KR20230006200A - Chatbot system for vehicle sales and maintenance and method of vehicle sales and maintenance service using chatbot system - Google Patents

Abstract

A chatbot system for vehicle sales and maintenance that can provide highly reliable chatbot services for each domain such as vehicle sales, purchase, and maintenance, and a chatbot-based vehicle sales and maintenance service method using the same are disclosed. The chatbot system for vehicle sales and maintenance, according to an embodiment of the present invention, comprises: a domain classification module that analyzes input conversation data received from a user terminal and classifies a domain of the input conversation data into one or more vehicle service domains among a plurality of vehicle-related domains comprising a vehicle sales domain, a vehicle purchase domain and a vehicle maintenance domain; and a chatbot module that vectorizes a text of the input conversation data to generate a user request vector and provide a Q&A service corresponding to the user request vector according to a scenario set for the vehicle service domain classified by the domain classification module.

Description

Chatbot system for vehicle sales and maintenance and chatbot-based vehicle sales and maintenance service method using the same

The present invention relates to a chatbot system for vehicle sales and maintenance, and more particularly, to a chatbot system specialized in a vehicle sales and maintenance domain and a chatbot-based vehicle sales and maintenance service method using the same.

In general, vehicle sales-related service systems or vehicle maintenance-related service systems are used to provide appropriate sales or maintenance services to users who want to sell or purchase vehicles or users who need maintenance (maintenance) of vehicles in use. . A user accesses a vehicle sales-related service system or a vehicle maintenance-related service system and receives a vehicle sales or maintenance service by making an inquiry or searching. However, conventional vehicle sales or maintenance-related services have limitations in providing a variety of information required when a user sells, purchases, or maintains a vehicle.

A chatbot is an interactive messenger system that provides answers (response) in the form of a conversation with a person in everyday language when a user inputs a question in the form of a chat (conversation) into a messenger. It is used for purposes such as customer support or information acquisition in various application fields such as institutions and educational institutions. Utilizing these chatbots can provide a lot of help to users who want to receive vehicle sales or maintenance-related services.

However, most of the conventional chatbot solutions register and operate Q&A, and there is a limit to appropriately responding to various scenarios that require detailed answers in relation to vehicle-related domains, such as vehicle sales or maintenance, in addition to general customer inquiries. . In addition, due to the nature of vehicle-related services, there are cases where a user needs to make a pre-suggestion before making a vehicle maintenance-related inquiry or before a problem occurs with the vehicle.

The present invention is to provide a chatbot system for vehicle sales and maintenance capable of providing highly reliable chatbot services for each domain, such as vehicle sales, purchase, and maintenance, and a chatbot-based vehicle sales and maintenance service method using the same.

In addition, the present invention is to provide a chatbot system for vehicle sales and maintenance that allows users to make inquiries about vehicle maintenance or to make suggestions to users before problems occur in the vehicle, and a chatbot-based vehicle sales and maintenance service method using the same. .

In addition, the present invention provides a chatbot system for vehicle sales and maintenance that pre-proposes appropriate recommended maintenance items by analyzing the user's driving habits and safe driving scores, and a chatbot-based vehicle sales and maintenance service method using the same will be.

A chatbot system for vehicle sales and maintenance according to an embodiment of the present invention: analyzes input dialog data received from a user terminal and sets the domain of the input dialog data to a vehicle sales domain, a vehicle purchase domain, and a vehicle maintenance domain. a domain classification module configured to classify into one or more vehicle service domains among a plurality of vehicle-related domains; and vectorizing text of the input dialogue data to generate a user request vector, and providing a query response service conforming to the user request vector according to a scenario set for the vehicle service domain classified by the domain classification module. Chatbot module; includes.

The chatbot system for vehicle sales and maintenance according to an embodiment of the present invention may further include: a vehicle database for storing various vehicle sales data and maintenance information for each vehicle type.

The chatbot module: When the input conversation data corresponds to the vehicle sales domain, provides a question and answer service related to vehicle sales according to a set vehicle sales scenario, and corresponds to a specification related to the user's vehicle for sale in the vehicle database. a vehicle sales chatbot module configured to search for a similar vehicle to be sold and provide a price of a vehicle with a similar specification and an estimated sales price of the vehicle for sale; When the input conversation data corresponds to the vehicle purchase domain, a question answering service related to vehicle purchase is provided according to a set vehicle purchase scenario, and a similarity between a vehicle for sale stored in the vehicle database and the user's vehicle purchase requirements is determined. a vehicle purchase chatbot module configured to analyze and search for and recommend a vehicle for sale that meets the vehicle purchase requirements; and when the input conversation data corresponds to the vehicle maintenance domain, a question and answer service related to vehicle maintenance is provided according to a set vehicle maintenance scenario, and an inspection history, accumulated driving mileage, and driving habit data of the user's vehicle are provided. and a vehicle maintenance chatbot module configured to analyze and recommend recommended maintenance items based on the maintenance information for each vehicle type.

The vehicle sales chatbot module: compares the user request vector with a set vehicle sales vector, and requests sales vehicle information for a missing item among items including vehicle model, year, and mileage to the user terminal; generating a sales request vector by converting sales vehicle information input from the user terminal into text vectors; generating vehicle sales request data by mapping the user request vector and the sales request vector to a hash value set for the vehicle sales domain; comparing the vehicle sales request data with vehicle listing data stored in the vehicle database to output listing prices of similar vehicles in order of similarity; analyzing a difference item having a difference between the sale vehicle information and the similar vehicle for sale; And it may be configured to correct the sale price according to the similarity and the difference item to predict and output the expected selling price.

The vehicle purchase chatbot module: compares the user request vector with a set vehicle purchase vector, and requests purchase vehicle information for a missing item among items including vehicle model, model year, mileage, and budget to the user terminal; generating a purchase request vector by converting purchase vehicle information input from the user terminal into text vectors; generating vehicle purchase request data by mapping the user request vector and the purchase request vector to a hash value set for the vehicle purchase domain; comparing the vehicle purchase request data with vehicle listing data stored in the vehicle database to determine a plurality of recommended products in order of relevance; correcting the selling price of the recommended product according to the vehicle purchase request data to calculate a converted selling price; And it may be configured to sort the plurality of recommended products in the order of converted sales price, list them, and output them.

The vehicle maintenance chatbot module: collects vehicle driving-related data from an on-board vehicle diagnosis system installed in the vehicle of the user; generating a driving habit vector of the user by analyzing the driving-related data; comparing the user request vector and the driving habit vector with a set vehicle maintenance vector and requesting vehicle maintenance information for a missing item among items including vehicle information, inspection history, and mileage from the user terminal; generating a maintenance request vector by converting vehicle maintenance information received from the user terminal into text vectors; generating and storing vehicle maintenance request data in a user database by mapping the user request vector, the driving habit vector, and the maintenance request vector to a hash value set for the vehicle maintenance domain; comparing the vehicle maintenance request data with maintenance information for each vehicle type stored in the vehicle database, and recommending the recommended maintenance item and maintenance time point; And it may be configured to send a maintenance recommendation notification to the user terminal in advance before the recommended maintenance time point.

A chatbot-based vehicle sales and maintenance service method according to an embodiment of the present invention includes: (a) a domain classification module analyzes input dialog data received from a user terminal, and sets the domain of the input dialog data to a vehicle sales domain and a vehicle classifying into one or more vehicle service domains among a plurality of vehicle-related domains including a purchasing domain and a vehicle maintenance domain; and (b) by a chatbot module, text of the input conversation data is vectorized to generate a user request vector, and the user request vector is matched according to a scenario set for the vehicle service domain classified by the domain classification module. It includes; providing a question answering service.

The step (b) includes: (b-1) providing, by the vehicle sales chatbot module, a question answering service related to vehicle sales according to a set vehicle sales scenario when the input conversation data corresponds to the vehicle sales domain; Searching for a similar vehicle corresponding to a specification related to the user's sales vehicle in a vehicle database storing various vehicle listing data and maintenance information for each vehicle type, and providing a price of a vehicle with similar specifications and an estimated sales price of the sales vehicle; (b-2) When the input conversation data corresponds to the vehicle purchase domain, the vehicle purchase chatbot module provides a question-and-answer service related to vehicle purchase according to a set vehicle purchase scenario, and vehicle listings stored in the vehicle database and analyzing the degree of similarity between the user's vehicle purchase requirements and searching for and recommending vehicles that meet the vehicle purchase requirements; and (b-3) by the car maintenance chatbot module, if the input conversation data corresponds to the car maintenance domain, a car maintenance question and answer service is provided according to a set car maintenance scenario, and the user's own vehicle and recommending recommended maintenance items based on the maintenance information for each vehicle type by analyzing inspection history, accumulated driving mileage, and driving habit data.

The step (b-1) may include: comparing the user request vector with a set vehicle sales vector and requesting, from among items including vehicle model, year, and mileage, vehicle information for sale for an omitted item, to the user terminal. ; generating a sales request vector by converting vehicle information for sale received from the user terminal into text vectors; generating vehicle sales request data by mapping the user request vector and the sales request vector to a hash value set for the vehicle sales domain; comparing the vehicle sale request data with vehicle listing data stored in the vehicle database and outputting listing prices of similar vehicles in order of similarity; analyzing a difference item having a difference between the sale vehicle information and the similar vehicle for sale; and correcting the sale price according to the similarity and the difference item to predict and output an expected selling price.

Step (b-2): Comparing the user request vector with a set vehicle purchase vector and requesting purchase vehicle information for an omitted item among items including vehicle model, model year, mileage, and budget to the user terminal. step; generating a purchase request vector by converting purchase vehicle information input from the user terminal into text vectors; generating vehicle purchase request data by mapping the user request vector and the purchase request vector to a hash value set for the vehicle purchase domain; Comparing the vehicle purchase request data with vehicle sale data stored in the vehicle database to determine a plurality of recommended products in order of relevance; calculating; and arranging the plurality of recommended products in order of converted sales price, listing them, and outputting them.

The step (b-3) may include: collecting vehicle driving-related data from an on-board vehicle diagnosis system installed in the vehicle of the user; generating a driving habit vector of the user by analyzing the driving-related data; comparing the user request vector and the driving habit vector with a set vehicle maintenance vector and requesting vehicle maintenance information for a missing item among items including vehicle information, inspection history, and mileage to the user terminal; generating a maintenance request vector by converting vehicle maintenance information received from the user terminal into text vectors; generating vehicle maintenance request data by mapping the user request vector, the driving habit vector, and the maintenance request vector to a hash value set for the vehicle maintenance domain, and storing the generated vehicle maintenance request data in a user database; comparing the vehicle maintenance request data with maintenance information for each vehicle type stored in the vehicle database and recommending the recommended maintenance item and maintenance time point; and sending a maintenance recommendation notification to the user terminal before a recommended maintenance time point.

According to an embodiment of the present invention, a computer program recorded on a computer-readable recording medium is provided to execute the chatbot-based vehicle sales and maintenance service method.

According to an embodiment of the present invention, a chatbot system for vehicle sales and maintenance capable of providing highly reliable chatbot services for each domain such as vehicle sales, purchase, and maintenance, and a chatbot-based vehicle sales and maintenance service method using the same are provided do.

In addition, according to an embodiment of the present invention, it is possible to make suggestions to the user before the user inquires about vehicle maintenance or a problem occurs with the vehicle, and recommends appropriate maintenance items by analyzing the user's driving habits and safe driving scores. line can be suggested.

1 is a configuration diagram of a chatbot system for vehicle sales and maintenance according to an embodiment of the present invention.
2 is a flowchart of a chatbot-based vehicle sales and maintenance service method according to an embodiment of the present invention.
FIG. 3 is a flowchart illustrating step S130 of FIG. 2 .
4 is a flowchart illustrating step S140 of FIG. 2 .
5 is a flowchart illustrating step S150 of FIG. 2 .

Advantages and features of the present invention, and methods of achieving them, will become clear with reference to the detailed description of the following embodiments taken in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only these embodiments make the disclosure of the present invention complete, and common knowledge in the art to which the present invention belongs. It is provided to completely inform the person who has the scope of the invention, and the present invention is only defined by the scope of the claims. Like reference numbers designate like elements throughout the specification. For components well known in the art related to the present invention, detailed descriptions may be omitted so as not to obscure the subject matter of the present invention.

In the present specification, when a part 'includes' a certain component, it means that it may further include other components without excluding other components unless otherwise stated. As used herein, '~ module, ~ unit, ~ system' is a unit that processes at least one function or operation, and may mean, for example, software, an FPGA, or a hardware component. Functions provided by '~module, ~unit, ~system' may be performed separately by a plurality of components or may be integrated with other additional components. The '~module, ~unit, ~system' in this specification is not necessarily limited to software or hardware, and may be configured to be in an addressable storage medium or configured to reproduce one or more processors. Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

1 is a configuration diagram of a chatbot system for vehicle sales and maintenance according to an embodiment of the present invention. Referring to FIG. 1 , a chatbot system 100 for vehicle sales and maintenance according to an embodiment of the present invention is to provide services related to vehicle sales and maintenance to users based on a chatbot, and an input unit 110 ), domain classification module 120, chatbot module 130, output unit 140, vehicle database (DB; database) 150, user database 160, mileage calculation module 170, safe driving score calculation A module 180 and a hash processing unit 190 may be included.

The input unit 110 may receive input conversation data in a conversation form from the user terminal 10 through an input interface. The user terminal 10 is a terminal of a user requesting service provision such as vehicle sales and maintenance, for example, a portable terminal such as a mobile phone, tablet, laptop, or smart terminal having a wireless communication function, or a wired/wireless communication function It may be provided with a desktop computer or the like having, but is not limited to these examples.

The input conversation data input by the input unit 110 is transferred to the domain classification module 120 for classification of vehicle service domains. In order to provide an effective question-answering service according to domain-specific scenarios of the input dialogue data, the domain classification module 120 classifies the domain of the input dialogue data into a plurality of domains including a vehicle sales domain, a vehicle purchase domain, and a vehicle maintenance domain. It can be classified into one or more vehicle service domains among two vehicle-related domains.

In an embodiment, the domain classification module 120 compares the text extracted from the input conversation data with keywords set for each domain, and classifies the text into one or a plurality of domains according to similarities. For example, when a user desires to purchase a new vehicle while selling a vehicle owned by the user, input conversation data may be classified into those corresponding to a vehicle sales domain and a vehicle purchase domain. In this case, a process of distinguishing between text related to the vehicle selling domain and text related to the vehicle purchasing domain may be additionally performed in the input dialogue data.

The chatbot module 130 analyzes the meaning of natural language, which is a language used in daily life, so that the computer can process it, and the text of the input conversation data can be understood by the computer through natural language processing. A user-requested vector may be generated by performing text vectorization processing in which a binary number of the form is converted into digits, etc. In this case, text vectorization methods include, for example, one-hot encoding, word embedding, and word2vec, but are not limited thereto.

To provide a chatbot service specialized for each domain, the user request vector generated by the chatbot module 130 may be different according to the vehicle service domain. The chatbot module 130 provides a query response service that meets the user request vector according to scenarios set for the vehicle service domains (vehicle sales domain, vehicle purchase domain, vehicle maintenance domain, etc.) classified by the domain classification module 120 can do.

In an embodiment, the chatbot module 130 may include a car sales chatbot module 132 , a car purchase chatbot module 134 , and a car maintenance chatbot module 136 . When it is determined by the domain classification module 120 that the input conversation data corresponds to the vehicle sales domain, the vehicle sales chatbot module 132 may provide a vehicle sales question answering service according to a set vehicle sales scenario.

The vehicle sales chatbot module 132 searches the vehicle database 150 for similar vehicles corresponding to specifications related to the vehicle the user intends to sell, and determines the sales price of the vehicle with similar specifications, the average selling price of a plurality of vehicles with similar specifications, and You can provide an estimated selling price for the vehicle you intend to sell. At this time, the vehicle sales chatbot module 132 may analyze the exterior/interior picture of the vehicle transmitted from the user terminal 10 to determine the vehicle management state, and reflect the result to the estimated sales price of the vehicle.

Through the user terminal 10, the user inputs a simple message such as "I want to sell a vehicle" together with vehicle information such as the model, year, and mileage of the vehicle he/she wants to sell, and takes pictures of the exterior and interior of the vehicle. It is possible to grasp the appropriate selling price of one's own vehicle (used car) only by the act of transmitting the information to the chatbot module 130. Therefore, even if the user does not know the exact price of the used car according to the model, year, and mileage of the used car he/she wants to sell, the vehicle sales chatbot module 132 checks the price of a car with similar specifications and the expected sales price to determine the appropriate price of the car to be sold. You can determine the selling price.

When it is determined by the domain classification module 120 that the input conversation data corresponds to the vehicle purchase domain, the vehicle purchase chatbot module 134 may provide a vehicle purchase question answering service according to a set vehicle purchase scenario. The vehicle purchase chatbot module 134 analyzes the similarity (eg, cosine similarity) between the vehicle listing stored in the vehicle database 150 and the user's vehicle purchase requirements, and obtains a vehicle that meets the vehicle purchase requirements. You can search for properties and recommend them. If the user agrees to receive the notification, the vehicle purchase chatbot module 134 may periodically transmit a notification to the user terminal 10 when a vehicle for sale having a high similarity is newly registered.

Through the user terminal 10, the user may request vehicle requirements such as the model, year, mileage, budget, used car/new car/foreign car of the vehicle he/she intends to purchase, and user requirements (eg, gender, age, marital status). , number of children, age of children, dependents, disabled people, etc.) together with various vehicle purchase-related information (requirements for purchasing a vehicle), along with recommendation information on vehicle sales simply by entering a simple message such as “I want to purchase a vehicle” can be provided.

User requirements are used to determine the recommendation ranking of vehicles for sale by assigning points (weights) according to the vehicle types that meet the user's requirements when the user does not limit the vehicle type or describes various vehicle types, For example, when there are many used cars for sale of different colors corresponding to the same car model, it can be used to recommend a car for sale in consideration of the color of the car.

When it is determined by the domain classification module 120 that the user's input conversation data corresponds to the vehicle maintenance domain, the vehicle maintenance chatbot module 136 may provide a vehicle maintenance question and answer service according to a set vehicle maintenance scenario. there is. The vehicle maintenance chatbot module 136 analyzes the user's vehicle type, year, inspection history including the last inspection date, cumulative driving mileage, and driving habit data to provide recommended maintenance items (recommended maintenance) based on maintenance information for each vehicle type. list) can be recommended.

The vehicle maintenance chatbot module 136 may perform a combination of a passive function of providing recommended maintenance items upon a user's request and an active function of actively providing recommended maintenance items even without a separate request from the user. When driving a vehicle, the user terminal 10 linked with the OBD can transmit GPS information to the server of the chatbot system, and based on this, the vehicle mileage and safe driving score can be calculated and stored in the user database 160. The safe driving score may be exemplarily calculated according to the conditions shown in Table 1 below.

driving mode condition safe driving If the following 3 conditions are not met Speeding When exceeding the speed limit standard ±10 km/h sudden acceleration For speed changes of more than +10 km/h in 1 second rapid deceleration per second - for speed changes of 10 km/h or less

The vehicle type, year, inspection history, accumulated driving mileage, and driving habit data of the vehicle owned by the user may be obtained from data input by the user through the user terminal 10, and are retained by the user to collect highly reliable safe driving scores. It may also be obtained through an on-board vehicle diagnostic system 20 such as OBDII (On-board diagnostics II) installed in the vehicle. The vehicle maintenance chatbot module 136 sends maintenance recommendation notifications to the user terminal 10 in accordance with the maintenance time before the actual maintenance demand occurs based on the recommended maintenance items and the time required for inspection of each recommended maintenance item. A pre-talk function that preemptively proposes maintenance items can be provided. For example, when conditions related to vehicle maintenance are satisfied, such as one month before the engine oil replacement period or when brake pads need to be replaced, a maintenance recommendation and suggestion notification may be sent to the user terminal 10 .

The output unit 140 outputs the expected sales price of the sales vehicle provided by the vehicle sales chatbot module 132, the sales price of a vehicle having similar specifications to the sales vehicle, and the recommended vehicle sale and vehicle sales provided by the vehicle purchase chatbot module 134. Information related to vehicle sales and maintenance, such as recommended maintenance items (list) provided by the maintenance chatbot module 136, may be output (transmitted) to the user terminal 10 through a wired/wireless communication interface.

The vehicle database 150 may store various vehicle listing data and maintenance information for each vehicle type. Vehicle listing data may include information such as model, year, cumulative mileage, purchase price, exterior management status, and interior management status of used cars for sale by various users. The maintenance information for each vehicle type may include information on a maintenance cycle for various maintenance items according to the vehicle type.

The user database 160 may store information such as a user's mileage, driving region, driving habit, and safe driving score. Information such as the user's mileage, driving area, driving habits, and safe driving score is based on driving-related information (GPS data, speed/acceleration profile, etc.) collected from the on-board vehicle diagnosis system 20 installed in the user's vehicle. can be obtained with

The mileage calculation module 170 may analyze the driving-related information collected from the on-board vehicle diagnosis system 20 to calculate the user's vehicle mileage (accumulated mileage). The safe driving score calculation module 180 may calculate the user's safe driving score by analyzing the driving-related information collected from the on-board vehicle diagnosis system 20 . The user's driving area can be analyzed based on GPS data, and depending on the characteristics of the driving area (city, suburbs, highway, unpaved road, etc.), the vehicle's maintenance item inspection cycle can be determined, or the sales price of a used car can be estimated. may be additionally reflected.

The hash processing unit 190 hash-processes various driving-related information of the user collected by the mileage calculation module 170 and the safe driving score calculation module 180 using a hash function, and stores the information in the user database 160. can be saved Accordingly, it is possible to prevent arbitrary modification of input and collected data, and secure data reliability by mapping to hash values.

The hash function may be a function that maps data of an arbitrary length to data of a fixed length. By storing the value obtained by the hash function in the user database 160 according to the one-way encryption technique without storing the user's driving-related information as it is, data integrity verification is possible because data cannot be tampered with, and information leakage is ensured by security. can prevent

As described above, the chatbot system for vehicle sales and maintenance according to an embodiment of the present invention can provide highly reliable chatbot services for each domain such as vehicle sales, purchase, and maintenance, and is applied regardless of the type or scale of the platform. this is possible In addition, it is possible to make preemptive suggestions according to the user's driving habits, safe driving score, vehicle information, maintenance-related conditions, etc., rather than simply responding to inquiries, so it can be used appropriately for vehicle maintenance management that requires tracking of driving mileage and period. there is.

2 is a flowchart of a chatbot-based vehicle sales and maintenance service method according to an embodiment of the present invention. Referring to FIGS. 1 and 2 , the domain classification module 120 classifies the domain of input conversation data into one or more vehicle service domains among a plurality of vehicle-related domains including a vehicle sales domain, a vehicle purchase domain, and a vehicle maintenance domain. It can (S110).

The chatbot module 130 vectorizes the text of the input conversation data to generate a user request vector, and for the vehicle service domain (vehicle sales domain, vehicle purchase domain, vehicle maintenance domain, etc.) classified by the domain classification module 120 According to the set scenario, a query response service corresponding to the user's request vector may be provided (S120).

If the input conversation data corresponds to the vehicle sales domain, the vehicle sales chatbot module 132 may provide a vehicle sales question and answer service according to the vehicle sales scenario set (S130). The vehicle sales chatbot module 132 may search for similar vehicles corresponding to specifications related to the user's sales vehicle in the vehicle database and provide a price of a vehicle with similar specifications and an estimated sales price of the vehicle for sale.

When the input conversation data corresponds to the vehicle purchase domain, the vehicle purchase chatbot module 134 may provide a vehicle purchase question answering service according to a set vehicle purchase scenario (S140). The vehicle purchase chatbot module 134 may search for and recommend a vehicle for sale that meets the vehicle purchase requirements by analyzing the similarity between the vehicle products stored in the vehicle database 150 and the user's vehicle purchase requirements.

When the input conversation data corresponds to the vehicle maintenance domain, the vehicle maintenance chatbot module 136 may provide a vehicle maintenance question and answer service according to a set vehicle maintenance scenario (S150). The vehicle maintenance chatbot module 136 may recommend recommended maintenance items based on maintenance information for each vehicle type by analyzing the user's vehicle inspection history, accumulated driving mileage, and driving habit data.

FIG. 3 is a flowchart illustrating step S130 of FIG. 2 . 1 to 3 , the vehicle sales chatbot module 132 compares a user request vector with a set vehicle sales vector and obtains sales vehicle information for the missing item among items including vehicle model, year, and mileage. A request can be made to the user terminal 10 (S131). For example, the vehicle sales chatbot module 132 may request input of information on the missing item in the form of "Please tell me the model of the vehicle to be sold."

The vehicle sales chatbot module 132 may generate a sales request vector by converting vehicle sales information input from the user terminal into text vectors (S132). The vehicle sales chatbot module 132 may generate vehicle sales request data by mapping the user request vector and the sales request vector to a hash value set for the vehicle sales domain (S133).

The vehicle sales chatbot module 132 may compare the vehicle sales request data with the vehicle listing data stored in the vehicle database 150 and output the listing prices of similar vehicles in order of similarity (S134). The vehicle sales chatbot module 132 may analyze a difference item having a difference between sales vehicle information and a similar vehicle for sale (S135). The vehicle sales chatbot module 132 may correct the sale price according to the similarity and difference items to predict and output the expected selling price (S136). For example, the vehicle sales chatbot module 132 predicts the sales price of a vehicle for sale based on a multiple regression analysis model between variables related to the vehicle (vehicle model, year, mileage, etc.) and the vehicle sales price, or for similarity and difference items. The selling price of the sales vehicle may be predicted by reflecting the set correction value.

4 is a flowchart illustrating step S140 of FIG. 2 . 1, 2, and 4, the vehicle purchase chatbot module 134 compares the user request vector with the set vehicle purchase vector and determines the missing item among items including vehicle model, model year, mileage and budget. Purchase vehicle information may be requested from the user terminal 10 (S141). The vehicle purchase chatbot module 134 may generate a purchase request vector by converting vehicle purchase information received from the user terminal into text vectors (S142).

The vehicle purchase chatbot module 134 may generate vehicle purchase request data by mapping the user request vector and the purchase request vector to a hash value set for the vehicle purchase domain (S143). The vehicle purchase chatbot module 134 may compare the vehicle purchase request data with the vehicle listing data stored in the vehicle database 150 to determine a plurality of recommended products in order of relevance (S144).

In an embodiment, the vehicle purchase chatbot module 134 may correct the selling price of the recommended product according to the vehicle purchase request data to calculate the converted sales price (S145). The vehicle purchase chatbot module 134 may list and output a plurality of recommended products by arranging them in the order of converted sales price (S146). Therefore, it is possible to recommend a more suitable vehicle for sale to the user by sorting and outputting the recommended items in the order of selling price converted based on the user's requirements, rather than simply sorting in the order of price of the recommended items.

5 is a flowchart illustrating step S150 of FIG. 2 . Referring to FIGS. 1, 2, and 5 , the vehicle maintenance chatbot module 136 may collect vehicle driving-related data from the on-board vehicle diagnosis system installed in the user's vehicle (S151). The vehicle maintenance chatbot module 136 may generate a driving habit vector of the user by analyzing vehicle driving-related data (S152).

The vehicle maintenance chatbot module 136 compares the user request vector and the driving habit vector with the set vehicle maintenance vector, and provides vehicle maintenance information for missing items among items including vehicle information, inspection history, and mileage to the user terminal 10 ) can be requested (S153). The car maintenance chatbot module 136 may generate a maintenance request vector by converting the car maintenance information input from the user terminal 10 into text vectors (S154).

The car maintenance chatbot module 136 may generate car maintenance request data by mapping the user request vector, the driving habit vector, and the maintenance request vector to a hash value set for the car maintenance domain, and store it in the user database 160 (S155). . The vehicle maintenance chatbot module 136 may compare the vehicle maintenance request data with maintenance information for each vehicle type stored in the vehicle database 150 to recommend recommended maintenance items and maintenance timing (S156). The car maintenance chatbot module 136 may send a maintenance recommendation notification to the user terminal 10 in advance before the recommended maintenance time (S157).

As described above, the chatbot-based vehicle sales and maintenance service method according to an embodiment of the present invention can provide users with highly reliable chatbot services for each domain, such as vehicle sales, purchases, and maintenance, regardless of the type or scale of the platform. , It is possible to make preemptive suggestions according to the user's driving habits, safe driving score, vehicle information, maintenance-related conditions, etc., rather than simply responding to inquiries, so it can be used appropriately for vehicle maintenance management that requires tracking of mileage and period. .

The embodiments described above may be implemented as hardware components, software components, and/or a combination of hardware components and software components. For example, the devices, methods and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate (FPGA) array), programmable logic units (PLUs), microprocessors, or any other device capable of executing and responding to instructions.

A processing device may run an operating system and one or more software applications running on the operating system. A processing device may also access, store, manipulate, process, and generate data in response to execution of software. For convenience of understanding, there are cases in which one processing device is used, but those skilled in the art know that a processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It will be understood that it can include

For example, a processing device may include a plurality of processors or a processor and a controller. Also, other processing configurations are possible, such as a parallel processor. Software may include a computer program, code, instructions, or a combination of one or more of the foregoing, which configures a processing device to operate as desired or processes independently or collectively. You can command the device.

Software and/or data may be any tangible machine, component, physical device, virtual equipment, computer storage medium or device, intended to be interpreted by or provide instructions or data to a processing device. , or may be permanently or temporarily embodied in a transmitted signal wave. Software may be distributed on networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer readable media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer readable medium. Computer readable media may include program instructions, data files, data structures, etc. alone or in combination. Program commands recorded on the medium may be specially designed and configured for the embodiment or may be known and usable to those skilled in computer software.

Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CDROMs and DVDs, and ROMs, RAMs, and flash memories. hardware devices specially configured to store and execute program instructions, such as; Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter, as well as machine language codes such as those produced by a compiler. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

As described above, although the embodiments have been described with limited examples and drawings, those skilled in the art can make various modifications and variations from the above description. For example, the described techniques may be performed in an order different from the method described, and/or components of the described system, structure, device, circuit, etc. may be combined or combined in a different form than the method described, or other components may be used. Or even if it is replaced or substituted by equivalents, appropriate results can be achieved. Therefore, other implementations, other embodiments, and equivalents of the claims are within the scope of the following claims.

100: Chatbot system for vehicle sales and maintenance
110: input unit
120: domain classification module
130: chatbot module
132: vehicle sales chatbot module
134: vehicle purchase chatbot module
136: vehicle maintenance chatbot module
140: output unit
150: vehicle database
160: user database
170: mileage calculation module
180: safe driving score calculation module
190: hash processing unit

Claims (11)

A chatbot system for vehicle sales and maintenance to provide services related to vehicle sales and maintenance based on chatbots,
Domain classification configured to analyze input dialogue data received from a user terminal and classify a domain of the input dialogue data into one or more vehicle service domains among a plurality of vehicle-related domains including a vehicle sales domain, a vehicle purchase domain, and a vehicle maintenance domain. module; and
A chatbot configured to vectorize text of the input conversation data to generate a user request vector, and to provide a query response service conforming to the user request vector according to a scenario set for the vehicle service domain classified by the domain classification module. Chatbot system for vehicle sales and maintenance, including a module; According to claim 1,
Further comprising a vehicle database for storing various vehicle listing data and maintenance information for each vehicle type;
The chatbot module is:
If the input conversation data corresponds to the vehicle sales domain, a question and answer service related to vehicle sales is provided according to a set vehicle sales scenario, and a similar vehicle corresponding to specifications related to the user's vehicle for sale is searched in the vehicle database. a vehicle sales chatbot module configured to provide a similar specification vehicle price and an estimated sales price of the vehicle for sale;
When the input conversation data corresponds to the vehicle purchase domain, a question answering service related to vehicle purchase is provided according to a set vehicle purchase scenario, and a similarity between a vehicle for sale stored in the vehicle database and the user's vehicle purchase requirements is determined. a vehicle purchase chatbot module configured to analyze and search for and recommend a vehicle for sale that meets the vehicle purchase requirements; and
When the input conversation data corresponds to the vehicle maintenance domain, a question and answer service related to vehicle maintenance is provided according to a set vehicle maintenance scenario, and the inspection history, accumulated driving mileage, and driving habit data of the user's vehicle are analyzed. A chatbot system for vehicle sales and maintenance comprising a vehicle maintenance chatbot module configured to recommend a recommended maintenance item based on the maintenance information for each vehicle type. According to claim 2,
The vehicle sales chatbot module:
comparing the user-requested vector with a set vehicle sales vector and requesting sales vehicle information for a missing item among items including vehicle model, model year, and mileage from the user terminal;
generating a sales request vector by converting sales vehicle information input from the user terminal into text vectors;
generating vehicle sales request data by mapping the user request vector and the sales request vector to a hash value set for the vehicle sales domain;
comparing the vehicle sales request data with vehicle listing data stored in the vehicle database to output listing prices of similar vehicles in order of similarity;
analyzing a difference item having a difference between the sale vehicle information and the similar vehicle for sale; And
A chatbot system for vehicle sales and maintenance configured to predict and output an estimated selling price by correcting the sale price according to the similarity and the difference item. According to claim 2,
The vehicle purchase chatbot module:
comparing the user request vector with a set vehicle purchase vector and requesting purchase vehicle information for a missing item among items including vehicle model, model year, mileage, and budget to the user terminal;
generating a purchase request vector by converting purchase vehicle information input from the user terminal into text vectors;
generating vehicle purchase request data by mapping the user request vector and the purchase request vector to a hash value set for the vehicle purchase domain;
comparing the vehicle purchase request data with vehicle listing data stored in the vehicle database to determine a plurality of recommended products in order of relevance;
correcting the selling price of the recommended product according to the vehicle purchase request data to calculate a converted selling price; And
A chatbot system for vehicle sales and maintenance configured to list and output the plurality of recommended products in order of converted sales price. According to claim 2,
The vehicle maintenance chatbot module:
collecting vehicle driving-related data from an on-board vehicle diagnostic system installed in the vehicle of the user;
generating a driving habit vector of the user by analyzing the driving-related data;
comparing the user request vector and the driving habit vector with a set vehicle maintenance vector and requesting vehicle maintenance information for a missing item among items including vehicle information, inspection history, and mileage from the user terminal;
generating a maintenance request vector by converting vehicle maintenance information received from the user terminal into text vectors;
generating and storing vehicle maintenance request data in a user database by mapping the user request vector, the driving habit vector, and the maintenance request vector to a hash value set for the vehicle maintenance domain;
comparing the vehicle maintenance request data with maintenance information for each vehicle type stored in the vehicle database, and recommending the recommended maintenance item and maintenance time point; And
A chatbot system for vehicle sales and maintenance configured to send a maintenance recommendation notification to the user terminal in advance before a recommended maintenance time. As a chatbot-based vehicle sales and maintenance service method that provides services related to vehicle sales and maintenance based on chatbots,
(a) The domain classification module analyzes the input dialogue data received from the user terminal and determines the domain of the input dialogue data as one or more vehicles among a plurality of vehicle-related domains including a vehicle sales domain, a vehicle purchase domain, and a vehicle maintenance domain. classifying into service domains; and
(b) a chatbot module vectorizes the text of the input dialogue data to generate a user request vector, and a query conforming to the user request vector according to a scenario set for the vehicle service domain classified by the domain classification module; Providing a response service; chatbot-based vehicle sales and maintenance service method including. According to claim 6,
The step (b) is:
(b-1) When the input conversation data corresponds to the vehicle sales domain, the vehicle sales chatbot module provides a question and answer service related to vehicle sales according to a set vehicle sales scenario, and various vehicle sales data and vehicle types searching for a similar vehicle corresponding to a specification related to the user's sales vehicle in a vehicle database storing maintenance information, and providing a similar specification vehicle price and an estimated sales price of the sales vehicle;
(b-2) When the input conversation data corresponds to the vehicle purchase domain, the vehicle purchase chatbot module provides a question-and-answer service related to vehicle purchase according to a set vehicle purchase scenario, and vehicle listings stored in the vehicle database and analyzing the degree of similarity between the user's vehicle purchase requirements and searching for and recommending vehicles that meet the vehicle purchase requirements; and
(b-3) When the input conversation data corresponds to the vehicle maintenance domain, the vehicle maintenance chatbot module provides a question and answer service related to vehicle maintenance according to a set vehicle maintenance scenario, and checks the vehicle owned by the user A chatbot-based vehicle sales and maintenance service method comprising analyzing history, accumulated driving mileage, and driving habit data and recommending recommended maintenance items based on the maintenance information for each vehicle type. According to claim 7,
The step (b-1) is:
comparing the user-requested vector with a set vehicle sales vector and requesting, from the user terminal, vehicle information for sale for an omitted item among items including vehicle model, model year, and mileage;
generating a sales request vector by converting vehicle information for sale received from the user terminal into text vectors;
generating vehicle sales request data by mapping the user request vector and the sales request vector to a hash value set for the vehicle sales domain;
comparing the vehicle sale request data with vehicle listing data stored in the vehicle database and outputting listing prices of similar vehicles in order of similarity;
analyzing a difference item having a difference between the sale vehicle information and the similar vehicle for sale; and
and estimating and outputting an estimated selling price by correcting the sale price according to the similarity and the difference item. According to claim 7,
The step (b-2) is:
comparing the user request vector with a set vehicle purchase vector and requesting purchase vehicle information for a missing item among items including vehicle model, model year, mileage, and budget to the user terminal;
generating a purchase request vector by converting purchase vehicle information input from the user terminal into text vectors;
generating vehicle purchase request data by mapping the user request vector and the purchase request vector to a hash value set for the vehicle purchase domain;
comparing the vehicle purchase request data with vehicle listing data stored in the vehicle database to determine a plurality of recommended products in order of relevance;
correcting a selling price of the recommended item according to the vehicle purchase request data and calculating a converted selling price; and
A chatbot-based vehicle sales and maintenance service method comprising the step of sorting the plurality of recommended products in order of converted sales price, listing them, and outputting them. According to claim 7,
The step (b-3) is:
collecting vehicle driving-related data from an on-board vehicle diagnostic system installed in the vehicle of the user;
generating a driving habit vector of the user by analyzing the driving-related data;
comparing the user request vector and the driving habit vector with a set vehicle maintenance vector and requesting vehicle maintenance information for a missing item among items including vehicle information, inspection history, and mileage to the user terminal;
generating a maintenance request vector by converting vehicle maintenance information received from the user terminal into text vectors;
generating vehicle maintenance request data by mapping the user request vector, the driving habit vector, and the maintenance request vector to a hash value set for the vehicle maintenance domain, and storing the generated vehicle maintenance request data in a user database;
comparing the vehicle maintenance request data with maintenance information for each vehicle type stored in the vehicle database and recommending the recommended maintenance item and maintenance time point; and
A chatbot-based vehicle sales and maintenance service method comprising the step of sending a maintenance recommendation notification to the user terminal in advance before a recommended maintenance time point. A computer program recorded on a computer-readable recording medium to execute the chatbot-based vehicle sales and maintenance service method of any one of claims 6 to 10.

Images