Vehicle information identification method and device, terminal and computer readable storage medium
Through word cutting processing and automotive professional dictionary search, combined with inverted indexing technology and speech fuzzy matching algorithm, the problem of inaccurate identification of vehicle models and spare parts in the existing technology is solved, and a high accuracy and user-friendly search experience is achieved.
Patent Information
- Application Number
- CN202411005439.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2020-08-26
- Publication Date
- 2024-11-15
AI Technical Summary
The prior art cannot accurately identify information when searching for vehicle models and auto parts, resulting in poor user query experience and lack of professionalism and accuracy.
By obtaining vehicle information, performing word-cutting processing to obtain atomic information, and finding matching atomic information in the automotive professional dictionary, combining inverted indexing technology and speech fuzzy matching algorithms to achieve accurate identification and sorting of vehicle information.
It improves the recognition accuracy of vehicle model and spare parts information, improves user experience, and provides a search engine that can accurately search vehicle model and spare parts information to meet the needs of professional and non-professional users.
Smart Images

Figure CN118964381A_ABST
Abstract
Description
[0001] This application is a divisional application based on patent application number 202010872786.1 filed on August 26, 2020 and entitled “Vehicle Information Identification Method, Device, Terminal and Computer-Readable Storage Medium”. Technical Field
[0002] The present invention relates to the field of vehicle technology, and in particular to a vehicle information identification method, device, terminal and computer-readable storage medium. Background Art
[0003] With the rapid development of my country's economy, people's consumption level has gradually increased, and cars have gradually changed from consumables to daily means of transportation for families.
[0004] Nowadays, there are tens of thousands of auto parts. Auto parts products are searched through general platforms or search engines, but product searches are vague and cannot accurately locate the auto parts they are looking for. It usually takes consumers a long time to find the right car model and auto parts. For example, even if the existing search engine uses advanced search, such as inputting "Passat 2007 model", it cannot recognize the "2007 model" information and cannot return the result of "Passat 2007 model", or the returned results of inputting "Passat" and "Passat 2007 model" are exactly the same, or the returned results of inputting "Passat" have no sorting logic; sometimes users make input errors, such as inputting "Toyota Crown Road" with incorrect logic, and the existing search engine cannot return the requested content; for familiar sales version descriptions, such as inputting "Geely King Kong automatic village Taobao version is on the market", the existing search engine cannot return the requested content; for inputting random data such as "Volkswagen taxi", the existing search engine cannot return the requested content; for input such as "BMW 520" or "520", the existing search engine cannot return the requested content for "BMW 520", indicating that the input content has not been segmented and the key information in the user's query statement cannot be recognized. For example, if you enter "520", the priority return is "Lamborghini LP520". The data accuracy is not high, which means there is no inherent or dynamic result sorting logic, and the user query experience is not good; some car models can be searched by entering professional information, such as entering the engine model "N20B20D", and the existing search engine cannot return the request content; some car models can be searched by entering professional information, such as entering the Toyota EPC model code "GRJ120", which the existing search engine cannot recognize and cannot return the request content; when the user wants to enter complex information to search, such as the information "F20M135i" seen on other comprehensive automotive websites, the existing search engine cannot recognize and cannot return the request content; the existing technology, especially in the field of accessories e-commerce, such as entering the spark plug of Passat, cannot accurately locate the product of the e-commerce, and cannot return the result of whether the applicable model of this spark plug is the same as the applicable model of the spark plug of the well-known big brand.
[0005] Currently, there is no mature search engine for car models and spare parts in the automotive field, including some e-commerce websites that do not have the professionalism or accuracy in car model and spare parts retrieval. Summary of the invention
[0006] In view of this, an embodiment of the present invention provides a vehicle information identification method, device, terminal and computer-readable storage medium to solve the problem in the prior art that vehicle model and auto parts information cannot be accurately identified when searching for vehicle model and auto parts information.
[0007] To this end, the embodiment of the present invention provides the following technical solutions:
[0008] In a first aspect, the present invention provides a vehicle information recognition method, comprising:
[0009] Acquire vehicle information; wherein the vehicle information includes at least one of the following: vehicle model information, spare parts information, and vehicle maintenance information;
[0010] Performing word segmentation processing on the vehicle information to obtain one or more atomic information;
[0011] Searching for matching atomic information in a predetermined automobile professional dictionary according to the one or more atomic information; wherein the predetermined automobile professional dictionary includes at least one of the following: a predetermined vehicle model atomic library, a predetermined spare parts atomic library, and predetermined vehicle maintenance information;
[0012] Searching for matching vehicle information in the predetermined automobile professional dictionary according to the matching atomic information; wherein the matching vehicle information includes at least one of the following: matching vehicle model information, spare parts information, and matching vehicle maintenance information; wherein the matching vehicle information includes standard vehicle model information, standard spare parts information, standard vehicle maintenance information, standard vehicle model information and its corresponding standard spare parts information, standard vehicle model information and its corresponding standard vehicle maintenance information, standard spare parts information and its corresponding standard vehicle maintenance information, or standard vehicle model information and its corresponding standard spare parts information and its corresponding standard vehicle maintenance information;
[0013] The matching vehicle information is output.
[0014] Optionally, the method further comprises:
[0015] When no matching atomic information can be found in the predetermined automobile professional dictionary according to the one or more atomic information, converting the one or more atomic information into one or more first atomic pinyin information according to the vehicle field initial consonant and final similarity algorithm;
[0016] Using an inverted indexing technique to process the one or more first atomic phonetic information to obtain one or more second atomic phonetic information;
[0017] The predetermined automobile professional dictionary is searched for designated atomic pinyin information having the smallest difference in the number of Chinese characters between the one or more second atomic pinyin information and the designated atomic pinyin information, and the designated atomic pinyin information is used as the matching atomic information.
[0018] Optionally, when the vehicle information includes vehicle model information and OE code information, before performing word segmentation processing on the vehicle information to obtain one or more atomic information, the method further includes:
[0019] Determine the designated vehicle model information corresponding to the OE coding information;
[0020] When the designated vehicle model information is consistent with the vehicle model information, performing word segmentation processing on the vehicle model information to obtain the one or more atomic information;
[0021] In the case that the designated vehicle model information is inconsistent with the vehicle model information, word segmentation processing is performed on the designated vehicle model information to obtain the one or more atomic information.
[0022] Optionally, before performing word segmentation processing on the vehicle information to obtain one or more atomic information, the method further includes:
[0023] Prioritizing the plurality of sub-vehicle information included in the vehicle information;
[0024] The one or more atomic information are obtained by performing word segmentation processing on all or part of the multiple sub-vehicle information according to the priority order.
[0025] Optionally, when the vehicle information includes first model year information, outputting the matching vehicle information includes:
[0026] When the second model year information corresponding to the standard vehicle model information has no intersection with the first model year information, obtaining the third model year information which has a difference of less than a predetermined threshold in years from the first model year information and has an intersection with the second model year information;
[0027] The vehicle model information and the corresponding third model year information are output.
[0028] Optionally, outputting the matching vehicle information includes:
[0029] In the case where there are multiple pieces of standard vehicle model information, obtaining multiple vehicle model ownership quantities corresponding to the multiple pieces of standard vehicle model information;
[0030] The plurality of standard vehicle model information are output in descending order according to the descending order of the number of vehicle models owned.
[0031] Optionally, the vehicle model information includes vehicle model data and vehicle model parameter data; wherein the vehicle model data is used to define the vehicle model, and the vehicle model parameter data is used to define the vehicle model configuration.
[0032] In a second aspect of the present invention, a vehicle information recognition device is provided, comprising:
[0033] An acquisition module, used to acquire vehicle information; wherein the vehicle information includes at least one of the following: vehicle model information, spare parts information, and vehicle maintenance information;
[0034] A word segmentation module, used for performing word segmentation processing on the vehicle information to obtain one or more atomic information;
[0035] A first search module is used to search for matching atomic information in a predetermined automobile professional dictionary according to the one or more atomic information; wherein the predetermined automobile professional dictionary includes at least one of the following: a predetermined vehicle model atomic library, a predetermined spare parts atomic library, and predetermined vehicle maintenance information;
[0036] A second search module is used to search for matching vehicle information in the predetermined automobile professional dictionary according to the matching atomic information; wherein the matching vehicle information includes at least one of the following: matching vehicle model information, spare parts information, and matching vehicle maintenance information; wherein the matching vehicle information includes standard vehicle model information, standard spare parts information, standard vehicle maintenance information, standard vehicle model information and its corresponding standard spare parts information, standard vehicle model information and its corresponding standard vehicle maintenance information, standard spare parts information and its corresponding standard vehicle maintenance information, or standard vehicle model information and its corresponding standard spare parts information and its corresponding standard vehicle maintenance information;
[0037] An output module is used to output the matching vehicle information.
[0038] In a third aspect of the present invention, a terminal is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the one processor, and the instructions are executed by the at least one processor so that the at least one processor executes any vehicle information identification method described in the first aspect above.
[0039] In a fourth aspect of the present invention, a computer-readable storage medium is provided, on which computer instructions are stored, and when the instructions are executed by a processor, the vehicle information recognition method described in any one of the first aspects is implemented. The technical solution of the embodiment of the present invention has the following advantages:
[0040] The embodiment of the present invention provides a vehicle information identification method, device, terminal and computer-readable storage medium, wherein the method comprises: obtaining vehicle information; the vehicle information comprises at least one of the following: vehicle model information, spare parts information, vehicle maintenance information; performing word segmentation processing on the vehicle information to obtain one or more atomic information; searching for matching atomic information in a predetermined automobile professional dictionary according to the one or more atomic information; wherein the predetermined automobile professional dictionary comprises at least one of the following: a predetermined vehicle model atomic library, a predetermined spare parts atomic library, and predetermined vehicle maintenance information; searching for matching vehicle information in a predetermined automobile professional dictionary according to the matching atomic information; wherein the matching vehicle information comprises at least one of the following: matching vehicle model information, spare parts information, and matching vehicle maintenance information; wherein the matching vehicle information comprises standard vehicle model information, standard spare parts information, standard vehicle maintenance information, standard vehicle model information and its corresponding standard spare parts information, standard vehicle model information and its corresponding standard vehicle maintenance information, standard spare parts information and its corresponding standard vehicle maintenance information, or standard vehicle model information and its corresponding standard spare parts information and corresponding standard vehicle maintenance information; and outputting the matching vehicle information. Through the above method, the input vehicle information molecule is segmented to obtain atoms, the standard atom corresponding to the atom is searched in the standard automobile professional dictionary, and then the standard vehicle information molecule is obtained through the standard atom. This solves the problem in the prior art that when searching for information such as vehicle models and auto parts, the vehicle model and spare parts cannot be accurately identified, improves the accuracy of identifying information such as vehicle models and auto parts, and enhances the user experience. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0042] Figure 1 is a flow chart of a vehicle information identification method according to an embodiment of the present invention;
[0043] Figure 2 is a structural block diagram of a vehicle information identification device according to an embodiment of the present invention;
[0044] Figure 3 It is a schematic diagram of the hardware structure of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0045] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.
[0046] In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, which are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first" and "second" are used only for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.
[0047] In this application, the word "exemplary" is used to mean "used as an example, illustration, or description." Any embodiment described in this application as "exemplary" is not necessarily to be construed as being preferred or advantageous over other embodiments. The following description is given to enable any technician in the field to implement and use the present application. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present application can be implemented without using these specific details. In other instances, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present application with unnecessary details. Therefore, the present application is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present application.
[0048] According to an embodiment of the present invention, a vehicle information identification method embodiment is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0049] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.
[0050] In this embodiment, a vehicle information recognition method is provided, which can be used in the above-mentioned mobile terminal, such as a mobile phone, a tablet computer, etc. Figure 1 is a flow chart of a vehicle information identification method according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:
[0051] Step S101, obtaining vehicle information, wherein the vehicle information includes at least one of the following: vehicle model information, spare parts information, and vehicle maintenance information. Specifically, the vehicle model information includes vehicle model data and vehicle model parameter data, wherein the vehicle model data is used to define the vehicle model, and the vehicle model parameter data is used to define the vehicle model configuration. There are many ways to obtain vehicle information, such as obtaining vehicle information through physical buttons or virtual buttons, or obtaining vehicle information through voice. Those skilled in the art should know that the above-mentioned vehicle information acquisition methods are only for illustration and do not constitute a limitation of the present invention. Other acquisition methods that can be thought of by those skilled in the art are also within the protection scope of the present invention.
[0052] Step S102, segment the vehicle information to obtain one or more atomic information. Specifically, on the one hand, it is necessary to segment natural sentences, and it is necessary to innovate a segmentation algorithm based on automotive professional dictionaries; on the other hand, call nearly 100 automotive experts in the automotive aftermarket to perform point-to-point queries or voice search operations according to their usual work habits to collect original materials. Then machine learning is performed to perform intelligent semantic analysis. At the same time, compare public vehicle model information such as Autohome model descriptions, Yiwang model descriptions, and model description information on the public websites of major well-known manufacturers, all of which are used as input information for algorithm training.
[0053] Step S103, searching for matching atomic information in a predetermined automobile professional dictionary according to one or more atomic information; wherein the predetermined automobile professional dictionary includes at least one of the following: a predetermined vehicle model atomic library, a predetermined spare parts atomic library, and a predetermined vehicle maintenance information. Specifically, vehicle model atomic library: in the automotive aftermarket, the definition of a car is usually specified by a vehicle model, but the description of the vehicle model has both specifications defined by various data manufacturers and popular names on the market. The vehicle model atomic library of the embodiment of the present invention is established to accurately identify various popular vehicle model expressions on the market, and the purpose is to break the vehicle model field to be identified into atoms, such as "Range Rover Sport" corresponding to the atom "Range Rover Sport", "Passat Lingyu" corresponding to "Passat Lingyu" or "Lingyu", etc., to facilitate the logical processing after word segmentation and recognition, and convert vehicle model descriptions with different semantics and dimensions into the finest dimensional vehicle model data defined in the standard vehicle model library. Detailed atoms include "brand, manufacturer, chassis, car series, model, displacement, year, engine, gearbox, sales version..."
[0054] Parts Atomic Library: In the automotive professional field, the same parts name or maintenance operation has both written names, such as front bumper skin, engine hood, center grille, sheet metal, and spray paint, and also common names in the industry, such as front bumper, head cover, ghost mask, and repair spray paint. The front bumper is the front bumper skin, the head cover is the engine hood, the ghost mask is the center grille, and the repair spray paint is sheet metal and spray paint. Moreover, an object often has many different common names. The common names of the front bumper skin include "front bar, front bumper, front rod, front pump handle, front bumper skin..."
[0055] In order to adapt to the general natural speech and semantic AI intelligent analysis, the embodiment of the present invention has created a professional automotive dictionary with tens of millions of words, and created a set of "atomic molecular material theory" to identify professional terms. Molecules are the names of auto parts or maintenance operations. We call it "automotive professional dictionary". The atoms that make up molecules are the smallest units of the professional dictionary, and matter is the sentence.
[0056] Atoms are divided into five categories: A, B, C, F, and E, which represent different attributes respectively. For example, F stands for direction. The "left" in the "left headlight" is the direction word F. E stands for maintenance operations, such as replacement, painting...
[0057] Standard vehicle model library: vehicle model is vehicle model data, which refers to the sum of defined fields required to accurately describe a vehicle model in the vehicle model database. The more defined fields there are, the longer the length of different defined fields arranged according to a certain logic, and the more information the defined fields contain, the more accurate the vehicle model data is. It is specifically divided into vehicle model data and attached vehicle model parameter data. The function of vehicle model data is to define the vehicle model, and vehicle model parameter data is only used to describe the vehicle model configuration. It can usually be subdivided into various forms such as: vehicle model name of the main engine factory, announcement number of the Ministry of Industry and Information Technology, sales version of the distribution channel, body form, country, parameter configuration defined by classification, etc. The standard vehicle model library of the embodiment of the present invention is to organize the descriptions of different vehicle models that have been used in the market of the Chinese automotive aftermarket, and organize a more standardized vehicle model description information after the field definition is standardized, and at the same time combine the vehicle model information of the main engine factory, such as chassis number, engine model, and gearbox model. A set of vehicle group definition specifications are formed from the chassis dimension of the main engine factory, and the vehicle model definition is more accurate and easy to understand. The entire standard vehicle model database has more than 150,000 pieces of information horizontally, covering 99.9% of passenger cars, and each vehicle model has more than 150 detailed and reasonably valuable fields to represent it vertically, including chassis number data, vehicle chassis generation data, engine and replacement number data, Ministry of Industry and Information Technology announcement number data, suspension and shock absorber data, tire and wheel data, listing status, and generation and model change information.
[0058] Standard adaptation library: In order to ensure the quality of brand parts adaptation data, the embodiment of the present invention extracts the public adaptation data of major well-known manufacturers, combines the OE uniqueness data for reverse verification, and forms a standard database with typical reference value in the market. It contains: a complete database of product information, adaptation relationships, OE and replacement chains, and parameter images. This has important guiding significance for the correction and error correction of data of parts of the same brand, adaptation recommendations, and sales management. In the algorithm of the embodiment of the present invention, on the one hand, it can provide a massive atomic library, and on the other hand, it is also an important part of the query result set.
[0059] Specifically, various checks in the case of atomic conflicts in car models: For example, the whole word recognition of "BMW 3 Series / MG6, Mercedes-Benz E200, BMW 3 Series" will first get: "Brand-BMW", and then recognize "Car Series-Mazda 3", but "BMW and Mazda 3" have conflicting characters. So the first step is to check whether "BMW and Mazda 3" are logical car models. If they are not, the program determines that it is an incorrect word segmentation result; the second step is to remove BMW and perform word segmentation again to obtain the correct result "3 Series". For MG6, there is also a character conflict, but "MG and MG6" are logical car models, and the program determines that it is a correct word segmentation result.
[0060] Step S104, searching for matching vehicle information in a predetermined automobile professional dictionary according to the matching atomic information; wherein the matching vehicle information includes at least one of the following: matching vehicle model information, spare parts information, matching vehicle maintenance information; wherein the matching vehicle information includes standard vehicle model information, standard spare parts information, standard vehicle maintenance information, standard vehicle model information and its corresponding standard spare parts information, standard vehicle model information and its corresponding standard vehicle maintenance information, standard spare parts information and its corresponding standard vehicle maintenance information, or standard vehicle model information and its corresponding standard spare parts information and its corresponding standard vehicle maintenance information. Specifically, for "Mercedes-Benz E200, E200" is both a Mercedes-Benz and Zotye vehicle model atom and a Baojun chassis atom, then at this time, it is necessary to judge that E200 should be a Mercedes-Benz vehicle model atom based on the brand Mercedes-Benz, thereby discarding the chassis atom.
[0061] Step S105, outputting the matching vehicle information.
[0062] Through the above steps, the input vehicle information molecule is segmented to obtain atoms, the standard atom corresponding to the atom is searched in the standard automobile professional dictionary, and then the standard vehicle information molecule is obtained through the standard atom. This solves the problem in the prior art that when searching for vehicle models and auto parts and other information, the vehicle model and spare parts information cannot be accurately identified, improves the accuracy of identifying vehicle model and auto parts and other information, and enhances the user experience.
[0063] In a specific optional embodiment, when identifying vehicle information, (1) Pre-processing stage: Step 1: Atoms are pre-written by car model experts. Such as brand atoms, manufacturer atoms, car series atoms, chassis atoms, car model atoms and engine atoms, version atoms, etc. Step 2: These atoms are stored in separate tables and marked for easy distinction. For example, Corolla->'O: Corolla', chassis'CM'->'V: CM', version'Comfort Classic Edition'->'SALE_VERSION: Comfort' and'SALE_VERSION: Classic'. (2) Actual word segmentation moment: Step 3: Determine whether the atom is already in the cache. If not, read the entire table of the type atom to be identified; if it already exists, omit this step to improve matching efficiency. For example, for the input character 'Toyota Corolla Comfort Classic Edition', the car series atom table will be read into the cache in the first case. The second and subsequent times will be matched directly. Step 4: ① Starting from the first character, read in up to 20 bits in sequence, and perform maximum length matching on the input characters in a loop with the atoms in the cache. For example, in the above case, if we have the car series atoms 'Carlo' and 'Corolla', we will match 'Corolla'. Then start from 'Comfort Classic Edition' and repeat this process until all input characters are matched. ② For atoms composed of pure English and letters, in addition to matching the maximum length, it will also determine whether it is an independent word. For example, if you enter 'fiats', although ATS is a Cadillac car series, fiats is obviously a whole word, and no word segmentation is performed at this time. ③ After identification, the brand, manufacturer, model, chassis and model atoms will be removed to prevent the same character from being repeatedly identified as different types of atoms, causing interference with accuracy. The remaining characters are then matched with other configuration items, such as year, displacement, etc. Identification of accessory codes: Analyze the input characters to see if there is a character group consisting of 5 or more consecutive letters and numbers. Match the elements of the character group in the accessory code table in turn. If a match is found, it is considered that the accessory matches the vehicle model.
[0064] In the process of voice-to-text conversion, due to the influence of human accent, external noise, etc., translation errors often occur. For example, when the voice says "front bumper bracket", it is sometimes translated into "wallet bracket". In order to solve this problem, in an optional embodiment, when the matching atomic information cannot be found in the predetermined automobile professional dictionary according to the above one or more atomic information, the above one or more atomic information is converted into one or more first atomic pinyin information according to the vehicle field initial consonant and final similarity algorithm, and the one or more first atomic pinyin information is processed by inverted index technology to obtain one or more second atomic pinyin information, and the designated atomic pinyin information with the smallest difference in the number of Chinese characters with the one or more second atomic pinyin information is searched in the above predetermined automobile professional dictionary, and the designated atomic pinyin information is used as the above matching atomic information. Based on this, the embodiment of the present invention proposes a "speech fuzzy matching algorithm", which first translates the smallest unit of automobile professional dictionary atoms into pinyin combinations, and then innovates the Chinese pinyin similarity algorithm of automobile professional terms according to the similar rules of initial consonants and finals, and translates the voice into the correct professional dictionary. Specifically, the smallest dictionary of automobile profession, i.e., atom, is translated into Chinese Pinyin; the automobile profession dictionary of tens of millions created by the embodiment of the present invention is combined with atoms in Pinyin according to the theory of combining atoms into molecules, and finally forms a automobile profession dictionary of tens of millions in Chinese Pinyin style; in the recognition process of speech-to-text, the Chinese characters are first exhausted in Pinyin, and the Pinyin tones are processed independently. Then the initial consonants and finals are classified according to their similarity, such as the initial consonants g, k, h are similar groups, and the finals en, eng are similar groups. Chinese characters form atoms, and atoms form molecules. In the recognition process of speech-to-text, the "similarity" between Pinyin strings is calculated to find the atoms with the most similar Pinyin, in which the inverted index technology is used, and all the Pinyin strings in the atom table are decomposed according to sliding N bytes (such as "abcde" is decomposed into ab, abc, bcd, cde, de) and indexed, and then the Pinyin string of the input Chinese is also decomposed, and the number of the same and different substrings of the two are quickly located in the index, and then the familiarity is calculated.
[0065] Among all similar target strings in the similarity, find the atomic word that is replaceable by pinyin and has the smallest edit distance as the most similar atomic word. Because the algorithm uses inverted index technology, it can quickly locate the most similar automotive professional dictionary (i.e., molecule) in the pinyin strings of tens of millions of professional dictionaries within a few milliseconds, and finally intelligently recognize the voice as a sentence composed of professional dictionaries, and then perform semantic intelligent analysis on the sentence.
[0066] In an optional embodiment, when the above vehicle information includes vehicle model information and OE coding information, before the vehicle information is segmented to obtain one or more atomic information, the designated vehicle model information corresponding to the OE coding information is determined. When the designated vehicle model information is consistent with the above vehicle model information, the vehicle model information is segmented to obtain the one or more atomic information. When the designated vehicle model information is inconsistent with the vehicle model information, the designated vehicle model information is segmented to obtain the one or more atomic information. Specifically, when OE and vehicle models are mixed, first determine whether there is an intersection. If there is an intersection, take the intersection. If there is no intersection, the vehicle model retrieved by OE is displayed first.
[0067] In an optional embodiment, before the vehicle information is segmented to obtain one or more atomic information, the plurality of sub-vehicle information included in the vehicle information is prioritized, and all or part of the plurality of sub-vehicle information is segmented according to the priority order to obtain the one or more atomic information. Specifically, dynamic screening: when there are both version and year, if there is no displacement atom, the year is matched first, such as Corolla 2014 luxury version. If there is also a displacement atom, the displacement and version are matched first, and the year is considered last, such as Corolla 2014 1.2t luxury version. Sorting algorithm: First, the segmentation algorithm will obtain the OE and model data and model configuration data in the user's random input information. If the OE and model are mixed, the order of the sorting algorithm is to prioritize the model identified by the OE and model information. Secondly, if the user input information contains the year, but the sales year does not exist in the identified model data, the vehicle group with the closest year range is displayed first. For example, "Corolla 2015" will be displayed first as "Corolla_E18_2014-2019"; secondly, for the same weighted result information, it will be sorted in descending order according to the corresponding model ownership information. If the results contain both Porsche and Zotye model information, Porsche's model information will be ranked first because Porsche's ownership is higher than Zotye. Thirdly, in the case of mixed input models and multiple types of configurations. When the model data is unique and there are contradictions among the configurations, dynamic filtering will be performed according to the type of input configuration data. For example, for the 2008 Corolla 1.2T luxury version, because the displacement configuration and sales version configuration combination have a higher priority than the year in the sorting algorithm, the obtained year information will be filtered and only the result of "Corolla 1.2T luxury version" will be displayed; if the input data is "Corolla 2019 premium version", because the year has a higher priority than the sales version when there are only year and sales version configurations, the obtained sales version information will be filtered and only the result of "Corolla 2019 model" will be displayed.
[0068] In another optional embodiment, the scoring and screening algorithm is configured: for the detailed records matched according to the vehicle model, the configuration field atoms in the records and the identified configuration atoms are scored one by one. Intersection is positive, and non-intersection is negative. Then, the record with the highest score is taken according to the set configuration field priority, and each record is compared with the highest score record. The record with a score not lower than the highest score is returned.
[0069] In an optional embodiment, when the vehicle information includes the first model year information, when the second model year information corresponding to the standard vehicle model information has no intersection with the first model year information, the third model year information is obtained, which has a difference of less than a predetermined threshold in years from the first model year information and has an intersection with the second model year information, and the vehicle model information and the corresponding third model year information are output. Specifically, if the user input information contains a model year, but the sales model year does not exist in the identified vehicle model data, the vehicle group with the closest model year range will be displayed first. For example, "Corolla 2015" will be displayed first as "Corolla_E18_2014-2019";
[0070] In the above step S105, specifically, when there are multiple standard vehicle model information, multiple vehicle model ownership corresponding to the multiple standard vehicle model information is obtained, and the multiple standard vehicle model information is output in descending order according to the descending order of the multiple vehicle model ownership. Specifically, for the result information with the same weight, the corresponding vehicle model ownership information is sorted in descending order. If the result contains vehicle model information of both Porsche and Zotye, since the ownership of Porsche is higher than that of Zotye, the vehicle model information of Porsche is ranked first.
[0071] The following is a detailed description with reference to a specific embodiment.
[0072] S1, enter keywords, such as Corolla 3 Series 1.2T;
[0073] S2, segmenting the above-mentioned input keywords according to the underlying segmentation library, the processing process includes English number pinyin, semantic logic processing, correction, etc.;
[0074] S3, determining whether the result after word segmentation includes the vehicle model atom; if it does not include the vehicle model atom, executing step S4; if it includes the vehicle model atom, executing step S6;
[0075] S4, determine whether it is an engine series atom, if the determination result is yes, execute step S5, otherwise no result;
[0076] S5, search for vehicle models by engine / engine serial number / engine replacement number, and output vehicle model results;
[0077] S6, judging whether the vehicle type atom includes a brand or brand atom, and if the judging result is yes, executing step S7, and if the judging result is no, outputting a chassis atom (A) and a configuration atom (B);
[0078] S7, filter vehicle model data;
[0079] S8, determine whether the brand, brand atom and model atom coincide or conflict. If they coincide, output the model atom (A); if they conflict, output the model atom (A);
[0080] S9, processing the above chassis atom (A), configuration atom (B), vehicle type atom (A) and vehicle series model atom (A) in the atomic system, and outputting the vehicle type result.
[0081] In this embodiment, a vehicle information identification device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.
[0082] This embodiment provides a vehicle information recognition device, such as Figure 2 As shown, including:
[0083] The acquisition module 201 is used to acquire vehicle information; wherein the vehicle information includes at least one of the following: vehicle model information, spare parts information, and vehicle maintenance information;
[0084] A word segmentation module 202 is used to perform word segmentation on the vehicle information to obtain one or more atomic information;
[0085] The first search module 203 is used to search for matching atomic information in a predetermined automobile professional dictionary according to the one or more atomic information; wherein the predetermined automobile professional dictionary includes at least one of the following: a predetermined vehicle model atomic library, a predetermined spare parts atomic library, and predetermined vehicle maintenance information;
[0086] The second search module 204 is used to search for matching vehicle information in the predetermined automobile professional dictionary according to the matching atomic information; wherein the matching vehicle information includes standard vehicle model information, standard spare parts information, standard vehicle maintenance information, standard vehicle model information and its corresponding standard spare parts information, standard vehicle model information and its corresponding standard vehicle maintenance information, standard spare parts information and its corresponding standard vehicle maintenance information, or standard vehicle model information and its corresponding standard spare parts information and its corresponding standard vehicle maintenance information;
[0087] The output module 205 is used to output the matching vehicle information.
[0088] The vehicle information identification device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.
[0089] The further functional description of each of the above modules is the same as that of the above corresponding embodiments and will not be repeated here.
[0090] The embodiment of the present invention also provides a mobile terminal having the above Figure 2 The vehicle information identification device shown.
[0091] See also Figure 3 , Figure 3 is a schematic diagram of the structure of a terminal provided by an optional embodiment of the present invention, such as Figure 3 As shown, the terminal may include: at least one processor 301, such as a CPU (Central Processing Unit), at least one communication interface 303, a memory 304, and at least one communication bus 302. The communication bus 302 is used to realize the connection and communication between these components. The communication interface 303 may include a display screen (Display), a keyboard (Keyboard), and the optional communication interface 303 may also include a standard wired interface and a wireless interface. The memory 604 may be a high-speed RAM memory (Random Access Memory) or a non-volatile memory (non-volatile memory), such as at least one disk storage. The memory 304 may optionally be at least one storage device located away from the aforementioned processor 301. The processor 301 may be combined with Figure 2 In the described device, the memory 304 stores an application program, and the processor 301 calls the program code stored in the memory 304 to execute the steps of any of the above-mentioned vehicle information identification methods.
[0092] The communication bus 302 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The communication bus 302 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.
[0093] Among them, the memory 304 may include a volatile memory (English: volatile memory), such as a random access memory (English: random-access memory, abbreviated: RAM); the memory may also include a non-volatile memory (English: non-volatile memory), such as a flash memory (English: flash memory), a hard disk (English: hard disk drive, abbreviated: HDD) or a solid-state drive (English: solid-state drive, abbreviated: SSD); the memory 604 may also include a combination of the above types of memory.
[0094] The processor 301 may be a central processing unit (CPU), a network processor (NP), or a combination of a CPU and a NP.
[0095] The processor 301 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The PLD may be a complex programmable logic device (CPLD), a field programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.
[0096] Optionally, the memory 304 is also used to store program instructions. The processor 301 can call the program instructions to implement the present application Figure 1 The vehicle information identification method shown in the embodiment.
[0097] The embodiment of the present invention further provides a non-transitory computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions can execute the vehicle information identification method in any of the above method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk (HDD) or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above types of memory.
[0098] In summary, 1. The present invention is currently the only search engine in the automotive field that can accurately search for vehicle models and spare parts information. It allows users to input at will, and the returned results can accurately reflect the needs of users, and the order of the returned results meets the needs of users, which improves the user experience and achieves a free, accurate and intelligent search effect. 2. Compared with the search engines in the traditional automotive field, the present invention provides a variety of effective search methods, which can be used by both professional and non-professional users. Therefore, it can be applied to different industries in the automotive field, such as spare parts manufacturers, spare parts dealers, auto repair shops, insurance companies, second-hand car platforms, e-commerce platforms, and car owners, thereby improving the business practicality in the industry. 3. The present invention only needs to input one information parameter to obtain the required vehicle model or spare parts information, and the response and return results are faster, which greatly improves the user experience. 4. Voice search brings high efficiency, intelligence and convenience, and also breaks through the single operation of manual input or step-by-step screening of traditional search engines. 5. The present invention also provides intelligent point-to-point service, and provides a car body appearance map as a reference standard for selection, which greatly improves the user's search efficiency and accuracy. 6. The present invention continuously analyzes and learns user behavior, conducts a comprehensive analysis of the information input by the user, and self-learns to continuously optimize and supplement the atomic library, output results, and sorting algorithms.
[0099] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.
Claims
1. A vehicle information recognition method, It is characterized in that include: Acquire vehicle information; wherein the vehicle information includes at least one of the following: vehicle model information, spare parts information, and vehicle maintenance information; Performing word segmentation processing on the vehicle information to obtain one or more atomic information; Searching for matching atomic information in a predetermined automobile professional dictionary according to the one or more atomic information obtained by word segmentation; wherein the predetermined automobile professional dictionary includes at least one of the following: a predetermined vehicle model atomic library, a predetermined spare parts atomic library, and predetermined vehicle maintenance information; Performing a logic check on the matching atom information based on a predetermined atom library; Searching for matching vehicle information in the predetermined automobile professional dictionary according to the matching atomic information after logical verification; wherein the matching vehicle information includes standard vehicle model information, standard spare parts information, standard vehicle maintenance information, standard vehicle model information and its corresponding standard spare parts information, standard vehicle model information and its corresponding standard vehicle maintenance information, standard spare parts information and its corresponding standard vehicle maintenance information, or standard vehicle model information and its corresponding standard spare parts information and its corresponding standard vehicle maintenance information; The matching vehicle information is output.
2. The vehicle information recognition method according to claim 1, It is characterized in that Performing a logic check on the matching atom information based on a predetermined atom library includes: In response to the verification result that there is illogical matching atomic information in the plurality of matching atomic information, an operation of removing the logical atomic information from the vehicle information is performed to re-segment the vehicle information after the removal process.
3. The vehicle information recognition method according to claim 1 or 2, It is characterized in that The vehicle information is segmented to obtain one or more atomic information, including: based on a predetermined segmentation algorithm model, the vehicle information is segmented. The word segmentation algorithm model is established through the following steps: Obtaining the words queried or searched by the automotive professional and the vehicle model description information disclosed on the website as input information; and Machine learning and algorithm training are performed on the input information to establish the word segmentation algorithm model.
4. The vehicle information recognition method according to claim 1, It is characterized in that The method further comprises: When no matching atomic information can be found in the predetermined automobile professional dictionary according to the one or more atomic information, converting the one or more atomic information into one or more first atomic pinyin information according to the vehicle field initial consonant and final similarity algorithm; Using an inverted indexing technique to process the one or more first atomic phonetic information to obtain one or more second atomic phonetic information; The predetermined automobile professional dictionary is searched for designated atomic pinyin information having the smallest difference in the number of Chinese characters between the one or more second atomic pinyin information and the designated atomic pinyin information, and the designated atomic pinyin information is used as the matching atomic information.
5. The vehicle information recognition method according to claim 1, It is characterized in that In the case where the vehicle information includes vehicle model information and OE code information, before performing word segmentation processing on the vehicle information to obtain one or more atomic information, the method further includes: Determine the designated vehicle model information corresponding to the OE coding information; When the designated vehicle model information is consistent with the vehicle model information, performing word segmentation processing on the vehicle model information to obtain the one or more atomic information; In the case that the designated vehicle model information is inconsistent with the vehicle model information, word segmentation processing is performed on the designated vehicle model information to obtain the one or more atomic information.
6. The vehicle information recognition method according to claim 1, It is characterized in that Before performing word segmentation processing on the vehicle information to obtain one or more atomic information, the method further includes: Prioritizing the plurality of sub-vehicle information included in the vehicle information; The one or more atomic information are obtained by performing word segmentation processing on all or part of the multiple sub-vehicle information according to the priority order.
7. The vehicle information recognition method according to claim 1, It is characterized in that When the vehicle information includes the first model year information, outputting the matching vehicle information includes: When the second model year information corresponding to the standard vehicle model information has no intersection with the first model year information, obtaining the third model year information which has a difference of less than a predetermined threshold in years from the first model year information and has an intersection with the second model year information; The vehicle model information and the corresponding third model year information are output.
8. The vehicle information recognition method according to claim 1, It is characterized in that Outputting the matching vehicle information includes: In the case where there are multiple pieces of standard vehicle model information, obtaining multiple vehicle model ownership quantities corresponding to the multiple pieces of standard vehicle model information; The plurality of standard vehicle model information are output in descending order according to the descending order of the number of vehicle models owned.
9. The vehicle information recognition method according to any one of claims 1 to 8, It is characterized in that The vehicle model information includes vehicle model data and vehicle model parameter data; wherein the vehicle model data is used to define the vehicle model, and the vehicle model parameter data is used to define the vehicle model configuration.
10. A vehicle information recognition device, It is characterized in that include: An acquisition module, used to acquire vehicle information; wherein the vehicle information includes at least one of the following: vehicle model information, spare parts information, and vehicle maintenance information; A word segmentation module, used for performing word segmentation processing on the vehicle information to obtain one or more atomic information; A first search module is used to search for matching atomic information in a predetermined automobile professional dictionary according to the one or more atomic information obtained by word segmentation processing; wherein the predetermined automobile professional dictionary includes at least one of the following: a predetermined vehicle model atomic library, a predetermined spare parts atomic library, and predetermined vehicle maintenance information; A logic verification module, performing logic verification on the matching atom information based on a predetermined atom library; A second search module is used to search for matching vehicle information in the predetermined automobile professional dictionary according to the matching atomic information after logical verification; wherein the matching vehicle information includes standard vehicle model information, standard spare parts information, standard vehicle maintenance information, standard vehicle model information and its corresponding standard spare parts information, standard vehicle model information and its corresponding standard vehicle maintenance information, standard spare parts information and its corresponding standard vehicle maintenance information, or standard vehicle model information and its corresponding standard spare parts information and its corresponding standard vehicle maintenance information; An output module is used to output the matching vehicle information.
11. A terminal, It is characterized in that include: at least one processor; And a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the one processor, and the instructions are executed by the at least one processor so that the at least one processor executes the vehicle information identification method described in any one of claims 1-9 above.
12. A computer-readable storage medium having computer instructions stored thereon, It is characterized in that When the instruction is executed by the processor, the vehicle information identification method described in any one of claims 1-9 is implemented.
