vehicle exceeds the preset viewing time, it indicates that the user may have a comprehensive understanding of the vehicle and may want to rent it; therefore, the vehicle data is defined as "vehicle of interest data." For example, if viewing data for a BMW 330i takes at least 3 minutes, then after 3 minutes, the BMW 330i data is defined as "vehicle of interest data."

[0060] Step 93: Obtain interest-based vehicle models based on interest vehicle data.

[0061] The "Interest-Based Vehicle Selection" refers to the vehicle models that the user might be interested in from the final selection. This is obtained by the system based on the user's viewing duration. For example, if a user views the BMW 330i and Mercedes-Benz C260, both German sports sedans, then German sports sedans will be selected as the "Interest-Based Vehicle Selection."

[0062] Step 94: Filter car models based on interests. Update the final filtered car models and output them to the car model display interface.

[0063] The update method here is to replace the original final selected model with the intersection of the models selected by interest and the models selected by final selection. For example, if the models selected by interest only include German sports sedans, but the final selected models do not include Porsche sports sedans or French supercars, then the Porsche models from the French supercars and German sports sedans will be deleted before outputting.

[0064] Reference Figure 5 It also includes a method for filtering vehicle models with a viewing duration of 0, which includes: Step 910: Obtain the selection dwell time on the vehicle display interface based on the vehicle display interface.

[0065] The selection dwell time refers to the amount of time a user spends on the vehicle display screen. This is obtained by measuring the time the user spends on the vehicle display screen.

[0066] Step 911: If the selection dwell time exceeds the preset selection time or a preset refresh signal is received, then non-intersecting filter models are obtained based on the required model and the converted model.

[0067] The selected time refers to the maximum time the vehicle model is displayed on the screen. This time selection is set by staff based on their own viewing time of the vehicle model display screen. The refresh signal refers to the signal the user wants to refresh the vehicle model displayed on the screen. This signal is received by the user clicking the refresh button on the vehicle model display screen or by swiping up and releasing the button to refresh.

[0068] Non-intersection filtering refers to vehicle models that are neither part of the demand model nor the conversion model. This is achieved by analyzing vehicle models that do not simultaneously belong to either the demand model or the conversion model. For example, if the demand model includes American sports cars and domestically produced coupes, and the conversion model includes European coupes and domestically produced coupes, then the non-intersection filtering would include American sports cars and European coupes.

[0069] If the user stays in the selection process for longer than the preset selection time or receives a refresh signal, it indicates that the user is not viewing the currently displayed models. Therefore, in this case, a non-overlapping selection of models will be generated based on the desired and converted models. For example, if the currently displayed models are domestically produced sports sedans, and the user has not clicked on any vehicle image within 5 minutes, then a non-overlapping selection of models will be generated based on the desired and converted models.

[0070] Step 912: Output the non-intersecting filtered models as the final filtered models to the model display interface.

[0071] Step 913: Upon receiving a vehicle viewing signal, stop accumulating the selection dwell time and do not update the final selected vehicle model.

[0072] Upon receiving a vehicle viewing signal, it indicates that the user is viewing vehicles from the currently displayed filter models. Therefore, the cumulative selection dwell time is stopped and the final filter models are not updated.

[0073] The methods for obtaining the duration of the selection process include: Step 9101: If the vehicle viewing time exceeds the preset basic loading time of the vehicle data interface, output the preset basic vehicle information to the vehicle data interface.

[0074] The basic loading time refers to the time required to load basic information. This basic loading time is set by those skilled in the art based on the duration of data loading. For example: 3 seconds, 5 seconds.

[0075] like Figure 4 Interface 1, Vehicle Basic Information, refers to the vehicle's key data, including vehicle length, passenger capacity, fuel consumption, and front / rear trunk capacity. This basic information is entered pre-processed by staff using a basic information form; the system then inputs relevant vehicle data found online based on the form's content.

[0076] If the vehicle viewing time exceeds the basic loading time of the vehicle data interface, it means that the user wants to get a preliminary understanding of the vehicle, so the basic vehicle information is output.

[0077] Step 9102: Upon receiving the preset detailed data signal, output the preset vehicle detailed data to the vehicle data interface after a preset detailed loading time.

[0078] Detailed data signals refer to signals that users want to learn more about the vehicle. These signals can be received by clicking the "View Detailed Data" button on the vehicle data interface or by swiping through the vehicle data interface.

[0079] like Figure 4 Interface 2, Vehicle Detailed Data, refers to all known data about the vehicle, including rental price, vehicle performance, 0-100 km / h acceleration, and top speed. This detailed vehicle data is pre-entered into a data table by staff, and the system then inputs relevant vehicle data found online based on the table's content. Detailed Loading Time refers to the time required to load the vehicle detailed data. This loading time is set by staff based on the data loading duration, for example, 3 seconds or 5 seconds.

[0080] When a preset detailed data signal is received, it indicates that the user wants to know more about the vehicle. At this time, the detailed vehicle data is output after detailed loading time.

[0081] Step 9103: Calculate the total loading time based on the basic loading time and detailed loading time.

[0082] The total loading time is the time spent loading basic vehicle information and loading detailed vehicle data. It is obtained by adding the basic loading time and the detailed loading time together.

[0083] Step 9104: If the vehicle viewing time exceeds the total loading time, stop accumulating the selection dwell time.

[0084] If the vehicle viewing time exceeds the total loading time, it means that there are vehicles that meet the user's needs in the final selection of models, so the cumulative selection time is stopped.

[0085] Step 9105: If the vehicle viewing time does not exceed the total loading time, the selection dwell time will continue to accumulate.

[0086] If the vehicle viewing time does not exceed the total loading time, it indicates that the user did not intentionally enter the vehicle's data interface, and therefore the selection dwell time continues to accumulate. The accumulation method here is based on the previous selection dwell time. For example: if the user spends 3 minutes and 14 seconds on the vehicle model display interface, the total loading time is 8 seconds, and the user is on the vehicle data interface for 3 seconds, and then returns to the vehicle model display interface after 3 seconds, then the selection dwell time continues to accumulate from 3 minutes and 14 seconds.

[0087] The methods for obtaining vehicle viewing duration include: Step 91050: If the user is in the vehicle data interface, the vehicle viewing time will be continuously accumulated.

[0088] If a user is on the vehicle data interface, it means that the user wants to continue viewing the vehicle data, so the vehicle viewing time will be continuously accumulated.

[0089] Step 91051: If the user is on the vehicle model display interface, stop accumulating vehicle viewing time and output it.

[0090] If the user is on the vehicle model display screen, it means the user will not continue to view the vehicle data. At this time, the accumulated vehicle viewing time will be stopped and output.

[0091] Step 91052: When the user's vehicle viewing time on the vehicle data interface exceeds the total loading time, the vehicle viewing time is continuously accumulated based on the previous vehicle viewing time.

[0092] When the time a user spends viewing a vehicle on the vehicle data interface exceeds the total loading time, it indicates that the user wants to review or continue to view the vehicle data. Therefore, the vehicle viewing time is continuously accumulated based on the previous vehicle viewing time.

[0093] Step 91053: If the user's vehicle viewing time on the vehicle data interface has not exceeded the total loading time, then output the previous vehicle viewing time.

[0094] If the user's vehicle viewing time on the vehicle data interface does not exceed the total loading time, it means that the user does not intend to view the vehicle data again or continue to view it. Therefore, the vehicle viewing time is not accumulated and the previous vehicle viewing time is output.

[0095] Reference Figure 6 It also includes an optimization method for the final selection of vehicle models, which includes: Step 910510: Upon receiving vehicle data viewed by the user, obtain the vehicle type corresponding to the vehicle data.

[0096] Vehicle type refers to the category to which the vehicle belongs. This is obtained by the system finding the vehicle type corresponding to the vehicle data pre-entered by the staff.

[0097] Upon receiving vehicle data viewed by a user, it indicates that the vehicle is one the user intends to rent, and therefore the vehicle type corresponding to the vehicle data is obtained.

[0098] Step 910511: Sort vehicle types based on the vehicle viewing time corresponding to the vehicle type to obtain the vehicle model viewing time sort.

[0099] The vehicle model viewing duration sort refers to the sorting of vehicle models based on the viewing duration. This is achieved by sorting vehicle types according to the length of time they have been viewed.

[0100] Step 910512: Sort the vehicle types in the vehicle display interface based on vehicle model duration and then output them.

[0101] For example, if a user spends 5 minutes looking at Porsche, 10 minutes looking at Alfa Romeo, and 7 minutes looking at Bentley, then Alfa Romeo will be displayed first on the car display screen, followed by Bentley, and finally Porsche.

[0102] Reference Figure 7 It also includes a method for optimizing the final vehicle model selection based on vehicle viewing time, which includes: Step 9105110: If the vehicle viewing time exceeds the preset viewing time threshold, obtain the vehicle type's advantage data.

[0103] The viewing time threshold refers to the longest reference time for viewing data for that vehicle. This threshold is determined by staff through the longest reading time of relevant data found online.

[0104] Advantageous data refers to the data that makes a vehicle stand out among other vehicles of the same type. This data is obtained by comparing the vehicle's data with that of other vehicles. Examples include: spacious interior, rapid vehicle response, good sound insulation, high safety, and low fuel consumption.

[0105] If the viewing time for a vehicle exceeds the viewing time threshold, it indicates that the user is interested in the vehicle but may not be satisfied with some of its shortcomings. Therefore, we need to obtain the advantages data of the vehicle type.

[0106] Step 9105111: Based on the advantageous data, find the vehicle type with advantageous data and whose other preset key data are all better than that vehicle type in the corresponding vehicle type, and output it to the vehicle model display interface.

[0107] Other key data refers to data that may affect the user during driving. The search method here involves the system retrieving relevant vehicle model key data from the information provided by staff through their experience and online research, then filtering for the most relevant data. Examples include: in-vehicle functions, vehicle performance, and vehicle exterior data.

[0108] Step 9105112: If the number of vehicle types exceeding the viewing time threshold exceeds 1, then the advantage data of the corresponding vehicle types will be integrated to obtain comprehensive advantage data.

[0109] The comprehensive advantage data refers to the advantage data possessed by multiple vehicles individually. This data is obtained by the system aggregating the advantage data of each vehicle whose user viewing time exceeds a certain threshold.

[0110] If more than one vehicle type exceeds the viewing time threshold, it indicates that users have different points of satisfaction and dissatisfaction with different vehicle models. Therefore, the advantage data of the corresponding vehicle types will be integrated to obtain comprehensive advantage data.

[0111] Step 9105113: Based on the comprehensive advantages data, the final selected models will be updated and output.

[0112] The update method here is to select models with comprehensive advantages from the final selected models and update them as the new final selected models.

[0113] Reference Figure 8 It also includes a method for filtering car models based on historical car rental data, which includes: Step 80: Obtain the similarity of the user's input car rental needs and historical car rental data.

[0114] Historical car rental data refers to the models of cars a user has previously rented and their corresponding vehicle information. Demand similarity refers to the degree of similarity between a user's car rental needs and historical car rental data. This is obtained by comparing key information from the rental needs with historical car rental data.

[0115] Step 81: If the similarity of the demand is greater than the preset similarity threshold, then the similar car models are obtained based on the user's input car rental demand and historical car rental data as the final filtered car models and output to the car model display interface.

[0116] The similarity threshold is a standard value used to judge the similarity between car rental requests and historical car rental data. This similarity threshold is determined experimentally by professionals in the field. The experiment involves setting a threshold where the submitted request contains most of the characteristics of the desired car model, and the model falls into the selection pool at a certain similarity level. For example, if the rental request includes a sports sedan, convertible, and BMW, and the historical rental data shows that the BMW Z4 is a BMW, a Z series car, a sports sedan, and a convertible, then the selected models based on the user's request include the BMW Z4, achieving a similarity of 75%. Therefore, 75% is set as the similarity threshold.

[0117] Similar car models are those that are similar to the car rental needs and historical car rental data. This is obtained by combining the car models selected based on rental needs with those selected based on historical car rental data.

[0118] If the similarity of the demand is greater than the preset similarity threshold, it means that the user may want to rent the same car model as the historical car rental data. Therefore, similar car models are obtained based on the user's input car rental demand and historical car rental data.

[0119] Step 82: If the similarity of the requirements is less than the similarity threshold, then the corresponding car model is selected from the car model database based on the user's input car rental requirements.

[0120] If the similarity of the requests is less than the similarity threshold, it means the user does not want to rent a car model related to historical data. In this case, the system filters the car model database based on the user's input rental request. The filtering method here is for the system to find a suitable car model in the database based on the rental request.

[0121] Step 83: Based on the required vehicle type, filter the vehicle types in the historical car rental data to obtain vehicles without a history of rentals as the final filtered vehicles and output them to the vehicle display interface.

[0122] "No historical vehicle models" refers to the final selection of vehicle models that do not include those from historical rental data. This is obtained by filtering historical vehicle models from the desired vehicle type.

[0123] One method for filtering historical car rental data by vehicle type based on the desired vehicle model includes: Step 830: Find the corresponding historical car rental patterns based on the user's historical car rental data.

[0124] Historical car rental patterns refer to the patterns observed in users' past car rental data. The search method here involves the system analyzing historical car rental data to determine if there are cyclical patterns.

[0125] Step 831: If a historical car rental pattern exists, then based on the car rental demand and the historical car rental pattern, find the corresponding historical car models in the car model database as the final filtered car models and output them to the car model display interface.

[0126] Historical vehicle models refer to the final selection of models that have been rented in the past. The search method here is to obtain the models based on the intersection of the rental needs and the historical patterns of the models.

[0127] If a historical car rental pattern exists, it means that the user has been renting cars according to this pattern for a long time, and is likely to do so again this time. Therefore, based on the car rental demand and historical car rental pattern, the corresponding historical car models can be found in the car model database.

[0128] Step 832: If no historical car rental pattern exists, then based on the car rental needs and identity information, select the corresponding required car models from the car model database as the final selected car models and output them to the car model display interface.

[0129] The filtering method here has been introduced in step 82 and will not be repeated here.

[0130] This also includes a method for selecting the desired vehicle model from a vehicle model database based on historical rental data when no historical rental pattern exists. This method includes: Step 8320: If no historical car rental pattern exists, then obtain the user's repeat car rental data based on the historical car rental data.

[0131] Repeat rental data refers to data related to vehicles rented repeatedly by a user. This data is obtained by the system searching for vehicle models that have been rented more than once in the historical rental data.

[0132] If there is no historical pattern in car rentals, it means that the user rents a car based on their needs. In this case, the user's repeat car rental data is obtained based on the historical car rental data.

[0133] Step 8321: If duplicate car rental data exists, then filter the final selected car models from the car model database based on the car rental demand and duplicate car rental data, and output them to the car model display interface.

[0134] The filtering method here is for the system to find the final selected models that meet the requirements and have duplicate models in the vehicle model database.

[0135] If duplicate rental data exists, it means that the user is satisfied with the car model and may rent the same car model again. Therefore, the final car model is selected from the car model database based on the rental needs and duplicate rental data.

[0136] Step 8322: If duplicate car rental data does not exist, then filter the required car models and historical car rental data to obtain the models without historical data as the final filtered models and output them to the car model display interface.

[0137] The filtering method here is obtained by the system filtering historical models from the required models.

[0138] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for intelligent vehicle selection on a car rental platform, characterized in that, include: Step 1: In response to the car rental signal, obtain the user's input car rental request and identity information, including historical car rental data; Step 2: Based on the car rental requirements, filter out the corresponding car models from the preset car model database; Step 3: Determine the initial screening quantity based on the required vehicle models; Step 4: When the number of vehicles selected in the initial screening is less than the preset threshold for quick browsing, define the required vehicle models as the final selected vehicle models and output them. Step 5: When the number of initial screenings exceeds the threshold number for quick browsing, the corresponding model selection information is retrieved from the preset model association database based on the identity information; Step 6: Find the corresponding conversion requirements from the preset conversion database using the selection association information; Step 7: Based on the conversion requirements, select the corresponding conversion models from the vehicle model database; Step 8: Determine the comprehensive screening models based on the required vehicle type and the converted vehicle type; Step 9: The comprehensively selected vehicle models are used as the final selected vehicle models and output on the preset vehicle model display interface. The vehicle model display interface displays the basic vehicle information corresponding to the comprehensively selected vehicle models.

2. The intelligent vehicle selection method for a car rental platform according to claim 1, characterized in that, It also includes a method for updating the final selected vehicle models, which includes: Step 90: After the final selected vehicle model is output, in response to the vehicle viewing signal, the corresponding vehicle data is obtained and output to the preset vehicle data interface. The vehicle data interface displays detailed vehicle data, vehicle pictures from various angles, and vehicle rental plans. Step 91: When the user is in the vehicle data interface, obtain the corresponding vehicle data and vehicle viewing duration; Step 92: When the viewing time of the vehicle exceeds the preset viewing time, the vehicle data is defined as interest vehicle data; Step 93: Obtain the vehicle model selection based on the interest vehicle data; Step 94: Based on the interest-based vehicle selection, update the final selected vehicle models and output them to the vehicle model display interface.

3. The intelligent vehicle selection method for a car rental platform according to claim 2, characterized in that, It also includes a method for filtering vehicle models with a viewing duration of 0, the method comprising: Step 910: Obtain the model selection dwell time on the model display interface based on the model display interface; Step 911: When the selection dwell time exceeds the preset selection time or a preset refresh signal is received, a non-intersection filter model is obtained based on the required model and the converted model; Step 912: Output the non-intersection filtered models as the final filtered models to the model display interface; Step 913: Upon receiving the vehicle viewing signal, stop accumulating the selection dwell time and do not update the final filtered vehicle model.

4. The intelligent vehicle selection method for a car rental platform according to claim 3, characterized in that, The methods for obtaining the selection dwell time include: Step 9101: If the vehicle viewing time exceeds the preset basic loading time of the vehicle data interface, output the preset basic vehicle information to the vehicle data interface. Step 9102: Upon receiving the preset detailed data signal, output the preset vehicle detailed data to the vehicle data interface after a preset detailed loading time; Step 9103: Obtain the total loading time based on the basic loading time and the detailed loading time; Step 9104: If the vehicle viewing time exceeds the total loading time, stop accumulating the selection dwell time; Step 9105: If the vehicle viewing time does not exceed the total loading time, then the selection dwell time continues to accumulate.

5. The intelligent vehicle selection method for a car rental platform according to claim 4, characterized in that, The methods for obtaining the vehicle viewing duration include: Step 91050: If the user is in the vehicle data interface, the vehicle viewing time will be continuously accumulated; Step 91051: If the user is in the vehicle display interface, stop accumulating the vehicle viewing time and output it; Step 91052: When the user's vehicle viewing time on the vehicle data interface exceeds the total loading time, the vehicle viewing time is continuously accumulated based on the previous vehicle viewing time. Step 91053: If the user's vehicle viewing time on the vehicle data interface has not exceeded the total loading time, then the previous vehicle viewing time is output.

6. The intelligent vehicle selection method for a car rental platform according to claim 5, characterized in that, It also includes an optimization method for the final selected vehicle models, the method comprising: Step 910510: Upon receiving the vehicle data viewed by the user, obtain the vehicle type corresponding to the vehicle data; Step 910511: Sort the vehicle types according to the vehicle viewing duration corresponding to the vehicle type to obtain the vehicle model viewing duration sort. Step 910512: Based on the vehicle model duration sorting, sort the vehicle types on the vehicle model display interface and output them.

7. The intelligent vehicle selection method for a car rental platform according to claim 6, characterized in that, It also includes a method for optimizing the final filtered vehicle models based on the vehicle viewing duration, the method comprising: Step 9105110: If the vehicle viewing time exceeds the preset viewing time threshold, obtain the advantage data of the vehicle type; Step 9105111: Based on the advantage data, find the vehicle type in the corresponding vehicle type that has the advantage data and whose other preset key data are all better than the vehicle type, and then output it to the vehicle model display interface; Step 9105112: If the number of vehicle types exceeding the viewing time threshold exceeds 1, then the advantage data of the corresponding vehicle types are integrated to obtain comprehensive advantage data. Step 9105113: Update the final selected vehicle models based on the comprehensive advantage data and output the updated models.

8. The intelligent vehicle selection method for a car rental platform according to claim 1, characterized in that, It also includes a method for filtering car models based on the historical car rental data, the method comprising: Step 80: Obtain the similarity between the user's input of the car rental request and the historical car rental data; Step 81: If the similarity of the demand is greater than the preset similarity threshold, then based on the user's input car rental demand and the historical car rental data, obtain similar filter models as the final filter models and output them to the model display interface. Step 82: If the similarity of the demand is less than the similarity threshold, then the corresponding car model is selected from the car model database based on the car rental demand input by the user. Step 83: Based on the required vehicle type, filter the vehicle types in the historical car rental data to obtain vehicles with no history as the final filtered vehicle types and output them to the vehicle display interface.

9. The intelligent vehicle selection method for a car rental platform according to claim 8, characterized in that, The method for filtering the vehicle types in the historical car rental data based on the required vehicle type includes: Step 830: Based on the user's historical car rental data, find the corresponding historical car rental patterns; Step 831: If the historical car rental pattern exists, then based on the car rental demand and the historical car rental pattern, find the corresponding historical car model in the car model database and output it as the final filtered car model to the car model display interface; Step 832: If the historical car rental pattern does not exist, then based on the car rental demand and the identity information, the corresponding demanded car model is selected from the car model database and output as the final selected car model to the car model display interface.

10. The intelligent vehicle selection method for a car rental platform according to claim 9, characterized in that, It also includes a method for selecting the corresponding desired vehicle model from the vehicle model database based on the historical car rental data when the historical car rental pattern does not exist. This method includes: Step 8320: If the historical car rental pattern does not exist, then obtain the user's repeated car rental data based on the historical car rental data; Step 8321: If the duplicate car rental data exists, then the final selected car models are filtered from the car model database according to the car rental demand and the duplicate car rental data and output to the car model display interface; Step 8322: If the duplicate car rental data does not exist, then the car models without historical records are obtained by filtering based on the required car model and the historical car rental data and output as the final filtered car models to the car model display interface.

Images