An automobile marketing service response method based on artificial intelligence semantic analysis

Abstract

The present invention discloses a car marketing service answering method based on artificial intelligence semantic analysis, comprising the following steps: obtaining character information of a user's target text, and using a car enterprise field algorithm to perform semantic analysis on the character information of the user's target text; encrypting the text after semantic analysis, generating a string token and uploading it to a cloud server algorithm center, wherein the cloud server algorithm center translates the string token into a character set; using an intelligent labeling algorithm to match and analyze the character set generated by the cloud server algorithm center with a car enterprise field dictionary, and generating a weighted label for each sentence; using a label mapping algorithm to map and associate the weighted labels with a preset content label system, and derive the closest multiple groups of content; and returning the closest multiple groups of content to the user. The present invention can better solve the problem of unclear user pointing.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an automobile marketing service response method based on artificial intelligence semantic analysis. Background Art

[0002] Natural Language Processing (NLP) is an important direction in the fields of computer science and artificial intelligence. It studies various theories and methods that can achieve effective communication between people and computers using natural language. Natural language processing is a science that integrates linguistics, computer science, and mathematics. Therefore, research in this field will involve natural language, that is, the language people use in daily life, so it is closely related to the study of linguistics, but there are important differences. Natural language processing is not a general study of natural language, but the development of computer systems that can effectively realize natural language communication, especially the software systems therein. Therefore, it is a part of computer science. Natural language processing is mainly used in machine translation, public opinion monitoring, automatic summarization, opinion extraction, text classification, question answering, text semantic comparison, speech recognition, Chinese OCR, etc.

[0003] Natural language processing is a discipline that uses computer technology to analyze, understand and process natural language, taking language as the object. That is, using computers as a powerful tool for language research, quantitatively studying language information with the support of computers, and providing language descriptions that can be used by both humans and computers. It includes two parts: Natural Language Understanding (NLU) and Natural Language Generation (NLG). It is a typical marginal interdisciplinary subject, involving language science, computer science, mathematics, cognitive science, logic, etc., and focuses on the interaction between computers and human (natural) language. People call the process of processing natural language with computers natural language understanding (NLU), human language technology (HLT), computational linguistics HL (Computational Linguistics), quantitative linguistics (Quantitative Linguistics), and mathematical linguistics (Mathematical Linguistics) in different periods or with different focuses.

[0004] In the field of automobile marketing, automobiles are an outstanding product of the industrial age. They are quite complex for customers. There are a lot of questions that need to be answered before and after purchase. Communicating with users through intelligent semantic analysis robots is a breakthrough expansion and application in the field of natural language processing. With the development of artificial intelligence, natural language processing has become an application-enabling technology. Given the complexity of natural language itself and the large number of proper nouns and technical nouns in the field of automobile marketing, traditional semantic recognition has been challenged, with problems such as failure to translate proper nouns and failure to resolve synonyms. Therefore, upgrading the existing semantic recognition in the field of automobile marketing is the key to solving this problem. Summary of the invention

[0005] In response to the above technical problems, the purpose of the present invention is to provide an automobile marketing service response method based on artificial intelligence semantic analysis, so as to solve the technical problems in the field of automobile marketing, such as how to quickly resolve customer questions, reduce the workload of manual services, improve the quality of professional services, and improve customer satisfaction.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is: a method for responding to automobile marketing services based on artificial intelligence semantic analysis, comprising the following steps:

[0007] Obtain the character information of the user's target text, and use the automotive industry algorithm to perform semantic analysis on the character information of the user's target text;

[0008] Encrypt the semantically analyzed text to generate a string token and upload it to the cloud server algorithm center, which translates the string token into a character set;

[0009] The intelligent labeling algorithm is used to match and analyze the character set generated by the cloud server algorithm center with the automotive industry dictionary, and a weighted label is generated for each sentence.

[0010] A tag mapping algorithm is used to map and associate weighted tags with a preset content tag system to obtain the most relevant sets of content;

[0011] Return the most relevant sets of content to the user.

[0012] Furthermore, the automotive enterprise field algorithm includes a natural language processing (NLP) algorithm, and the natural language processing (NLP) algorithm performs semantic analysis on the user target text, and the semantic analysis includes at least lexical analysis, syntactic analysis, pragmatic analysis and contextual analysis.

[0013] Furthermore, the cloud server algorithm center translates the string token into a character set by splitting the string token into multiple sentences through the cloud server algorithm center, and then splitting the single sentence into subject, predicate, object and adverbial words to form a character set.

[0014] Furthermore, the smart label algorithm includes the following steps:

[0015] Step A: assign different weights x to characters in the m character sets generated by the cloud server algorithm center according to different parts of speech, and obtain n valid characters;

[0016] Step B: Match the valid characters with the dictionary of the automotive industry, assign the characters a label A and label fit y according to the matching degree, and obtain the label coefficient s1 of a single character, s1=(x*y)A;

[0017] Step C: Collect all n characters in the sentence and summarize them according to the label to obtain the label feature expression of the input character sentence: S=Sum(Sa)A+Sum(Sb)B+...Sum(Sx)X, where Sum(Sa) is the sum of the label coefficients of all valid characters labeled A, Sum(Sb) is the sum of the label coefficients of all valid characters labeled B, and Sum(Sx) is the sum of the label coefficients of all valid characters labeled X.

[0018] Furthermore, the weight x is a floating weight, and the value of weight x is the exclusive weight value for different types of parts of speech in the field of automobile marketing. After the floating weight is granted, a weight of 0 is added to some non-critical analyses such as modal particles and predicates to improve the overall algorithm efficiency. The label fit y is the label fit obtained by matching the characters in the intelligent label algorithm with the system's built-in label dictionary, label category, and application-level label. It is an application of label analysis and summary in the field of automobile marketing.

[0019] Furthermore, the label mapping algorithm includes the following steps:

[0020] Step a: Generate a label weight Pareto chart according to the number of label coefficients and the weights of label coefficients in the label feature expression generated by the intelligent label algorithm;

[0021] Step b: Obtain the core indicator labels according to the label weight Pareto chart. The number is N, and the mapping method is determined according to different core indicator labels.

[0022] Furthermore, the number N of the core indicator labels is less than 10.

[0023] Furthermore, the mapping method is determined according to different core indicator labels:

[0024] When N=1, the absolute mapping method is used to directly output the content with the highest content quality ranking to the user;

[0025] When N=2 or N=3, the main tag mapping method is used to perform ratio mapping based on the main tags and their proportions with the main tags of the content tags, and the content with the highest fit is selected and output to the user;

[0026] When N>3, a combined tag mapping method is used to select multiple suitable contents for the user to continue to select or view according to the tag combination.

[0027] The beneficial effects of the present invention are: user-oriented, scenario-based solution to user problems, improving industry capabilities, and better solving problems that are unclear to users; it can quickly resolve customer questions, reduce the workload of manual services, improve professional service quality, and improve customer satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0029] Figure 1 It is a flow chart of the present invention;

[0030] Figure 2 It is a flowchart of the present invention;

[0031] Figure 3 This is the flowchart of the smart label algorithm;

[0032] Figure 4 This is a flowchart of the label mapping algorithm. DETAILED DESCRIPTION

[0033] With the development of artificial intelligence, natural language processing has become an application-enabling technology. In view of the complexity of natural language itself and the existence of a large number of proper nouns and technical nouns in the field of automobile marketing, traditional semantic recognition has been challenged, and there are problems such as failure to translate proper nouns and failure to resolve synonyms. Therefore, upgrading the existing semantic recognition in the field of automobile marketing is the key to solving this problem. Based on the existing technology, the present invention combines the intelligent label algorithm and the label mapping algorithm to make the application of natural language recognition in the field of automobile marketing more mature. Please refer to the embodiments of the present invention for further details.

[0034] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0035] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0036] Some embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0037] Embodiment 1:

[0038] See also Figure 1-4 , an automobile marketing service response method based on artificial intelligence semantic analysis, comprising the following steps:

[0039] Obtain the character information of the user's target text, and use the automotive industry algorithm to perform semantic analysis on the character information of the user's target text;

[0040] Encrypt the semantically analyzed text to generate a string token and upload it to the cloud server algorithm center, which translates the string token into a character set;

[0041] The intelligent labeling algorithm is used to match and analyze the character set generated by the cloud server algorithm center with the automotive industry dictionary, and a weighted label is generated for each sentence.

[0042] A tag mapping algorithm is used to map and associate weighted tags with a preset content tag system to obtain the most relevant sets of content;

[0043] The most relevant groups of content are returned to the user, completing the execution of the automobile marketing service response method based on artificial intelligence semantic analysis.

[0044] In this embodiment, the car enterprise field algorithm includes a natural language processing NLP algorithm, which performs semantic analysis on the user target text, and the semantic analysis includes at least lexical analysis, syntactic analysis, pragmatic analysis and context analysis. Further, the core of the natural language processing NLP algorithm is semantic analysis. Semantic analysis is a method for analyzing semantic information based on natural language. It not only performs grammatical analysis such as lexical analysis and syntactic analysis, but also involves the meaning contained in words, phrases, sentences and paragraphs. The purpose is to use the semantic structure of the sentence to represent the structure of the language. The semantic analysis technology specifically includes the following points: 1. Lexical analysis. Lexical analysis includes two aspects: word form analysis and vocabulary analysis. Generally speaking, word form analysis is mainly manifested in the analysis of prefixes and suffixes of words, while vocabulary analysis is manifested in the control of the entire vocabulary system, so that the characteristics of the user input information can be analyzed more accurately, and the search process can be finally completed accurately. 2. Syntactic analysis. Syntactic analysis is the analysis of vocabulary phrases of the natural language input by the user. The purpose is to identify the syntactic structure of the sentence to realize the process of automatic syntactic analysis. 3. Pragmatic analysis: Compared with semantic analysis, pragmatic analysis adds analysis of context, language background, and context, that is, it extracts additional information such as images and interpersonal relationships from the structure of the article. It is a more advanced linguistic analysis that links the content of the sentence with the details of real life to form a dynamic semantic structure. 4. Contextual analysis: Contextual analysis mainly refers to the technology of analyzing a large number of "gaps" outside the original query text in order to more accurately interpret the query language. These "gaps" include general knowledge, knowledge in specific fields, and the needs of query users.

[0045] In this embodiment, the cloud server algorithm center translates the string token (token, lexical analysis is the process of converting a character sequence into a token sequence in computer science, and the process of generating tokens from an input character stream is called tokenization. In this process, the lexical analyzer also classifies the tokens) into a character set by splitting the string token into multiple sentences through the cloud server algorithm center, and then splitting a single sentence into subject, predicate, object and adverbial words to form a character set.

[0046] In this embodiment, the smart tag algorithm includes the following steps:

[0047] Step A: assign different weights x to characters in the m character sets generated by the cloud server algorithm center according to different parts of speech, and obtain n valid characters;

[0048] Step B: Match the valid characters with the dictionary of the automotive industry, assign the characters a label A and label fit y according to the matching degree, and obtain the label coefficient s1 of a single character, s1=(x*y)A;

[0049] Step C: Collect all n characters in the sentence and summarize them according to the label to obtain the label feature expression of the input character sentence: S=Sum(Sa)A+Sum(Sb)B+...Sum(Sx)X, where Sum(Sa) is the sum of the label coefficients of all valid characters labeled A, Sum(Sb) is the sum of the label coefficients of all valid characters labeled B, and Sum(Sx) is the sum of the label coefficients of all valid characters labeled X.

[0050] In this embodiment, the weight x is a floating weight, and the value of weight x is the exclusive weight value for different types of parts of speech in the field of automobile marketing. After the floating weight is granted, a weight of 0 is added to some non-critical analyses such as modal particles and predicates to improve the overall algorithm efficiency. The label fit y is the degree of matching and mapping between the characters in the intelligent label algorithm and the system's built-in label dictionary, label category, and application-level labels, and its label fit is obtained. It is an application of label analysis and summary in the field of automobile marketing. Furthermore, Sum (Sa) to Sum (Sx) are arranged from high to low, which can improve the efficiency of subsequent model comparisons.

[0051] In this embodiment, the label mapping algorithm includes the following steps:

[0052] Step a: Generate a label weight Pareto chart based on the number of label coefficients and the weight of the label coefficients in the label feature expression generated by the intelligent label algorithm (Pareto chart is a chart that arranges the quality problems and quality improvement projects in order of importance. It is named after the Italian economist V. Pareto. Pareto chart is also called arrangement chart or primary and secondary chart. It is a histogram drawn in order of frequency of occurrence, indicating how many results are caused by confirmed types or categories of reasons);

[0053] Step b: Obtain the core indicator labels according to the label weight Pareto chart. The number is N, and the mapping method is determined according to different core indicator labels.

[0054] In this embodiment, the number N of the core indicator labels is less than 10.

[0055] In this embodiment, the mapping method is determined according to different core indicator labels:

[0056] When N=1, the absolute mapping method is used to directly output the content with the highest content quality ranking to the user;

[0057] When N=2 or N=3, the main tag mapping method is used to perform ratio mapping based on the main tags and their proportions with the main tags of the content tags, and the content with the highest fit is selected and output to the user;

[0058] When N>3, a combined tag mapping method is used to select multiple suitable contents for the user to continue to select or view according to the tag combination.

[0059] After obtaining the questions asked by the user, the present invention performs intelligent semantic analysis according to the content of the user's consultation, and the analysis results include but are not limited to: user historical portraits, user emotions, user demands, user concerns and customer operation forms; after the analysis, the user problem decomposition points are intelligently analyzed in combination with the user operation algorithm, and the analysis results are matched and combined with the content to form a customized problem solution push suitable for the user, and guide the user to gradually clarify the user problem and solve it properly; the push and analysis algorithms are improved according to each customer inquiry; finally, the user operation content is analyzed and improved through customer satisfaction and push content frequency to form a closed loop, quickly resolve customer questions, reduce the workload of manual services, improve the quality of professional services, and improve customer satisfaction.

[0060] The present invention accumulates historical experience and machine learning, accumulates semantic analysis dictionaries and semantic tag parsing, and forms an efficient operation algorithm for content empowerment and content reach that is user-oriented. It solves user problems in a user-oriented and scenario-based manner and improves industry capabilities.

[0061] The present invention is applicable to the technical fields of natural language analysis and natural language processing, and provides a text semantic analysis method, a semantic analysis scenario, and a semantic analysis application carrier in the scenario of viewing and purchasing a car.

[0062] The present invention better solves the problem of unclear user directions by mapping text labels with content label systems. At the same time, it conducts in-depth analysis based on dimensions such as basic customer information, basic customer labels, customer tone, and customer exclamations to form a weighted content recommendation mechanism, thus opening up a new path for quickly connecting with customers and solving customer problems.

[0063] It should be noted that, for the above embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification belong to preferred embodiments, and the actions involved are not necessarily required by the present application.

[0064] The above embodiments describe the basic principles and main features of the present invention and the advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the changes and modifications made by those skilled in the art shall be within the scope of protection of the appended claims of the present invention without departing from the spirit and scope of the present invention.

Claims

1. A method for responding to automobile marketing services based on artificial intelligence semantic analysis, characterized in that: The steps include: Obtain the character information of the user's target text, and use the automotive industry algorithm to perform semantic analysis on the character information of the user's target text; Encrypt the text after semantic analysis, generate a string token and upload it to the cloud server algorithm center, and the cloud server algorithm center translates the string token into a character set; The intelligent labeling algorithm is used to match and analyze the character set generated by the cloud server algorithm center with the automotive enterprise domain dictionary to generate a weighted label for each sentence; the intelligent labeling algorithm includes the following steps: Step A: assign different weights x to characters in the m character sets generated by the cloud server algorithm center according to different parts of speech, and obtain n valid characters; Step B: Match the valid characters with the dictionary of the automotive industry, assign the characters a label A and label fit y according to the matching degree, and obtain the label coefficient s1 of a single character, s1=(x*y)A; Step C: Collect all n characters in the sentence and summarize them according to the labels to obtain the label feature expression of the input character sentence: S = Sum (Sa) A + Sum (Sb) B + ... Sum (Sx) X, where Sum (Sa) is the sum of the label coefficients of all valid characters marked with label A, Sum (Sb) is the sum of the label coefficients of all valid characters marked with label B, and Sum (Sx) is the sum of the label coefficients of all valid characters marked with label X; A label mapping algorithm is used to map and associate weighted labels with a preset content label system to obtain the closest multiple groups of content; the label mapping algorithm includes the following steps: Step a: Generate a label weight Pareto chart according to the number of label coefficients and the weights of label coefficients in the label feature expression generated by the intelligent label algorithm; Step b: Obtain the core indicator labels according to the label weight Pareto chart, the number is N, and determine the mapping method according to different core indicator labels; Return the most relevant sets of content to the user.

2. The automobile marketing service response method based on artificial intelligence semantic analysis according to claim 1, characterized in that: The automotive enterprise field algorithm includes a natural language processing (NLP) algorithm, which performs semantic analysis on the user's target text. The semantic analysis includes at least lexical analysis, syntactic analysis, pragmatic analysis and contextual analysis.

3. The automobile marketing service response method based on artificial intelligence semantic analysis according to claim 1, characterized in that: The method by which the cloud server algorithm center translates the string token into a character set is as follows: the cloud server algorithm center splits the string token into multiple sentences, and then splits the single sentence into subject, predicate, object and adverbial words to form a character set.

4. The automobile marketing service response method based on artificial intelligence semantic analysis as claimed in claim 1, characterized in that: The weight x is a floating weight, and the value of the weight x is an exclusive weight value for different types of part-of-speech words in the field of automobile marketing.

5. The automobile marketing service response method based on artificial intelligence semantic analysis as claimed in claim 1, characterized in that: The number N of the core indicator labels is less than 10.

6. The automobile marketing service response method based on artificial intelligence semantic analysis according to claim 1, characterized in that: The mapping method is determined according to different core indicator labels: When N=1, the absolute mapping method is used to directly output the content with the highest content quality ranking to the user; When N=2 or N=3, the main tag mapping method is used to perform ratio mapping based on the main tags and their proportions with the main tags of the content tags, and the content with the highest fit is selected and output to the user; When N>3, a combined tag mapping method is used to select multiple suitable contents for the user to continue to select or view according to the tag combination.

Images